diff --git a/.changeset/bash-parser-service.md b/.changeset/bash-parser-service.md
new file mode 100644
index 0000000000..05e5f34208
--- /dev/null
+++ b/.changeset/bash-parser-service.md
@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": patch
+---
+
+Add an internal bash parsing capability that turns shell command strings into syntax trees, in preparation for per-command permission analysis. No user-facing behavior change yet.
diff --git a/AGENTS.md b/AGENTS.md
index 2349080c17..26bf9dcdba 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -28,6 +28,7 @@ This is a TypeScript monorepo built for agent-assisted development. Keep the roo
 - `packages/kap-server`: the Kimi Code server, backed by the DI × Scope agent engine (`@moonshot-ai/agent-core-v2`). Exposes sessions over REST + WebSocket (`/api/v1` + `/api/v1/ws`); bootstrapped from `src/start.ts` and consumed by `apps/kimi-code`. The RPC surface is `/api/v1/debug/*` — a reflection dispatcher over the ENTIRE scoped DI registry (every Service callable, no whitelist; `src/transport/registerDebugRoutes.ts` + `serviceDispatcherRoutes.ts`), mounted only with `--debug-endpoints` on a loopback bind and gated by the global bearer auth; repo dev scripts pass the flag.
 - `packages/klient`: the client SDK — a contract-driven facade over agent-core-v2 with aggregated `global.*` / `session(id).*` / `agent(id).*` methods, zod validation on every call, and klient-level typed event forwarding. Transport is chosen once at creation via subpath entry (`@moonshot-ai/klient/ipc|memory`); both return the same `Klient`. The package also hosts the e2e suites: the legacy `/api/v1` live suites (`test/e2e/legacy/`) and the docker e2e runner (`pnpm --filter @moonshot-ai/klient docker:e2e`). See `packages/klient/AGENTS.md`.
 - `packages/server-e2e`: live e2e tests and scenarios against a running server (`KIMI_SERVER_URL`, default `http://127.0.0.1:58627`). See `packages/server-e2e/AGENTS.md`.
+- `packages/tree-sitter-bash`: a pure-TypeScript bash parser (no runtime deps, no wasm) that produces a syntax tree with tree-sitter-bash 0.25.0 named-node type names and UTF-16 code-unit offsets. `parse(source, { timeoutMs, maxNodes })` runs under a deterministic budget (default 50 ms / 50k nodes, plus per-chain recursion depth caps) and returns a discriminated `ParseResult` (`{ ok, rootNode, hasError }` or `{ ok: false, reason: 'aborted' }`) — callers must treat aborted/hasError trees as "cannot analyze" and degrade. Parser only, no safety judgments; consumers (e.g. Bash tool permission matching) live elsewhere. Known deviations from the reference are tracked in the package README's "Known differences" section, pinned by differential fixtures tested against the real `tree-sitter-bash` wasm (dev-only).
 
 ## Environment Requirements
 
diff --git a/flake.nix b/flake.nix
index 7f56b35994..5261a5474c 100644
--- a/flake.nix
+++ b/flake.nix
@@ -77,6 +77,7 @@
         ./packages/protocol
         ./packages/telemetry
         ./packages/transcript
+        ./packages/tree-sitter-bash
         ./apps/kimi-code
         ./apps/vscode
         ./apps/kimi-inspect
@@ -103,6 +104,7 @@
         "@moonshot-ai/protocol"
         "@moonshot-ai/kimi-telemetry"
         "@moonshot-ai/transcript"
+        "@moonshot-ai/tree-sitter-bash"
         "@moonshot-ai/kimi-code"
         "kimi-code"
         "@moonshot-ai/kimi-inspect"
@@ -160,7 +162,7 @@
               inherit (finalAttrs) pname version src pnpmWorkspaces;
               inherit pnpm;
               fetcherVersion = 3;
-              hash = "sha256-+pzJfoWJwVXIUU8oc56LVpfNjSY6MABID5g11Cm92xw=";
+              hash = "sha256-niK61WYaqCJLUUlp4AGgq2/DOF3sqst6N9Kpk9noXyE=";
             };
 
             nativeBuildInputs = [
diff --git a/packages/agent-core-v2/package.json b/packages/agent-core-v2/package.json
index caba0689d6..1eb599fdc7 100644
--- a/packages/agent-core-v2/package.json
+++ b/packages/agent-core-v2/package.json
@@ -63,6 +63,7 @@
     "@moonshot-ai/kimi-code-oauth": "workspace:^",
     "@moonshot-ai/minidb": "workspace:^",
     "@moonshot-ai/protocol": "workspace:^",
+    "@moonshot-ai/tree-sitter-bash": "workspace:^",
     "@mozilla/readability": "^0.6.0",
     "ajv": "^8.18.0",
     "ajv-formats": "^3.0.1",
diff --git a/packages/agent-core-v2/scripts/check-domain-layers.mjs b/packages/agent-core-v2/scripts/check-domain-layers.mjs
index b93a586984..cb2cfc7685 100644
--- a/packages/agent-core-v2/scripts/check-domain-layers.mjs
+++ b/packages/agent-core-v2/scripts/check-domain-layers.mjs
@@ -102,6 +102,11 @@ const DOMAIN_LAYER = new Map([
   // Depends only on `_base`; sits in L1 beside the other program-control
   // layer substrates.
   ['task', 1],
+  // `bashParser` is the App-scope adapter over the pure
+  // `@moonshot-ai/tree-sitter-bash` package (bash source → syntax tree DTO).
+  // It injects no services, so it sits in L1 beside the other pure
+  // capabilities.
+  ['bashParser', 1],
   // persistence/ and os/ — the two-level scopes. `interface` holds contracts
   // (same layer as the old domains they replace); `backends` holds
   // implementations that may depend on cross-domain services at various layers.
diff --git a/packages/agent-core-v2/src/app/bashParser/bashParser.ts b/packages/agent-core-v2/src/app/bashParser/bashParser.ts
new file mode 100644
index 0000000000..154c8550f2
--- /dev/null
+++ b/packages/agent-core-v2/src/app/bashParser/bashParser.ts
@@ -0,0 +1,42 @@
+/**
+ * `bashParser` domain (L1) — bash source parsing capability.
+ *
+ * Defines the `IBashParserService` that parses a bash source string into a
+ * syntax tree through the pure `@moonshot-ai/tree-sitter-bash` package, plus
+ * the wire-safe DTO types it returns: `BashSyntaxNode` drops the cyclic
+ * `parent` link so results can cross the RPC boundary, and offsets are
+ * UTF-16 code units (`text` always equals `source.slice(start, end)`). The
+ * parse runs under a deterministic budget — budget exhaustion yields
+ * `{ ok: false, reason: 'aborted' }` and malformed input yields
+ * `hasError: true`, never a throw; callers that cannot analyze a command
+ * must degrade on either signal. Bound at App scope.
+ */
+
+import { createDecorator, type ServiceIdentifier } from '#/_base/di/instantiation';
+
+export interface BashSyntaxNode {
+  readonly type: string;
+  readonly text: string;
+  readonly startIndex: number;
+  readonly endIndex: number;
+  readonly isNamed: boolean;
+  readonly children: readonly BashSyntaxNode[];
+}
+
+export interface BashParseOptions {
+  readonly timeoutMs?: number;
+  readonly maxNodes?: number;
+}
+
+export type BashParseResult =
+  | { readonly ok: true; readonly hasError: boolean; readonly root: BashSyntaxNode }
+  | { readonly ok: false; readonly reason: 'aborted' };
+
+export interface IBashParserService {
+  readonly _serviceBrand: undefined;
+
+  parse(source: string, options?: BashParseOptions): BashParseResult;
+}
+
+export const IBashParserService: ServiceIdentifier<IBashParserService> =
+  createDecorator<IBashParserService>('bashParserService');
diff --git a/packages/agent-core-v2/src/app/bashParser/bashParserService.ts b/packages/agent-core-v2/src/app/bashParser/bashParserService.ts
new file mode 100644
index 0000000000..60c06bcbba
--- /dev/null
+++ b/packages/agent-core-v2/src/app/bashParser/bashParserService.ts
@@ -0,0 +1,49 @@
+/**
+ * `bashParser` domain (L1) — `IBashParserService` implementation.
+ *
+ * Thin adapter over the pure `@moonshot-ai/tree-sitter-bash` package: runs
+ * its budgeted `parse` and snapshots the returned tree into the wire-safe
+ * `BashSyntaxNode` DTO (source-ordered children including anonymous tokens,
+ * `parent` links dropped). Owns no state and injects no services. Bound at
+ * App scope.
+ */
+
+import { parse } from '@moonshot-ai/tree-sitter-bash';
+import type { SyntaxNode } from '@moonshot-ai/tree-sitter-bash';
+
+import { InstantiationType } from '#/_base/di/extensions';
+import { LifecycleScope, registerScopedService } from '#/_base/di/scope';
+
+import type { BashParseOptions, BashParseResult, BashSyntaxNode } from './bashParser';
+import { IBashParserService } from './bashParser';
+
+function snapshot(node: SyntaxNode): BashSyntaxNode {
+  return {
+    type: node.type,
+    text: node.text,
+    startIndex: node.startIndex,
+    endIndex: node.endIndex,
+    isNamed: node.isNamed,
+    children: node.children.map(snapshot),
+  };
+}
+
+export class BashParserService implements IBashParserService {
+  declare readonly _serviceBrand: undefined;
+
+  parse(source: string, options: BashParseOptions = {}): BashParseResult {
+    const result = parse(source, options);
+    if (!result.ok) {
+      return { ok: false, reason: result.reason };
+    }
+    return { ok: true, hasError: result.hasError, root: snapshot(result.rootNode) };
+  }
+}
+
+registerScopedService(
+  LifecycleScope.App,
+  IBashParserService,
+  BashParserService,
+  InstantiationType.Delayed,
+  'bashParser',
+);
diff --git a/packages/agent-core-v2/src/index.ts b/packages/agent-core-v2/src/index.ts
index a6678675b3..99a67fda31 100644
--- a/packages/agent-core-v2/src/index.ts
+++ b/packages/agent-core-v2/src/index.ts
@@ -282,6 +282,8 @@ export * from '#/app/workspaceRegistry/workspacePersistence';
 export * from '#/app/workspaceRegistry/fileWorkspacePersistence';
 import '#/app/workspaceRegistry/workspaceQueryService';
 import '#/app/git/gitService';
+export * from '#/app/bashParser/bashParser';
+import '#/app/bashParser/bashParserService';
 export * from '#/session/process/processRunner';
 export * from '#/session/process/processRunnerService';
 export * from '#/session/sessionFs/errors';
diff --git a/packages/agent-core-v2/test/app/bashParser/bashParserService.test.ts b/packages/agent-core-v2/test/app/bashParser/bashParserService.test.ts
new file mode 100644
index 0000000000..2228f50cac
--- /dev/null
+++ b/packages/agent-core-v2/test/app/bashParser/bashParserService.test.ts
@@ -0,0 +1,72 @@
+import { afterEach, beforeEach, describe, expect, it } from 'vitest';
+
+import { DisposableStore } from '#/_base/di/lifecycle';
+import { createServices, type TestInstantiationService } from '#/_base/di/test';
+import type { BashSyntaxNode } from '#/app/bashParser/bashParser';
+import { IBashParserService } from '#/app/bashParser/bashParser';
+import { BashParserService } from '#/app/bashParser/bashParserService';
+
+describe('BashParserService', () => {
+  let disposables: DisposableStore;
+  let ix: TestInstantiationService;
+  let service: IBashParserService;
+
+  beforeEach(() => {
+    disposables = new DisposableStore();
+    ix = createServices(disposables, {
+      additionalServices: (reg) => {
+        reg.define(IBashParserService, BashParserService);
+      },
+    });
+    service = ix.get(IBashParserService);
+  });
+
+  afterEach(() => {
+    disposables.dispose();
+  });
+
+  it('splits a compound command into per-command nodes', () => {
+    const result = service.parse('git status && rm -rf /');
+    if (!result.ok) {
+      throw new Error('expected ok');
+    }
+    expect(result.hasError).toBe(false);
+    expect(result.root.type).toBe('program');
+    const commands: string[] = [];
+    const walk = (node: BashSyntaxNode): void => {
+      if (node.type === 'command') {
+        commands.push(node.text);
+      }
+      node.children.forEach(walk);
+    };
+    walk(result.root);
+    expect(commands).toEqual(['git status', 'rm -rf /']);
+  });
+
+  it('flags malformed input with hasError instead of throwing', () => {
+    const result = service.parse('echo "unterminated');
+    if (!result.ok) {
+      throw new Error('expected ok');
+    }
+    expect(result.hasError).toBe(true);
+  });
+
+  it('reports budget exhaustion as aborted', () => {
+    const result = service.parse('ls; '.repeat(20000), { maxNodes: 100 });
+    expect(result).toEqual({ ok: false, reason: 'aborted' });
+  });
+
+  it('returns a JSON-serializable tree with text/range fidelity', () => {
+    const source = 'echo "你好 🎉" | grep 你';
+    const result = service.parse(source);
+    if (!result.ok) {
+      throw new Error('expected ok');
+    }
+    const roundTripped = JSON.parse(JSON.stringify(result.root)) as BashSyntaxNode;
+    const check = (node: BashSyntaxNode): void => {
+      expect(node.text).toBe(source.slice(node.startIndex, node.endIndex));
+      node.children.forEach(check);
+    };
+    check(roundTripped);
+  });
+});
diff --git a/packages/tree-sitter-bash/README.md b/packages/tree-sitter-bash/README.md
new file mode 100644
index 0000000000..8ef128d62f
--- /dev/null
+++ b/packages/tree-sitter-bash/README.md
@@ -0,0 +1,273 @@
+# @moonshot-ai/tree-sitter-bash
+
+A pure-TypeScript bash parser that produces a syntax tree whose named node
+types match [tree-sitter-bash](https://github.com/tree-sitter/tree-sitter-bash)
+0.25.0 one-to-one, built for agent-side command permission analysis.
+
+- No native addons; offsets are UTF-16 code units (`node.text` is always a
+  direct `source.slice(startIndex, endIndex)`).
+- Parsing runs under a hard budget and never throws: budget exhaustion
+  returns `{ ok: false, reason: 'aborted' }`, malformed input returns a
+  degraded tree with `hasError: true`.
+- Correctness is enforced by a differential test suite that parses hundreds
+  of samples — plus the complete official tree-sitter-bash 0.25.0 corpus —
+  with both this parser and the real tree-sitter-bash (wasm) and compares
+  the trees byte-for-byte (see `test/fixtures/`). The only remaining
+  deviations are the documented ones listed under
+  [Known differences](#known-differences-from-tree-sitter-bash) below.
+
+## API
+
+```ts
+import { parse } from '@moonshot-ai/tree-sitter-bash';
+
+const result = parse('git status && rm -rf /');
+if (result.ok) {
+  // result.rootNode: program → list → command …
+}
+```
+
+- `parse(source: string, options?: ParseOptions): ParseResult`
+- `ParseResult` is one of:
+  - `{ ok: true; rootNode: SyntaxNode; hasError: boolean }`
+  - `{ ok: false; reason: 'aborted' }`
+- `ParseOptions`: `{ timeoutMs?: number; maxNodes?: number }` (defaults
+  50 ms / 50 000 nodes; `timeoutMs: Infinity` disables the time check).
+- `SyntaxNode`: `{ type, text, startIndex, endIndex, isNamed, parent,
+  children, namedChildren }`. Named node types come from tree-sitter-bash's
+  `node-types.json`; `children` includes the anonymous (punctuation /
+  keyword token) nodes as well, `namedChildren` only the named ones, and
+  `descendantsOfType(root, ...types)` walks named descendants.
+- Depth caps (pathological nesting degrades locally — ERROR nodes,
+  `hasError: true` — instead of overflowing the stack):
+  - `MAX_SUBSTITUTION_DEPTH = 150` for the $( … ) / `…` / <( … )
+    sub-parser chain (the deepest per level, ~13 frames — measured stack
+    overflow at ~380–500 levels on a default Node stack, so the cap keeps
+    a ≥2.5× margin);
+  - `MAX_PARSE_DEPTH = 500` for subshell/compound/`${…}`/expression
+    nesting (verified safe at those caps);
+  - `MAX_SCAN_DEPTH = 1024` for the lexer's own scan recursion.
+
+## Budget semantics
+
+The budget caps total work, not input size: a multi-hundred-KB string or
+heredoc body parses fine (it produces only a handful of nodes), but a
+source whose tree would exceed 50 000 nodes — a megabyte-long `case`
+statement, tens of thousands of arithmetic expressions — is aborted under
+the defaults. Every created node ticks the budget; long character-level
+scan loops re-check the deadline periodically. When either limit is hit,
+`parse` returns `{ ok: false, reason: 'aborted' }` and the abort is prompt
+(see Performance). Pass `timeoutMs` / `maxNodes` to raise the limits.
+
+## Error recovery
+
+Malformed input never throws either. Unterminated constructs (quotes,
+expansions, substitutions, heredocs, compound commands) are kept as
+partial nodes and flagged with `hasError: true`; tokens that cannot start
+or continue a statement (stray `)`, a leading `&&`, …) are wrapped in
+`ERROR` nodes and parsing continues. Newlines never appear in the tree:
+statement-terminating `\n`s produce no nodes at all (matching
+tree-sitter-bash, whose scanner only emits `\n` tokens in heredoc
+position), while `;` / `&` terminators are anonymous children. As a
+last-resort guard against parser bugs, any unexpected internal exception
+degrades to a `program` root with a single `ERROR` child spanning the
+source and `hasError: true` — callers still get a usable tree.
+
+## Known differences from tree-sitter-bash
+
+Named node types always come from tree-sitter-bash's `node-types.json`, but
+for the following constructs the tree shape deliberately deviates from what
+tree-sitter-bash 0.25.0 produces (each verified against the real parser,
+and pinned by a stored expectation in `test/fixtures/` — see
+`test/helpers/known-differences.ts`):
+
+- `<>` (read-write redirect) is parsed as a normal `file_redirect`
+  operator; tree-sitter-bash 0.25.0 fails to parse it.
+- Several heredocs on one line (`cat <<A <<B`) cannot be represented as an
+  overlap-free tree (their bodies interleave with the redirect nodes).
+  tree-sitter-bash errors; this parser degrades the second `<<` to an
+  `ERROR` node, skips its body registration, and parses the "body" lines as
+  ordinary commands, mirroring the reference's recovery shape. The same
+  applies to a `<<` inside a heredoc's line tail.
+- Statements after a heredoc on the same line (`cat <<EOF; echo x`, also
+  `&`) are absorbed into the `heredoc_redirect` node so the tree stays
+  overlap-free once the body (which textually follows the whole line) is
+  attached. tree-sitter-bash errors on the whole line.
+- Inside an unquoted `heredoc_body`, every plain text chunk — including the
+  leading one — becomes a `heredoc_content` node, and backtick command
+  substitutions are parsed as `command_substitution` children.
+  tree-sitter-bash's scanner hides the leading chunk and swallows backticks
+  into the content.
+- A `$'` sequence inside an unquoted heredoc body is plain text here
+  (bash does not expand ANSI-C strings in heredocs; the reference does the
+  same mid-line). Quirk: when a body line STARTS with `$'…'`, the
+  reference's scanner errors and drops the whole `heredoc_redirect`.
+- A heredoc redirect at statement start (`<<EOF cat`) parses normally;
+  tree-sitter-bash errors.
+- Unterminated or invalid constructs keep their partial nodes with
+  `hasError: true` (see above); tree-sitter-bash degrades them to `ERROR`
+  nodes, and the exact recovery shape differs. This covers invalid compound
+  commands — a fallthrough terminator on the last case item
+  (`a) x ;;& esac`), a missing separator before a compound keyword
+  (`if a; then b fi`, `{ ls }` — both flagged `hasError: true` here), a
+  heredoc inside an if/while condition, `foo () ls`, a regex with an
+  unquoted space after `=~` — and small recovery details:
+  - An unterminated test command always keeps its partial `test_command`
+    with a zero-width closer here. The reference only inserts that
+    zero-width closer for an unterminated `[` or for a `[[` ending in a
+    complete unary test (`[[ -f file`); other unterminated `[[` forms
+    (`[[ $a`, `[[ a == b`) become an `ERROR` node with no closer there.
+    (An empty `[[ ]]` sets the error flag in BOTH parsers — the reference
+    recovers by inserting the missing closer as a zero-width token — and
+    this parser matches its tree exactly: a concatenation of two `]` word
+    pieces plus a zero-width `]]`.)
+  - An unterminated compound command gets no synthetic keyword here
+    (`if a; then b;` ends after the last statement with `hasError: true`);
+    the reference inserts zero-width keyword nodes (a zero-width `fi`).
+  - `${x@}` (transformation operator with no letter) keeps an `expansion`
+    with `hasError: true` here; the reference splits it into an `ERROR`
+    node and a `}` command.
+- A trailing connector (`ls &&`, `ls |`) yields a single-child
+  `list` / `pipeline` with `hasError: true`; tree-sitter-bash inserts a
+  zero-width `command` recovery node.
+- An empty backtick substitution (`` `` ``), or a pair containing only
+  whitespace, is a `command_substitution` with no statements;
+  tree-sitter-bash treats the two backticks as a single `` `` `` token
+  that is only valid inside a concatenation and errors in argument
+  position.
+- An empty command substitution (`$( )`) is a clean `command_substitution`
+  with no statements here; the reference inserts a zero-width `command`
+  recovery node (and flags `hasError`).
+- In arithmetic, a hex literal is a `number` (`$((0x1F))` → number
+  "0x1F"); the reference's arithmetic number token does not cover hex and
+  produces a `variable_name` "0x1F" instead (a reference quirk — bash does
+  evaluate hex in arithmetic).
+- `[[ ((a) == x) && y ]]` (parenthesized test group followed by `&&`)
+  parses cleanly here as nested `parenthesized_expression`s; the reference
+  mis-reads it as an `arithmetic_expansion` with an embedded `ERROR` node
+  (a reference quirk — the valid `[[ ((a)) == x ]]` form matches exactly).
+- An extglob group the reference rejects — one that directly follows a
+  pure literal or a dot (`x@(y|z)w`, `*.@(a|b)`, `_@(y)`, `*@(y)`,
+  `.@(y)`) — is an error there (an `ERROR` node inside the comparison);
+  this parser ends the pattern at the group and recovers with a different
+  error shape. Groups the reference ACCEPTS — at the start of the right
+  side (`+(a|b)`) or after glob characters (`*/@(default).vim`,
+  `*_@(LIB|SYMLINK)`, `@(LLD|GNU\ ld)*`, `*([0-9])([0-9])`, `\"+(?)\"`) —
+  match exactly, as do right sides split around one substitution or quote
+  (`*${var}*`, `*/${a}*`, `*"7f 45"*`).
+- An extglob group the reference rejects in a case pattern (`x@(y))`, and
+  the same mechanism for `.(` (`x.(y))`, `*.(a|b))` — `.` is not a sigil,
+  but the reference rejects it identically): the reference reparses the
+  group as the start of a new case item; this parser degrades the whole
+  pattern to an `ERROR` node. (Accepted groups — `*(a|b))`,
+  `*([0-9])([0-9]))` — match exactly.)
+- A single-dash short-option case pattern (`-o)`) is a `word` — the form
+  this parser always produces — but the reference's classification flips
+  with leftover scanner state (a first case item right after the `in`
+  line's newline parses it as an `extglob_pattern`).
+- A case pattern directly after a line continuation (`| \<newline>`) is a
+  `word` in the reference even when its content would make it a glob
+  (`cmake_modules` — a scanner-state quirk); this parser classifies
+  patterns by content, so such a pattern is an `extglob_pattern` here. A
+  continuation INSIDE a pattern (`cont\<newline>ued)`) is an error in the
+  reference (word "cont" + `ERROR` "ued", `hasError: true`), and a
+  continuation fused with a digit (`a\<newline>1)`) is an
+  `extglob_pattern` there; this parser keeps the continuation and parses
+  the whole pattern cleanly as one `word` — which is also how bash itself
+  reads it.
+- A comparison right side with TWO substitutions or quotes
+  (`*${x}*${y}`, `*"s"*"t"`) does not fit the reference's one-construct
+  pattern rule either; it recovers with a nested `binary_expression`
+  there, while this parser keeps a single `concatenation` right side.
+- A few `[[ … ]]` comparison right-hand sides deviate:
+  - A negative decimal operand (`[[ $n == -0.5 ]]`) is a
+    `unary_expression` (`-` over word "0.5") in the reference; this parser
+    produces a single `extglob_pattern` "-0.5" (the unary-minus reading
+    only covers integers here).
+  - A test operator fused with an expansion (`[[ -x$f && … ]]`) is a
+    `unary_expression` (`-` over a `concatenation` of `x` and `$f`) in the
+    reference; this parser produces a flat `concatenation` of `-x` and
+    `$f` with no unary wrapper.
+  - An escaped `|` inside a pattern (`[[ $a == foo\|bar ]]`) splits the
+    expression in the reference — `binary_expression(==, extglob "foo\")`
+    then `|` then word "bar" (a reference quirk); this parser keeps one
+    `extglob_pattern` "foo\|bar".
+- `${!# }` and `${!## }` (and `${!##/}`) are pathological expansion forms:
+  the reference recovers with zero-width operator tokens, this parser with
+  a flagged partial `expansion`; the shapes differ. (The sane forms
+  `${#}`, `${!#}`, `${!##}`, `${#!}` all match exactly.)
+- A base prefix fused with an expansion (`10#${x}`) is one `number` node
+  spanning the expansion in the reference (a quirk); this parser produces
+  a `concatenation` of word "10#" and the expansion.
+- `${v:-(default)}` (a parenthesized default value) is a plain `word`
+  here; the reference parses it as an `array` node (a quirk).
+- `${=1}` (zsh-style word splitting flag) produces a `variable_name` "1"
+  in the reference; this parser produces a `word`.
+- An escaped space or tab between arguments is dropped by the reference
+  as invisible extras (its tree has gaps: `echo 1 \ 2` produces number
+  "1" and number "2", and `echo a\<TAB>b` splits into word "a" and word
+  "b"); this parser keeps the escape inside the word. (An escaped space
+  INSIDE a word, `a\ b`, is kept by both parsers and matches.)
+- A `$` directly fused with a backtick substitution (`` $`echo x` ``) is
+  one `command_substitution` with a `$`+backtick token in the reference;
+  this parser produces a `($)` token followed by the substitution.
+- A non-ASCII “identifier” in assignment position (`变量=值`, `é=1`) is a
+  `variable_assignment` flagged `hasError` in the reference; this parser
+  keeps it a plain command word — which is also how bash itself treats it
+  (variable names are ASCII-only, so the text runs as a command).
+- A negative literal after a compound assignment in a c-style for header
+  (`for (( … j *= -1, … ))`) is folded into a `number` "-1" by the
+  reference; this parser produces a `unary_expression` (`-` over number
+  "1") — matching what the reference itself produces everywhere else.
+- A double backslash in a replacement value (`${x// /\\|}`) loses its
+  first backslash in the reference's tree (a quirk — the `word` starts
+  one character later); this parser keeps both.
+- `string_content` is not split at newlines (tree-sitter-bash's scanner
+  splits it).
+
+For completeness, a few constructs that LOOK like deviations but are not —
+the reference behaves the same way (verified):
+
+- `coproc` has no grammar support in tree-sitter-bash 0.25.0 and parses as
+  an ordinary command; so does this parser.
+- `select` is parsed as a `for_statement` with a `select` keyword token —
+  that is exactly the reference's tree (there is no `select_statement`
+  node type).
+- `$"…"` (translated string): in command-argument position it is an
+  anonymous `$` argument followed by the `string` argument; everywhere
+  else (assignment values, redirect destinations, for-loop values, …) it
+  is a `translated_string` node. Both match the reference exactly.
+  `$'…'` is a single-child-free `ansi_c_string` node, also matching.
+- Brace expansion: only `{N..M}` (unsigned digits) is a `brace_expression`;
+  `{a..z}`, `{1..10..2}` and `{-5..5}` are plain word concatenations in the
+  reference and here.
+- Statement-position `((word++…))` (inner text starting with a bare
+  identifier plus a postfix `++`/`--`) is a `test_command` in the
+  reference — its statement-position arithmetic grammar does not accept
+  that form — while `((x = 1))`, `((b + a++))`, `((a[i]++))` and
+  argument-position `((a++))` are `arithmetic_expansion`. This parser
+  matches all of these, including after `!`.
+- Command-prefix redirects take exactly one destination (`> a cmd x` makes
+  `cmd` the `command_name`), while redirects after the command name consume
+  every following word (`cmd > out arg` puts both words in the redirect) —
+  this matches tree-sitter-bash's actual disambiguation.
+
+## Performance
+
+Measured on an Apple-silicon MacBook (Node 24, default 50 ms / 50 000-node
+budget):
+
+- A typical one-line command (`git status && rm -rf /`): ~4 µs per parse.
+- A 100 KB realistic deployment script (functions, loops, redirects,
+  expansions, heredocs): tens of milliseconds per parse (~20–60 ms across
+  runs and machines) — the same order of magnitude as the default 50 ms
+  budget, so such inputs may return `ok: true` or a prompt abort depending
+  on the machine; raise `timeoutMs` when parsing whole scripts.
+- A 500 KB heredoc body: parses within the default budget (the tree has
+  only a handful of nodes).
+- The abort path is prompt: a 400 KB node-budget bomb aborts in ~20 ms
+  (hard guarantee asserted by the test suite: < 100 ms).
+
+These numbers are smoke-tested as orders of magnitude in
+`test/performance.test.ts` to guard against accidental quadratic
+complexity; absolute timings vary by machine.
diff --git a/packages/tree-sitter-bash/package.json b/packages/tree-sitter-bash/package.json
new file mode 100644
index 0000000000..dab858aa31
--- /dev/null
+++ b/packages/tree-sitter-bash/package.json
@@ -0,0 +1,47 @@
+{
+  "name": "@moonshot-ai/tree-sitter-bash",
+  "version": "0.1.0",
+  "private": true,
+  "description": "A pure-TypeScript bash parser producing a syntax tree whose named node types match tree-sitter-bash one-to-one.",
+  "license": "MIT",
+  "author": "Moonshot AI",
+  "homepage": "https://github.com/MoonshotAI/kimi-code/tree/main/packages/tree-sitter-bash#readme",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/MoonshotAI/kimi-code.git",
+    "directory": "packages/tree-sitter-bash"
+  },
+  "bugs": {
+    "url": "https://github.com/MoonshotAI/kimi-code/issues"
+  },
+  "keywords": [
+    "kimi",
+    "bash",
+    "shell",
+    "parser",
+    "tree-sitter"
+  ],
+  "files": [
+    "dist"
+  ],
+  "type": "module",
+  "imports": {
+    "#/*": "./src/*.ts"
+  },
+  "exports": {
+    ".": {
+      "types": "./src/index.ts",
+      "default": "./src/index.ts"
+    }
+  },
+  "scripts": {
+    "build": "tsdown",
+    "typecheck": "tsc -p tsconfig.json --noEmit",
+    "test": "vitest run",
+    "clean": "rm -rf dist"
+  },
+  "devDependencies": {
+    "tree-sitter-bash": "0.25.0",
+    "web-tree-sitter": "0.25.10"
+  }
+}
diff --git a/packages/tree-sitter-bash/src/budget.ts b/packages/tree-sitter-bash/src/budget.ts
new file mode 100644
index 0000000000..35ffe9db76
--- /dev/null
+++ b/packages/tree-sitter-bash/src/budget.ts
@@ -0,0 +1,69 @@
+// src/budget.ts
+//
+// Parse budget: a hard cap on wall-clock time and on the number of syntax
+// nodes a single parse may create. The parser calls `budget.tick()` every
+// time it creates a node; long scan loops (word runs, strings, heredoc
+// bodies) call `budget.progress()` periodically so a pathological single
+// token still hits the deadline without inflating the node count. When
+// either limit is exceeded the methods throw `Aborted`, which the `parse`
+// entry point catches and reports as `{ ok: false, reason: 'aborted' }`.
+
+export const DEFAULT_TIMEOUT_MS = 50;
+export const DEFAULT_MAX_NODES = 50_000;
+
+/** Internal control-flow error. Never escapes the `parse` entry point. */
+export class Aborted extends Error {
+  constructor(message: string) {
+    super(message);
+    this.name = 'Aborted';
+  }
+}
+
+export interface BudgetOptions {
+  /** Wall-clock limit in milliseconds. `Infinity` disables the time check. */
+  timeoutMs?: number;
+  /** Maximum number of nodes the parse may create. */
+  maxNodes?: number;
+}
+
+export class ParseBudget {
+  private readonly deadline: number;
+  private readonly maxNodes: number;
+  private nodeCount = 0;
+
+  constructor(options: BudgetOptions = {}) {
+    const timeoutMs = options.timeoutMs ?? DEFAULT_TIMEOUT_MS;
+    this.deadline = Date.now() + timeoutMs;
+    this.maxNodes = options.maxNodes ?? DEFAULT_MAX_NODES;
+  }
+
+  /** Number of nodes created so far. */
+  get nodesUsed(): number {
+    return this.nodeCount;
+  }
+
+  /**
+   * Account for one created node and re-check the deadline. Throws `Aborted`
+   * when the node cap is exceeded or the deadline has been reached.
+   */
+  tick(): void {
+    this.nodeCount++;
+    if (this.nodeCount > this.maxNodes) {
+      throw new Aborted(`parse aborted: node budget exceeded (${this.nodeCount} > ${this.maxNodes})`);
+    }
+    if (Date.now() >= this.deadline) {
+      throw new Aborted(`parse aborted: timeout`);
+    }
+  }
+
+  /**
+   * Re-check the deadline WITHOUT counting a node. For long scan loops that
+   * run many iterations per produced node (character-level scanning), called
+   * at intervals so the deadline is still enforced promptly.
+   */
+  progress(): void {
+    if (Date.now() >= this.deadline) {
+      throw new Aborted(`parse aborted: timeout`);
+    }
+  }
+}
diff --git a/packages/tree-sitter-bash/src/grammar.ts b/packages/tree-sitter-bash/src/grammar.ts
new file mode 100644
index 0000000000..8a16268490
--- /dev/null
+++ b/packages/tree-sitter-bash/src/grammar.ts
@@ -0,0 +1,134 @@
+// src/grammar.ts
+//
+// Static grammar tables for bash: operator tables, keyword sets and
+// variable-name sets shared by the lexer and the parser.
+
+/** Single-character special parameters: $@ $* $# $? $- $$ $! $0 and $_. */
+export const SPECIAL_VARIABLES = ['@', '*', '#', '?', '-', '$', '!', '0', '_'] as const;
+
+/** Special parameter characters that may follow `$` directly, as a string
+ *  for `includes` checks. Derived from SPECIAL_VARIABLES by dropping `0`
+ *  and `_`, which are word characters matched by the `\w+` rule instead. */
+export const SPECIAL_VARIABLE_CHARS = SPECIAL_VARIABLES.filter((ch) => !/\w/.test(ch)).join('');
+
+/** Operators that open a `file_redirect` (heredoc and herestring operators
+ *  are handled separately). `<>` is included even though tree-sitter-bash
+ *  0.25.0 fails to parse it — it is a real bash operator (`exec 3<>file`). */
+export const FILE_REDIRECT_OPERATORS = ['<', '>', '>>', '>&', '<&', '&>', '&>>', '>|', '<>', '>&-', '<&-'] as const;
+
+/** Keywords that open a declaration_command at statement position. */
+export const DECLARATION_COMMAND_KEYWORDS = ['declare', 'typeset', 'export', 'readonly', 'local'] as const;
+
+/** Keywords that open an unset_command at statement position. */
+export const UNSET_COMMAND_KEYWORDS = ['unset', 'unsetenv'] as const;
+
+/** Reserved words that may not be used as a function name in the
+ *  `name() { ...; }` form (they are keywords in command position). */
+export const RESERVED_WORDS = [
+  'if',
+  'then',
+  'elif',
+  'else',
+  'fi',
+  'while',
+  'until',
+  'do',
+  'done',
+  'for',
+  'select',
+  'in',
+  'case',
+  'esac',
+  'function',
+  '{',
+  '}',
+  '!',
+  ...DECLARATION_COMMAND_KEYWORDS,
+  ...UNSET_COMMAND_KEYWORDS,
+] as const;
+
+/** Infix operators inside arithmetic / test expressions with their
+ *  tree-sitter-bash precedence level (grammar.js PREC table; higher binds
+ *  tighter). `test_operator` binaries sit at level 10 (TEST) and are handled
+ *  separately. Ternary `? :` is level 2, postfix `++`/`--` level 18, prefix
+ *  `++`/`--` level 17, prefix `!`/`~`/`+`/`-` level 11. */
+export const EXPRESSION_PRECEDENCE: Readonly<Record<string, number>> = {
+  '+=': 0,
+  '-=': 0,
+  '*=': 0,
+  '/=': 0,
+  '%=': 0,
+  '**=': 0,
+  '<<=': 0,
+  '>>=': 0,
+  '&=': 0,
+  '^=': 0,
+  '|=': 0,
+  '=': 1,
+  '=~': 1,
+  '||': 3,
+  '&&': 4,
+  '|': 5,
+  '^': 6,
+  '&': 7,
+  '==': 8,
+  '!=': 8,
+  '<': 9,
+  '<=': 9,
+  '>': 9,
+  '>=': 9,
+  '<<': 12,
+  '>>': 12,
+  '+': 13,
+  '-': 13,
+  '*': 14,
+  '/': 14,
+  '%': 14,
+  '**': 15,
+};
+
+/** All operator texts the expression lexer recognizes, longest first so
+ *  matching is greedy (`**=` before `**` before `*`). `;` and `,` only act
+ *  as separators (c-style for headers and comma lists), not as operators. */
+export const EXPRESSION_OPERATORS = [
+  '**=',
+  '<<=',
+  '>>=',
+  '++',
+  '--',
+  '**',
+  '<<',
+  '>>',
+  '<=',
+  '>=',
+  '==',
+  '!=',
+  '&&',
+  '||',
+  '+=',
+  '-=',
+  '*=',
+  '/=',
+  '%=',
+  '&=',
+  '^=',
+  '|=',
+  '=~',
+  '=',
+  '+',
+  '-',
+  '*',
+  '/',
+  '%',
+  '^',
+  '&',
+  '|',
+  '<',
+  '>',
+  '!',
+  '~',
+  '?',
+  ':',
+  ',',
+  ';',
+] as const;
diff --git a/packages/tree-sitter-bash/src/index.ts b/packages/tree-sitter-bash/src/index.ts
new file mode 100644
index 0000000000..2737d6529e
--- /dev/null
+++ b/packages/tree-sitter-bash/src/index.ts
@@ -0,0 +1,4 @@
+export * from './node';
+export * from './budget';
+export * from './parse';
+export * from './grammar';
diff --git a/packages/tree-sitter-bash/src/lexer.ts b/packages/tree-sitter-bash/src/lexer.ts
new file mode 100644
index 0000000000..807deb7ee9
--- /dev/null
+++ b/packages/tree-sitter-bash/src/lexer.ts
@@ -0,0 +1,629 @@
+// src/lexer.ts
+//
+// Hand-written bash tokenizer. Produces a flat token stream of:
+//
+//   word      a maximal run of adjacent word material: bare characters,
+//             '...' / "..." quotes, $var / ${...} / $(...) / `...` expansions,
+//             <(...) / >(...) process substitutions. Quotes and substitutions
+//             are skipped over with nesting awareness; the parser re-scans
+//             the token range to build the actual sub-tree. The single
+//             characters { } [ ] are emitted as their own word tokens (they
+//             are not word characters in bash), matching tree-sitter-bash's
+//             _special_character behaviour.
+//   op        statement operators (&& || | |& ; ;; & ( )) and redirect
+//             operators (< > >> >& <& &> &>> >| <> >&- <&- <<< << <<-).
+//   io_number a run of digits immediately followed by < or > — the file
+//             descriptor prefix of a redirect (2>/dev/null).
+//   newline   a statement-terminating \n. Carries any heredoc bodies that
+//             were queued on the lexer and are scanned right after the line.
+//   comment   # ... to end of line (only at token start; # inside a word is
+//             a plain word character).
+//   eof       end of the lexer's range. Also carries pending heredoc bodies.
+//
+// Additionally:
+//   - `;&` and `;;&` are single op tokens (case_item fallthrough
+//     terminators).
+//   - `((...))` at token start is scanned as one word token (arithmetic
+//     command / c-style for header), so the parser can re-scan the range
+//     with its expression parser; the word ends right after the balanced
+//     close and does not merge with following word characters.
+//   - peekAt(n) looks arbitrarily far ahead (function_definition
+//     detection); reposition(pos) rewinds the stream (test commands parse
+//     their [[ ... ]] range character-wise and then resume tokenizing past
+//     the closer).
+//
+// The lexer is range-bounded: sub-parsers lex $(...) / `...` bodies with
+// their own Lexer over the same source but a narrower [start, end) window.
+//
+// Heredoc queue: the parser registers a HeredocSpec for every << / <<- it
+// accepts. When the lexer produces the next newline (or eof) token it scans
+// one body per queued spec, in registration order, and attaches them to the
+// token; the parser completes the matching heredoc_redirect nodes when it
+// consumes that token.
+//
+// Budget: the lexer never creates nodes, so it only checks the deadline —
+// once per produced token and periodically (every SCAN_TICK_INTERVAL
+// characters) inside every long scan loop (word runs, quote/paren skipping,
+// comments, blanks, heredoc bodies) — via budget.progress(), so a
+// pathological single token cannot starve the deadline check. Node counting
+// (budget.tick()) is the parser's job.
+
+import type { ParseBudget } from '#/budget';
+import { SPECIAL_VARIABLE_CHARS } from '#/grammar';
+
+export type TokenType = 'word' | 'op' | 'io_number' | 'newline' | 'comment' | 'eof';
+
+export interface Token {
+  readonly type: TokenType;
+  readonly start: number;
+  readonly end: number;
+  /** Heredoc bodies scanned when this newline/eof token was produced. */
+  readonly heredocBodies: HeredocBody[];
+}
+
+export interface HeredocSpec {
+  /** Delimiter text with any quoting removed — what must appear at line start. */
+  readonly delimiter: string;
+  /** `<<-`: leading tabs of the first body line are stripped. */
+  readonly stripTabs: boolean;
+  /** The delimiter was quoted ('EOF', "EOF", \EOF): the body is not expanded. */
+  readonly quoted: boolean;
+}
+
+export interface HeredocBody {
+  /** Start of the heredoc_body node (after first-line tab stripping). */
+  readonly bodyStart: number;
+  /** End of the heredoc_body node: right before the end marker, so the final
+   *  newline and any tabs preceding the marker stay inside the body. */
+  readonly bodyEnd: number;
+  /** Start of the heredoc_end marker (the delimiter word itself). */
+  readonly endStart: number;
+  /** End of the heredoc_end marker. */
+  readonly endEnd: number;
+  /** False when the delimiter line never appeared: body runs to the end of
+   *  the lexer's range and there is no heredoc_end. */
+  readonly found: boolean;
+}
+
+/** How often long scan loops tick the budget, in scanned characters. */
+const SCAN_TICK_INTERVAL = 2048;
+
+/**
+ * Depth cap for the scanBalanced ↔ skipDoubleQuoted mutual recursion
+ * (`${a#"${a#"…` nests two scan frames per level). Beyond the cap the scan
+ * reports "unbalanced", letting the parser degrade the construct with
+ * hasError instead of overflowing the call stack. Sized above the parser's
+ * own nesting guards (which trigger first for parser-driven recursion);
+ * this cap only protects lexer-driven scans of pathological input.
+ */
+const MAX_SCAN_DEPTH = 1024;
+
+const CONTROL_OPERATORS = ['&>>', '&>', '&&', '&', '|&', '||', '|', ';;&', ';;', ';&', ';', '(', ')'] as const;
+const REDIRECT_OPERATORS = ['<<-', '<<<', '<<', '<&-', '<&', '<>', '<', '>&-', '>&', '>>', '>|', '>'] as const;
+
+function isWordChar(ch: string | undefined): boolean {
+  return ch !== undefined && /[\w]/.test(ch);
+}
+
+function isBlank(ch: string | undefined): boolean {
+  return ch === ' ' || ch === '\t' || ch === '\r';
+}
+
+function isDigitAt(source: string, i: number): boolean {
+  const ch = source[i]!;
+  return ch >= '0' && ch <= '9';
+}
+
+/** Skip a "..." quoted region starting at `i` (which points at the opening
+ *  quote). Returns the index just past the closing quote, or `end` when the
+ *  string is unterminated. Substitution-aware: $(...), ${...} and `...`
+ *  inside the string may themselves contain quotes. `depth` tracks the
+ *  scanBalanced ↔ skipDoubleQuoted recursion (see MAX_SCAN_DEPTH). */
+export function skipDoubleQuoted(source: string, budget: ParseBudget, i: number, end: number, depth = 0): number {
+  if (depth >= MAX_SCAN_DEPTH) return end;
+  let j = i + 1;
+  let sinceTick = 0;
+  while (j < end) {
+    if (++sinceTick >= SCAN_TICK_INTERVAL) {
+      budget.progress();
+      sinceTick = 0;
+    }
+    const ch = source[j]!;
+    if (ch === '\\') {
+      j += 2;
+      continue;
+    }
+    if (ch === '"') return j + 1;
+    if (ch === '`') {
+      j = skipBacktick(source, budget, j, end);
+      continue;
+    }
+    if (ch === '$') {
+      const next = source[j + 1];
+      if (next === '(') {
+        j = scanBalancedStatements(source, budget, j + 1, end, depth + 1).end;
+        continue;
+      }
+      if (next === '{') {
+        j = scanBalanced(source, budget, j + 1, end, '{', '}', depth + 1).end;
+        continue;
+      }
+    }
+    j++;
+  }
+  return end;
+}
+
+/** Skip a '...' region starting at `i`. No escapes exist in raw strings. */
+export function skipSingleQuoted(source: string, _budget: ParseBudget, i: number, end: number): number {
+  const close = source.indexOf("'", i + 1);
+  if (close === -1 || close >= end) return end;
+  return close + 1;
+}
+
+/** Skip a `...` region starting at `i`; \` is an escaped backtick. */
+export function skipBacktick(source: string, budget: ParseBudget, i: number, end: number): number {
+  let j = i + 1;
+  let sinceTick = 0;
+  while (j < end) {
+    if (++sinceTick >= SCAN_TICK_INTERVAL) {
+      budget.progress();
+      sinceTick = 0;
+    }
+    const ch = source[j]!;
+    if (ch === '\\') {
+      j += 2;
+      continue;
+    }
+    if (ch === '`') return j + 1;
+    j++;
+  }
+  return end;
+}
+
+/** Result of scanning a balanced region. */
+export interface BalancedScan {
+  /** Index just past the matching close, or `end` when unbalanced. */
+  end: number;
+  /** Whether the matching close was actually found. */
+  balanced: boolean;
+}
+
+/** Scan a balanced open/close region starting at `i` (which points at the
+ *  opening character). Quote- and escape-aware. `depth` tracks the
+ *  scanBalanced ↔ skipDoubleQuoted recursion (see MAX_SCAN_DEPTH). */
+export function scanBalanced(
+  source: string,
+  budget: ParseBudget,
+  i: number,
+  end: number,
+  open: string,
+  close: string,
+  depth = 0,
+): BalancedScan {
+  if (depth >= MAX_SCAN_DEPTH) return { end, balanced: false };
+  let nesting = 0;
+  let j = i;
+  let sinceTick = 0;
+  while (j < end) {
+    if (++sinceTick >= SCAN_TICK_INTERVAL) {
+      budget.progress();
+      sinceTick = 0;
+    }
+    const ch = source[j]!;
+    if (ch === '\\') {
+      j += 2;
+      continue;
+    }
+    if (ch === open) {
+      nesting++;
+    } else if (ch === close) {
+      nesting--;
+      if (nesting === 0) return { end: j + 1, balanced: true };
+    } else if (ch === '"') {
+      j = skipDoubleQuoted(source, budget, j, end, depth + 1);
+      continue;
+    } else if (ch === "'") {
+      j = skipSingleQuoted(source, budget, j, end);
+      continue;
+    } else if (ch === '`') {
+      j = skipBacktick(source, budget, j, end);
+      continue;
+    }
+    j++;
+  }
+  return { end, balanced: false };
+}
+
+/** Words after which a `case` word opens a case_statement (statement
+ *  position) rather than being an ordinary argument. */
+const CASE_ENABLING_WORDS: ReadonlySet<string> = new Set(['if', 'then', 'elif', 'else', 'while', 'until', 'do']);
+
+/**
+ * Case-aware variant of scanBalanced for `(` … `)` regions that may hold
+ * full statements (command/process substitution bodies). A naive paren
+ * count closes the region early on a case_item pattern (`a) 1;; esac`):
+ * every item's `)` would decrement the depth. While a `case` is open (seen
+ * in statement position, not yet closed by `esac`), a `)` at the paren
+ * depth where the `case` started is a pattern close and does NOT count;
+ * parens from other constructs inside the case (subshells, $(…), the
+ * optional `(` of an item) balance themselves normally.
+ *
+ * Statement position is approximated lexically: `case` counts as a keyword
+ * after `(`, `;`, `&`, `|`, a newline, `{`, or one of the compound
+ * keywords in CASE_ENABLING_WORDS — so `echo case` does not confuse the
+ * scan. `esac` pops the innermost open case regardless of position (the
+ * reference scanner emits the esac token even in argument position).
+ */
+export function scanBalancedStatements(
+  source: string,
+  budget: ParseBudget,
+  i: number,
+  end: number,
+  depth = 0,
+): BalancedScan {
+  if (depth >= MAX_SCAN_DEPTH) return { end, balanced: false };
+  let nesting = 0;
+  /** Paren depths at which each open case_statement started. */
+  const caseDepths: number[] = [];
+  let j = i;
+  /** What preceded the current position: 'start' | 'sep' | 'keyword' | 'word'. */
+  let previous: 'start' | 'sep' | 'keyword' | 'word' = 'start';
+  let sinceTick = 0;
+  while (j < end) {
+    if (++sinceTick >= SCAN_TICK_INTERVAL) {
+      budget.progress();
+      sinceTick = 0;
+    }
+    const ch = source[j]!;
+    if (ch === '\\') {
+      j += 2;
+      continue;
+    }
+    if (ch === '"') {
+      j = skipDoubleQuoted(source, budget, j, end, depth + 1);
+      previous = 'word';
+      continue;
+    }
+    if (ch === "'") {
+      j = skipSingleQuoted(source, budget, j, end);
+      previous = 'word';
+      continue;
+    }
+    if (ch === '`') {
+      j = skipBacktick(source, budget, j, end);
+      previous = 'word';
+      continue;
+    }
+    if (ch === ' ' || ch === '\t' || ch === '\r') {
+      j++;
+      continue;
+    }
+    if (ch === '\n' || ch === ';' || ch === '&' || ch === '|') {
+      previous = 'sep';
+      j++;
+      continue;
+    }
+    if (ch === '{') {
+      previous = 'sep';
+      j++;
+      continue;
+    }
+    if (ch === '(') {
+      nesting++;
+      previous = 'sep';
+      j++;
+      continue;
+    }
+    if (ch === ')') {
+      if (caseDepths.length > 0 && caseDepths.at(-1) === nesting) {
+        // A case_item pattern close: not a paren.
+        previous = 'sep';
+        j++;
+        continue;
+      }
+      nesting--;
+      if (nesting === 0) return { end: j + 1, balanced: true };
+      previous = 'word';
+      j++;
+      continue;
+    }
+    // A word run: check for the case/esac keywords.
+    let k = j;
+    while (k < end && isWordChar(source[k])) k++;
+    if (k > j) {
+      const word = source.slice(j, k);
+      if (word === 'esac' && caseDepths.length > 0) {
+        caseDepths.pop();
+        previous = 'sep';
+      } else if (word === 'case' && previous !== 'word') {
+        caseDepths.push(nesting);
+        previous = 'word';
+      } else {
+        previous = CASE_ENABLING_WORDS.has(word) ? 'keyword' : 'word';
+      }
+      j = k;
+      continue;
+    }
+    previous = 'word';
+    j++;
+  }
+  return { end, balanced: false };
+}
+
+/** Skip a $-construct starting at `i` (which points at the `$`). Handles
+ *  $(...), $((...)), ${...}, $'...' (escape-aware: \' does not close),
+ *  $name and the single-character specials. A `$` followed by anything else
+ *  (including a double quote) consumes just the `$`. */
+export function skipDollar(source: string, budget: ParseBudget, i: number, end: number): number {
+  const next = source[i + 1];
+  if (next === '(') return scanBalancedStatements(source, budget, i + 1, end).end;
+  if (next === '{') return scanBalanced(source, budget, i + 1, end, '{', '}').end;
+  // $[...] legacy arithmetic (only when `[` directly follows the `$`; `$a[0]`
+  // takes the word-character branch above).
+  if (next === '[') return scanBalanced(source, budget, i + 1, end, '[', ']').end;
+  if (next === "'") {
+    // ANSI-C string: \' is an escaped quote and does not terminate it.
+    let j = i + 2;
+    let sinceTick = 0;
+    while (j < end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        budget.progress();
+        sinceTick = 0;
+      }
+      const ch = source[j]!;
+      if (ch === '\\') {
+        j += 2;
+        continue;
+      }
+      if (ch === "'") return j + 1;
+      j++;
+    }
+    return end;
+  }
+  if (isWordChar(next)) {
+    let j = i + 1;
+    while (j < end && isWordChar(source[j])) j++;
+    return j;
+  }
+  if (next !== undefined && SPECIAL_VARIABLE_CHARS.includes(next)) return i + 2;
+  return i + 1;
+}
+
+export class Lexer {
+  /** Current scan position; exposed for the parser's heredoc bookkeeping. */
+  pos: number;
+  private lookahead: Token[] = [];
+  private readonly pendingHeredocs: HeredocSpec[] = [];
+
+  constructor(
+    private readonly source: string,
+    private readonly budget: ParseBudget,
+    start = 0,
+    private readonly end = source.length,
+  ) {
+    this.pos = start;
+  }
+
+  /** End of the lexer's range (exclusive). */
+  get rangeEnd(): number {
+    return this.end;
+  }
+
+  /** Queue a heredoc body to be scanned after the current line. */
+  queueHeredoc(spec: HeredocSpec): void {
+    this.pendingHeredocs.push(spec);
+  }
+
+  peek(): Token {
+    return this.peekAt(0);
+  }
+
+  /** Look `n` tokens ahead (0 = the next token). Tokens are scanned
+   *  on demand and buffered, so lookahead never re-scans. */
+  peekAt(n: number): Token {
+    while (this.lookahead.length <= n) this.lookahead.push(this.scanToken());
+    return this.lookahead[n]!;
+  }
+
+  next(): Token {
+    if (this.lookahead.length > 0) return this.lookahead.shift()!;
+    return this.scanToken();
+  }
+
+  /** Rewind the stream to `pos`, discarding buffered lookahead. Used when
+   *  the parser consumed a region character-wise (test commands). */
+  reposition(pos: number): void {
+    this.pos = pos;
+    this.lookahead = [];
+  }
+
+  text(token: Token): string {
+    return this.source.slice(token.start, token.end);
+  }
+
+  private scanToken(): Token {
+    this.budget.progress();
+    this.skipBlanks();
+    const start = this.pos;
+    if (this.pos >= this.end) return this.scanBoundary('eof', start, start);
+    const ch = this.source[this.pos]!;
+    if (ch === '\n') {
+      this.pos++;
+      return this.scanBoundary('newline', start, this.pos);
+    }
+    if (ch === '#') {
+      let sinceTick = 0;
+      while (this.pos < this.end && this.source[this.pos] !== '\n') {
+        if (++sinceTick >= SCAN_TICK_INTERVAL) {
+          this.budget.progress();
+          sinceTick = 0;
+        }
+        this.pos++;
+      }
+      return { type: 'comment', start, end: this.pos, heredocBodies: [] };
+    }
+    if (ch === '<' || ch === '>') {
+      // <( / >( start a process substitution, which is word material. The
+      // heredoc operators were already excluded: <<( is << + ( …
+      if (this.source[this.pos + 1] === '(') return this.scanWord();
+      return this.scanOp(REDIRECT_OPERATORS);
+    }
+    if (ch === '&' || ch === '|' || ch === ';' || ch === '(' || ch === ')') {
+      // `((` at token start opens arithmetic (an arithmetic command or a
+      // c-style for header), which is word material, not two parens.
+      if (ch === '(' && this.source[this.pos + 1] === '(') return this.scanWord();
+      return this.scanOp(CONTROL_OPERATORS);
+    }
+    if (ch === '{' || ch === '}' || ch === '[' || ch === ']') {
+      this.pos++;
+      return { type: 'word', start, end: this.pos, heredocBodies: [] };
+    }
+    if (ch >= '0' && ch <= '9') {
+      let i = this.pos;
+      while (i < this.end && isDigitAt(this.source, i)) i++;
+      const next = this.source[i];
+      if (next === '<' || next === '>') {
+        this.pos = i;
+        return { type: 'io_number', start, end: i, heredocBodies: [] };
+      }
+      return this.scanWord();
+    }
+    return this.scanWord();
+  }
+
+  /** Produce a newline/eof token, scanning any queued heredoc bodies that
+   *  start right after it. */
+  private scanBoundary(type: 'newline' | 'eof', start: number, end: number): Token {
+    const bodies: HeredocBody[] = [];
+    while (this.pendingHeredocs.length > 0) {
+      bodies.push(this.readHeredocBody(this.pendingHeredocs.shift()!));
+    }
+    return { type, start, end, heredocBodies: bodies };
+  }
+
+  private skipBlanks(): void {
+    let sinceTick = 0;
+    for (;;) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[this.pos];
+      if (isBlank(ch)) {
+        this.pos++;
+        continue;
+      }
+      // Line continuation is whitespace, not word material.
+      if (ch === '\\' && this.source[this.pos + 1] === '\n') {
+        this.pos += 2;
+        continue;
+      }
+      return;
+    }
+  }
+
+  private scanOp(table: readonly string[]): Token {
+    const start = this.pos;
+    for (const op of table) {
+      if (this.source.startsWith(op, this.pos) && this.pos + op.length <= this.end) {
+        this.pos += op.length;
+        return { type: 'op', start, end: this.pos, heredocBodies: [] };
+      }
+    }
+    // Unreachable for the callers above, but never loop forever.
+    this.pos++;
+    return { type: 'op', start, end: this.pos, heredocBodies: [] };
+  }
+
+  private scanWord(): Token {
+    const start = this.pos;
+    // `((...))` at token start: one word token ending right after the
+    // balanced close (the parser re-scans the range as arithmetic).
+    if (this.source[start] === '(' && this.source[start + 1] === '(') {
+      this.pos = scanBalanced(this.source, this.budget, start, this.end, '(', ')').end;
+      return { type: 'word', start, end: this.pos, heredocBodies: [] };
+    }
+    let i = this.pos;
+    let sinceTick = 0;
+    while (i < this.end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[i]!;
+      if (isBlank(ch) || ch === '\n') break;
+      if (ch === '&' || ch === '|' || ch === ';' || ch === '(' || ch === ')') break;
+      if (ch === '{' || ch === '}' || ch === '[' || ch === ']') break;
+      if (ch === '<' || ch === '>') {
+        if (this.source[i + 1] === '(') {
+          i = scanBalancedStatements(this.source, this.budget, i + 1, this.end).end;
+          continue;
+        }
+        break;
+      }
+      if (ch === '\\') {
+        // A line continuation ends the run (it acts as whitespace); a lone
+        // trailing backslash at end of range is consumed as word text.
+        if (this.source[i + 1] === '\n') break;
+        i += 2;
+        continue;
+      }
+      if (ch === '"') {
+        i = skipDoubleQuoted(this.source, this.budget, i, this.end);
+        continue;
+      }
+      if (ch === "'") {
+        i = skipSingleQuoted(this.source, this.budget, i, this.end);
+        continue;
+      }
+      if (ch === '`') {
+        i = skipBacktick(this.source, this.budget, i, this.end);
+        continue;
+      }
+      if (ch === '$') {
+        i = skipDollar(this.source, this.budget, i, this.end);
+        continue;
+      }
+      i++;
+    }
+    if (i === start) i++; // defensive: never emit a zero-width word token
+    this.pos = i;
+    return { type: 'word', start, end: i, heredocBodies: [] };
+  }
+
+  /**
+   * Scan one heredoc body, starting at the current position (right after the
+   * newline that ended the command line). Matches tree-sitter-bash's layout:
+   * for `<<-` only the first body line's leading tabs are stripped; the end
+   * marker is the bare delimiter word and any tabs before it belong to the
+   * body.
+   */
+  private readHeredocBody(spec: HeredocSpec): HeredocBody {
+    let bodyStart = this.pos;
+    if (spec.stripTabs) {
+      while (bodyStart < this.end && this.source[bodyStart] === '\t') bodyStart++;
+    }
+    let lineStart = this.pos;
+    while (lineStart < this.end) {
+      this.budget.progress();
+      let marker = lineStart;
+      if (spec.stripTabs) {
+        while (marker < this.end && this.source[marker] === '\t') marker++;
+      }
+      if (spec.delimiter.length > 0 && this.source.startsWith(spec.delimiter, marker)) {
+        const after = marker + spec.delimiter.length;
+        if (after >= this.end || this.source[after] === '\n') {
+          this.pos = after;
+          return { bodyStart, bodyEnd: marker, endStart: marker, endEnd: after, found: true };
+        }
+      }
+      const newline = this.source.indexOf('\n', lineStart);
+      if (newline === -1) break;
+      lineStart = newline + 1;
+    }
+    // Unterminated: the body runs to the end of this lexer's range.
+    this.pos = this.end;
+    return { bodyStart, bodyEnd: this.end, endStart: this.end, endEnd: this.end, found: false };
+  }
+}
diff --git a/packages/tree-sitter-bash/src/node.ts b/packages/tree-sitter-bash/src/node.ts
new file mode 100644
index 0000000000..09e0001892
--- /dev/null
+++ b/packages/tree-sitter-bash/src/node.ts
@@ -0,0 +1,121 @@
+// src/node.ts
+//
+// Syntax tree node model. Named node types correspond one-to-one with
+// tree-sitter-bash's named node types, so consumers can write tree queries
+// against either implementation. Two deliberate deviations from tree-sitter:
+//
+//   - startIndex / endIndex are UTF-16 code unit offsets (native JS string
+//     indexing), not tree-sitter's UTF-8 byte offsets. `node.text` therefore
+//     always equals `source.slice(node.startIndex, node.endIndex)`.
+//   - `text` is pre-computed when the node is created instead of being
+//     re-sliced on every access.
+//
+// Child semantics: `children` contains every child in source order, including
+// anonymous (punctuation / keyword token) nodes; `namedChildren` contains only
+// the named ones, in the same relative order. Anonymous nodes never appear in
+// `namedChildren`. `descendantsOfType` walks named descendants only, matching
+// how tree-sitter queries see the tree.
+
+export interface SyntaxNode {
+  /** Node type, e.g. 'program', 'command', 'word'. Matches tree-sitter-bash. */
+  readonly type: string;
+  /** Source text covered by this node (UTF-16 slice of the original source). */
+  readonly text: string;
+  /** Start offset in UTF-16 code units, inclusive. */
+  readonly startIndex: number;
+  /** End offset in UTF-16 code units, exclusive. */
+  readonly endIndex: number;
+  /** Whether this is a named node (false for punctuation/keyword tokens). */
+  readonly isNamed: boolean;
+  readonly parent: SyntaxNode | null;
+  /** All children in source order, named and anonymous. */
+  readonly children: readonly SyntaxNode[];
+  /** Named children only, in source order. */
+  readonly namedChildren: readonly SyntaxNode[];
+}
+
+export interface NodeInit {
+  type: string;
+  source: string;
+  startIndex: number;
+  endIndex: number;
+  isNamed?: boolean;
+}
+
+/**
+ * Mutable node under construction. The parser builds trees with this class and
+ * exposes them through the readonly `SyntaxNode` interface; once a node is
+ * handed out it must be treated as immutable.
+ */
+export class SyntaxNodeBuilder {
+  readonly type: string;
+  readonly text: string;
+  readonly startIndex: number;
+  readonly endIndex: number;
+  readonly isNamed: boolean;
+  parent: SyntaxNodeBuilder | null = null;
+  readonly children: SyntaxNodeBuilder[] = [];
+  readonly namedChildren: SyntaxNodeBuilder[] = [];
+
+  constructor(init: NodeInit) {
+    if (init.startIndex < 0 || init.endIndex < init.startIndex || init.endIndex > init.source.length) {
+      throw new RangeError(
+        `invalid node range [${init.startIndex}, ${init.endIndex}) for source of length ${init.source.length}`,
+      );
+    }
+    this.type = init.type;
+    this.startIndex = init.startIndex;
+    this.endIndex = init.endIndex;
+    this.isNamed = init.isNamed ?? true;
+    this.text = init.source.slice(init.startIndex, init.endIndex);
+  }
+
+  /** Attach a child, wiring its parent pointer. Named children are also added
+   *  to `namedChildren`. Returns the child for chaining.
+   *
+   *  Children must lie inside the parent's range and be appended in source
+   *  order without overlapping the previous sibling (a zero-width child may
+   *  start exactly where the previous sibling ends). */
+  addChild<T extends SyntaxNodeBuilder>(child: T): T {
+    if (child.parent !== null) throw new Error(`node '${child.type}' already has a parent`);
+    if (child.startIndex < this.startIndex || child.endIndex > this.endIndex) {
+      throw new RangeError(
+        `child '${child.type}' [${child.startIndex}, ${child.endIndex}) escapes parent '${this.type}' [${this.startIndex}, ${this.endIndex})`,
+      );
+    }
+    const last = this.children.at(-1);
+    if (last !== undefined && child.startIndex < last.endIndex) {
+      throw new RangeError(
+        `child '${child.type}' [${child.startIndex}, ${child.endIndex}) overlaps sibling '${last.type}' [${last.startIndex}, ${last.endIndex})`,
+      );
+    }
+    child.parent = this;
+    this.children.push(child);
+    if (child.isNamed) this.namedChildren.push(child);
+    return child;
+  }
+}
+
+/** Convenience factory for a detached node. */
+export function createNode(init: NodeInit): SyntaxNodeBuilder {
+  return new SyntaxNodeBuilder(init);
+}
+
+/**
+ * Pre-order traversal of the named descendants of `root` (not including
+ * `root` itself), filtered to the given types. With no types, returns every
+ * named descendant in pre-order. Iterative with an explicit stack so that
+ * pathologically deep trees cannot overflow the call stack.
+ */
+export function descendantsOfType(root: SyntaxNode, ...types: string[]): SyntaxNode[] {
+  const wanted = types.length > 0 ? new Set(types) : null;
+  const out: SyntaxNode[] = [];
+  const stack: SyntaxNode[] = [];
+  for (let i = root.namedChildren.length - 1; i >= 0; i--) stack.push(root.namedChildren[i]!);
+  while (stack.length > 0) {
+    const node = stack.pop()!;
+    if (wanted === null || wanted.has(node.type)) out.push(node);
+    for (let i = node.namedChildren.length - 1; i >= 0; i--) stack.push(node.namedChildren[i]!);
+  }
+  return out;
+}
diff --git a/packages/tree-sitter-bash/src/parse.ts b/packages/tree-sitter-bash/src/parse.ts
new file mode 100644
index 0000000000..1e28a83c9d
--- /dev/null
+++ b/packages/tree-sitter-bash/src/parse.ts
@@ -0,0 +1,46 @@
+// src/parse.ts
+//
+// Public parse entry point. Runs the hand-written lexer + recursive-descent
+// parser under a ParseBudget and returns the materialized syntax tree.
+//
+// Exit contract:
+//   - Budget exhaustion (time or node cap) → { ok: false, reason: 'aborted' }.
+//     `Aborted` never escapes.
+//   - Malformed input → { ok: true, hasError: true } with ERROR nodes and/or
+//     partial nodes where the parser recovered.
+//   - Any other parser-internal exception (a bug, not user input) is caught
+//     here as a last resort: the caller still gets a usable tree — a program
+//     root with a single ERROR child spanning the whole source and
+//     hasError: true — instead of an exception. Chosen over reporting
+//     'aborted' because the parse did not hit its budget; downstream
+//     consumers keep working on a degraded tree and hasError signals the
+//     failure. (See the catch block below.)
+
+import { Aborted, ParseBudget } from '#/budget';
+import type { BudgetOptions } from '#/budget';
+import { SyntaxNodeBuilder } from '#/node';
+import type { SyntaxNode } from '#/node';
+import { Parser, materialize } from '#/parser';
+
+export type ParseResult =
+  | { ok: true; rootNode: SyntaxNode; hasError: boolean }
+  | { ok: false; reason: 'aborted' };
+
+export type ParseOptions = BudgetOptions;
+
+export function parse(source: string, options: ParseOptions = {}): ParseResult {
+  const budget = new ParseBudget(options);
+  try {
+    const parser = new Parser(source, budget);
+    const root = parser.parseProgram();
+    const rootNode = materialize(root, source);
+    return { ok: true, rootNode, hasError: parser.hasError };
+  } catch (error) {
+    if (error instanceof Aborted) return { ok: false, reason: 'aborted' };
+    // Last-resort guard for parser bugs: degrade to an ERROR root instead of
+    // throwing into the caller (see the file header for why).
+    const root = new SyntaxNodeBuilder({ type: 'program', source, startIndex: 0, endIndex: source.length });
+    root.addChild(new SyntaxNodeBuilder({ type: 'ERROR', source, startIndex: 0, endIndex: source.length }));
+    return { ok: true, rootNode: root, hasError: true };
+  }
+}
diff --git a/packages/tree-sitter-bash/src/parser.ts b/packages/tree-sitter-bash/src/parser.ts
new file mode 100644
index 0000000000..be37427c4b
--- /dev/null
+++ b/packages/tree-sitter-bash/src/parser.ts
@@ -0,0 +1,3787 @@
+// src/parser.ts
+//
+// Recursive-descent bash parser. One method per tree-sitter-bash grammar
+// rule (parseList ↔ list, parsePipeline ↔ pipeline, parseCommand ↔ command,
+// …), producing node ranges and child layouts that match the real
+// tree-sitter-bash 0.25.0 tree.
+//
+// Construction strategy: the parser builds lightweight `Frame`s, not
+// SyntaxNodeBuilders, because heredoc bodies are scanned only when the line
+// ends — a heredoc_redirect node (and every ancestor) gets its final range
+// after the fact. Frames are mutable and carry parent pointers so the
+// heredoc drain can extend the ancestor chain. `materialize` converts the
+// finished frame tree into SyntaxNodeBuilders iteratively (explicit stack,
+// safe for pathologically deep trees) once parsing is done.
+//
+// Error recovery: unterminated constructs (quote, expansion, substitution,
+// heredoc, compound commands) keep their partial node and set hasError;
+// tokens that cannot start or continue a statement are wrapped in an ERROR
+// node and parsing continues. No parser-internal exception is expected to
+// escape; parse() still guards against it.
+//
+// Depth guards, all capped so pathological nesting degrades locally
+// (ERROR nodes, hasError) instead of overflowing the call stack:
+//   - word-level substitutions ($( … ) / `…` / <( … )) recurse via fresh
+//     sub-Parser instances bounded to their source range — the deepest
+//     chain per level (~13 frames), capped by MAX_SUBSTITUTION_DEPTH;
+//   - subshells and compound commands (if/while/for/case/{…}/function
+//     bodies) recurse within one parser (`scopeDepth`);
+//   - the parseLiteral ↔ parseString ↔ parseExpansion chain behind ${ … }
+//     nesting — including the pattern → string → $ cycle — has its own
+//     counter (`literalDepth`, incremented by both parseLiteral and
+//     parseExpansion);
+//   - parenthesized_expression nesting inside arithmetic/test expressions
+//     uses `exprDepth`.
+// The last three are capped by MAX_PARSE_DEPTH; beyond the caps the
+// construct degrades (compound keywords fall back to plain commands,
+// expressions and substitutions to ERROR nodes). The lexer's own scan
+// recursion (scanBalanced ↔ skipDoubleQuoted) is cheaper per level and
+// capped separately (MAX_SCAN_DEPTH in lexer.ts).
+//
+// Expression engine: arithmetic expansions ($((…)), ((…)), $[…]), c-style
+// for headers and test commands ([[ … ]] / [ … ]) share a small Pratt
+// parser (parseExpression over an ExprState) driven by the precedence table
+// in grammar.ts, mirroring grammar.js's PREC levels and the tree shapes they
+// produce (e.g. prefix -/+/~/! grab a full tighter-precedence expression:
+// $((-x + ~y)) is unary(-, binary(x, +, unary(~, y))) in the reference).
+
+import type { ParseBudget } from '#/budget';
+import {
+  DECLARATION_COMMAND_KEYWORDS,
+  EXPRESSION_OPERATORS,
+  EXPRESSION_PRECEDENCE,
+  FILE_REDIRECT_OPERATORS,
+  RESERVED_WORDS,
+  SPECIAL_VARIABLE_CHARS,
+  UNSET_COMMAND_KEYWORDS,
+} from '#/grammar';
+import { Lexer, scanBalanced, scanBalancedStatements, skipBacktick, skipDollar, skipDoubleQuoted, skipSingleQuoted } from '#/lexer';
+import type { BalancedScan, HeredocBody, HeredocSpec, Token } from '#/lexer';
+import { SyntaxNodeBuilder } from '#/node';
+
+/** Maximum nesting of scopes (subshells, compound commands, expressions,
+ *  ${…} chains). These chains cost only a handful of stack frames per
+ *  level and their guards skip iteratively, so 500 is comfortably safe —
+ *  verified: inputs thousands of levels deep degrade locally instead of
+ *  overflowing. */
+export const MAX_PARSE_DEPTH = 500;
+
+/**
+ * Maximum nesting of the $( … ) / `…` / <( … ) sub-parser chain (each
+ * level spawns a fresh Parser and costs ~13 stack frames through
+ * parseCommandSubstitution → parseScopedStatements → parseStatementList →
+ * parseCommand → parseLiteral → parseDollar). Measured overflow on a
+ * default Node stack: ~380–500 levels depending on environment, BEFORE the
+ * old 500 cap could fire. 150 keeps a ≥2.5× margin; beyond it the
+ * substitution degrades to a local ERROR node (hasError) instead of
+ * taking the whole tree down via a RangeError.
+ */
+export const MAX_SUBSTITUTION_DEPTH = 150;
+
+const FILE_REDIRECT_OP_SET: ReadonlySet<string> = new Set(FILE_REDIRECT_OPERATORS);
+const DECLARATION_COMMAND_SET: ReadonlySet<string> = new Set(DECLARATION_COMMAND_KEYWORDS);
+const UNSET_COMMAND_SET: ReadonlySet<string> = new Set(UNSET_COMMAND_KEYWORDS);
+const RESERVED_WORD_SET: ReadonlySet<string> = new Set(RESERVED_WORDS);
+const NUMBER_RE = /^-?(0x)?[0-9]+(#[0-9A-Za-z@_]+)?$/;
+const ASSIGNMENT_RE = /^[A-Za-z_][A-Za-z0-9_]*\+?=/;
+const ASSIGNMENT_SPLIT_RE = /^([A-Za-z_][A-Za-z0-9_]*)(\+?=)/;
+const SUBSCRIPT_ASSIGNMENT_RE = /^(\w+)\[([^\]\n]*)\](\+?=)/;
+const IDENTIFIER_RE = /^[A-Za-z_]\w*$/;
+/** Function names additionally allow `:` (`foo::bar() { … }` — the
+ *  reference accepts any such word as a function name). */
+const FUNCTION_NAME_RE = /^[A-Za-z_][\w:]*$/;
+const BRACE_EXPRESSION_RE = /^\{(\d+)\.\.(\d+)\}/;
+/** Statement-position `((…))` that tree-sitter-bash parses as a
+ *  test_command: inner text starting with an identifier-ish token
+ *  (letters/digits/`_`/`-`) ending in `++`/`--`. */
+const PAREN_TEST_RE = /^\(\(\s*[A-Za-z_][\w-]*(\+\+|--)/;
+const STOP_THEN: ReadonlySet<string> = new Set(['then']);
+const STOP_DO: ReadonlySet<string> = new Set(['do']);
+const STOP_DONE: ReadonlySet<string> = new Set(['done']);
+const STOP_IF_BODY: ReadonlySet<string> = new Set(['elif', 'else', 'fi']);
+const STOP_FI: ReadonlySet<string> = new Set(['fi']);
+const STOP_CLOSE_BRACE: ReadonlySet<string> = new Set(['}']);
+const STOP_ESAC: ReadonlySet<string> = new Set(['esac']);
+const CASE_TERMINATION_OPS: ReadonlySet<string> = new Set([';;', ';&', ';;&']);
+
+/** How often long scan loops tick the budget, in scanned characters. */
+const SCAN_TICK_INTERVAL = 2048;
+
+// Precedence levels from grammar.js's PREC table that are not in
+// EXPRESSION_PRECEDENCE.
+const PREC_TERNARY = 2;
+const PREC_TEST = 10;
+const PREC_UNARY = 11;
+const PREC_PREFIX = 17;
+const PREC_POSTFIX = 18;
+
+/** Mutable node under construction; see the file header. */
+export interface Frame {
+  type: string;
+  start: number;
+  end: number;
+  isNamed: boolean;
+  parent: Frame | null;
+  children: Frame[];
+}
+
+interface PendingHeredoc {
+  frame: Frame;
+  spec: HeredocSpec;
+}
+
+interface StatementListOptions {
+  readonly stopAtParen?: boolean;
+  readonly stopWords?: ReadonlySet<string>;
+  readonly stopOps?: ReadonlySet<string>;
+  /** True when the grammar requires a terminator before the stop keyword
+   *  (if/while/do bodies, `{ … }`): stopping with a pending statement and
+   *  no separator flags an error (`if a; then b fi` is an error in the
+   *  reference too). Case items and subshells do not require one. */
+  readonly terminatorRequired?: boolean;
+}
+
+/** Where the expression engine runs: arithmetic expansions, c-style for
+ *  headers, or test commands. */
+type ExprMode = 'arith' | 'c' | 'test';
+
+type ExprTokenKind =
+  | 'number'
+  | 'ident'
+  | 'word'
+  | 'subst'
+  | 'string'
+  | 'testop'
+  | 'op'
+  | 'lparen'
+  | 'rparen'
+  | 'unknown'
+  | 'end';
+
+interface ExprToken {
+  readonly kind: ExprTokenKind;
+  readonly start: number;
+  readonly end: number;
+  readonly text: string;
+  /** Pre-built node for subst/string/subscript tokens. */
+  readonly frame?: Frame;
+}
+
+interface ExprState {
+  pos: number;
+  readonly end: number;
+  readonly mode: ExprMode;
+  lookahead: ExprToken | null;
+  /** Nesting of parenthesized_expression inside test commands (the
+   *  reference emits word, not extglob_pattern, for a non-glob ==/!=
+   *  right-hand side inside parentheses). */
+  parenDepth: number;
+  /** Test mode only: true after a complete operand, where `=`/`!`-family
+   *  characters are comparison operators even when attached to the next
+   *  word (`a ==b` is `a == b`); false at operand position, where they are
+   *  word characters (`=b`, `!x`, `a!=b` are words in the reference). */
+  expectOperator: boolean;
+}
+
+function isFileRedirectOp(text: string): boolean {
+  return FILE_REDIRECT_OP_SET.has(text);
+}
+
+/** Strip quoting from a heredoc delimiter word; any quote or backslash in
+ *  the raw text marks the body as quoted (no expansions). */
+function extractHeredocSpec(raw: string, stripTabs: boolean): HeredocSpec {
+  let delimiter = '';
+  let quoted = false;
+  for (let i = 0; i < raw.length; i++) {
+    const ch = raw[i]!;
+    if (ch === '\\' && i + 1 < raw.length) {
+      delimiter += raw[i + 1];
+      quoted = true;
+      i++;
+    } else if (ch === '"' || ch === "'") {
+      quoted = true;
+    } else {
+      delimiter += ch;
+    }
+  }
+  return { delimiter, stripTabs, quoted };
+}
+
+export class Parser {
+  /** Set when any recovery path was taken. */
+  hasError = false;
+  private lexer!: Lexer;
+  private readonly heredocQueue: PendingHeredoc[] = [];
+  private scopeDepth = 0;
+  /** Recursion depth of parseLiteral ↔ parseString / parseExpansion (the
+   *  ${} nesting chain that stays inside this Parser instance). */
+  private literalDepth = 0;
+  /** Nesting of parenthesized_expression inside the expression parser. */
+  private exprDepth = 0;
+  /** True while parsing inside a heredoc's line tail: a second heredoc
+   *  cannot be represented and degrades to ERROR (see parseHeredocRedirect). */
+  private noHeredoc = false;
+  /** Nesting of case_item statement lists (see parseCaseItemStatements). */
+  private caseItemDepth = 0;
+  /** End of the most recently consumed `;`/`&`/`;;`/newline terminator in
+   *  any statement list. case_item uses it to extend its range over a
+   *  trailing terminator before `esac`, matching the reference (whose
+   *  last_case_item includes the final _terminator in its range). */
+  private lastTerminatorEnd = 0;
+
+  constructor(
+    private readonly source: string,
+    private readonly budget: ParseBudget,
+    private readonly depth = 0,
+  ) {}
+
+  // ---------------------------------------------------------------- helpers
+
+  private frame(type: string, start: number, end: number, children: Frame[] = [], isNamed = true): Frame {
+    this.budget.tick();
+    const frame: Frame = { type, start, end, isNamed, parent: null, children };
+    for (const child of children) child.parent = frame;
+    return frame;
+  }
+
+  private anon(type: string, start: number, end: number): Frame {
+    return this.frame(type, start, end, [], false);
+  }
+
+  private addKid(parent: Frame, child: Frame): void {
+    child.parent = parent;
+    parent.children.push(child);
+  }
+
+  private text(start: number, end: number): string {
+    return this.source.slice(start, end);
+  }
+
+  private tokenText(token: Token): string {
+    return this.source.slice(token.start, token.end);
+  }
+
+  private endOf(kids: Frame[], fallback: number): number {
+    return kids.length > 0 ? kids.at(-1)!.end : fallback;
+  }
+
+  private isStatementStart(token: Token): boolean {
+    if (token.type === 'word' || token.type === 'io_number') return true;
+    if (token.type !== 'op') return false;
+    const text = this.tokenText(token);
+    return text === '(' || text === '<<<' || text === '<<' || text === '<<-' || isFileRedirectOp(text);
+  }
+
+  /** Peeked word token whose text equals `keyword`. */
+  private peekKeyword(keyword: string): boolean {
+    const token = this.lexer.peek();
+    return token.type === 'word' && this.tokenText(token) === keyword;
+  }
+
+  /** Consume a keyword as an anonymous child; false (no consume) when the
+   *  next token is not that keyword. */
+  private consumeKeyword(kids: Frame[], keyword: string): boolean {
+    if (!this.peekKeyword(keyword)) return false;
+    const token = this.lexer.next();
+    kids.push(this.anon(keyword, token.start, token.end));
+    return true;
+  }
+
+  // ------------------------------------------------------------ entry point
+
+  /** program: the whole source as one statement list. */
+  parseProgram(): Frame {
+    this.lexer = new Lexer(this.source, this.budget, 0, this.source.length);
+    const children = this.parseStatementList();
+    return this.frame('program', 0, this.source.length, children);
+  }
+
+  /** Parse a sub-range as a statement list (body of $( … ), ` … `, <( … )).
+   *  The sub-parser chain is the deepest per-level recursion in the
+   *  parser, so it has its own, tighter cap — see MAX_SUBSTITUTION_DEPTH. */
+  private parseScopedStatements(start: number, end: number): Frame[] {
+    if (this.depth + 1 >= MAX_SUBSTITUTION_DEPTH) {
+      this.hasError = true;
+      return [this.frame('ERROR', start, end)];
+    }
+    const sub = new Parser(this.source, this.budget, this.depth + 1);
+    sub.lexer = new Lexer(this.source, this.budget, start, end);
+    const children = sub.parseStatementList();
+    if (sub.hasError) this.hasError = true;
+    return children;
+  }
+
+  // -------------------------------------------------------- statement lists
+
+  /**
+   * _statements: statements separated/terminated by ; & and newlines.
+   * `;`/`&`/`;;` terminators become anonymous children; newlines and
+   * comments do not (comments stay named children) — tree-sitter-bash never
+   * emits `\n` terminator nodes either. Parsing stops (without consuming)
+   * at a word in stopWords (then/do/done/fi/esac/…), an op in stopOps
+   * (case_item terminators), or a `)` when stopAtParen is set.
+   */
+  private parseStatementList(options: StatementListOptions = {}): Frame[] {
+    const children: Frame[] = [];
+    let needTerminator = false;
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'eof') {
+        this.completeHeredocs(token.heredocBodies);
+        this.failOpenHeredocs();
+        break;
+      }
+      if (token.type === 'newline') {
+        this.lexer.next();
+        this.completeHeredocs(token.heredocBodies);
+        this.lastTerminatorEnd = token.end;
+        needTerminator = false;
+        continue;
+      }
+      if (token.type === 'comment') {
+        this.lexer.next();
+        children.push(this.frame('comment', token.start, token.end));
+        continue;
+      }
+      if (token.type === 'word' && options.stopWords?.has(this.tokenText(token)) === true) {
+        if (needTerminator && options.terminatorRequired === true) this.hasError = true; // missing separator
+        break;
+      }
+      const op = token.type === 'op' ? this.tokenText(token) : '';
+      if (token.type === 'op' && options.stopOps?.has(op) === true) break;
+      if (token.type === 'op' && op === ')' && options.stopAtParen === true) {
+        this.failOpenHeredocs();
+        break;
+      }
+      if (token.type === 'op' && (op === ';' || op === '&' || op === ';;')) {
+        this.lexer.next();
+        children.push(this.anon(op, token.start, token.end));
+        this.lastTerminatorEnd = token.end;
+        needTerminator = false;
+        continue;
+      }
+      if (
+        token.type === 'op' &&
+        (op === ')' || op === '&&' || op === '||' || op === '|' || op === '|&' || op === ';&' || op === ';;&')
+      ) {
+        // A binary operator or closer with no left-hand side: recover.
+        this.hasError = true;
+        this.lexer.next();
+        children.push(this.frame('ERROR', token.start, token.end, [this.anon(op, token.start, token.end)]));
+        needTerminator = false;
+        continue;
+      }
+      if (needTerminator) this.hasError = true; // missing separator, e.g. `cmd (sub)`
+      children.push(this.parseList());
+      needTerminator = true;
+    }
+    return children;
+  }
+
+  /** Consume newlines and comments in operator-continuation position
+   *  (`a &&\n b`). Returns the comment nodes; newlines are dropped. */
+  private skipContinuation(): Frame[] {
+    const comments: Frame[] = [];
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'newline') {
+        this.lexer.next();
+        this.completeHeredocs(token.heredocBodies);
+        continue;
+      }
+      if (token.type === 'comment') {
+        this.lexer.next();
+        comments.push(this.frame('comment', token.start, token.end));
+        continue;
+      }
+      return comments;
+    }
+  }
+
+  // ------------------------------------------------------------- statements
+
+  /** list: statement ((&& | ||) statement)* — left associative. */
+  private parseList(): Frame {
+    let left = this.parsePipeline();
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type !== 'op') break;
+      const op = this.tokenText(token);
+      if (op !== '&&' && op !== '||') break;
+      this.lexer.next();
+      const extras = this.skipContinuation();
+      if (this.isStatementStart(this.lexer.peek())) {
+        const right = this.parsePipeline();
+        left = this.frame('list', left.start, right.end, [left, this.anon(op, token.start, token.end), ...extras, right]);
+      } else {
+        // Trailing connector (`ls &&`): keep the partial list, flag it.
+        this.hasError = true;
+        left = this.frame('list', left.start, token.end, [left, this.anon(op, token.start, token.end)]);
+        break;
+      }
+    }
+    return left;
+  }
+
+  /** pipeline: statement ((&#124; | &#124;&) statement)* */
+  private parsePipeline(): Frame {
+    return this.parsePipelineTail(this.parseStatementNotPipeline());
+  }
+
+  private parsePipelineTail(first: Frame): Frame {
+    const kids: Frame[] = [first];
+    let end = first.end;
+    let pipes = 0;
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type !== 'op') break;
+      const op = this.tokenText(token);
+      if (op !== '|' && op !== '|&') break;
+      this.lexer.next();
+      pipes++;
+      kids.push(this.anon(op, token.start, token.end));
+      end = token.end;
+      kids.push(...this.skipContinuation());
+      if (this.isStatementStart(this.lexer.peek())) {
+        const next = this.parseStatementNotPipeline();
+        kids.push(next);
+        end = next.end;
+      } else {
+        this.hasError = true; // dangling pipe (`ls |`)
+        break;
+      }
+    }
+    if (pipes === 0) return first;
+    return this.frame('pipeline', first.start, end, kids);
+  }
+
+  /** _statement_not_pipeline: any statement form plus trailing redirects. */
+  private parseStatementNotPipeline(): Frame {
+    let inner: Frame | null = this.parseStatementCore();
+    const trailing: Frame[] = [];
+    for (;;) {
+      const next = this.lexer.peek();
+      if (next.type === 'io_number') {
+        trailing.push(this.parseRedirect());
+        continue;
+      }
+      if (next.type === 'op') {
+        const op = this.tokenText(next);
+        if (isFileRedirectOp(op) || op === '<<<' || op === '<<' || op === '<<-') {
+          trailing.push(this.parseRedirect());
+          continue;
+        }
+      }
+      break;
+    }
+    if (trailing.length > 0) {
+      const kids = inner === null ? trailing : [inner, ...trailing];
+      inner = this.frame('redirected_statement', kids[0]!.start, kids.at(-1)!.end, kids);
+    }
+    if (inner === null) {
+      // Unreachable from well-formed dispatch; recover without looping.
+      this.hasError = true;
+      const skipped = this.lexer.next();
+      inner = this.frame('ERROR', skipped.start, skipped.end);
+    }
+    return inner;
+  }
+
+  /**
+   * Dispatch one statement (no pipeline, no trailing redirects) on its
+   * leading token: subshell, negation, compound commands, test commands,
+   * declaration/unset commands, function definitions, or a plain command.
+   * Reserved words only count at statement position — `echo if` keeps `if`
+   * as a word, and a prefix assignment (`x=1 if …`) disables the keyword
+   * reading just like in the reference.
+   */
+  private parseStatementCore(): Frame | null {
+    const token = this.lexer.peek();
+    if (token.type === 'op' && this.tokenText(token) === '(') {
+      return this.parseSubshell();
+    }
+    if (token.type === 'word') {
+      const text = this.tokenText(token);
+      if (text === '!') return this.parseNegatedCommand();
+      if (text === '{') {
+        // `{1..3}` at statement position is a brace_expression word.
+        if (!BRACE_EXPRESSION_RE.test(this.source.slice(token.start))) {
+          return this.parseCompoundGuarded(() => this.parseCompoundStatement());
+        }
+      } else if (text === 'if') {
+        return this.parseCompoundGuarded(() => this.parseIfStatement());
+      } else if (text === 'while' || text === 'until') {
+        return this.parseCompoundGuarded(() => this.parseWhileStatement());
+      } else if (text === 'for' || text === 'select') {
+        return this.parseCompoundGuarded(() => this.parseForStatement());
+      } else if (text === 'case') {
+        return this.parseCompoundGuarded(() => this.parseCaseStatement());
+      } else if (text === 'function') {
+        return this.parseCompoundGuarded(() => this.parseFunctionDefinition());
+      } else if (text === '[') {
+        return this.parseTestCommand();
+      } else if (text.startsWith('((') && PAREN_TEST_RE.test(text)) {
+        // `((word++…))` / `((word--…))` at statement position is a
+        // test_command in the reference (its statement-position arithmetic
+        // grammar does not accept a leading postfix operand), while other
+        // `((…))` forms are arithmetic_expansion command names.
+        return this.parseParenTestCommand();
+      } else if (DECLARATION_COMMAND_SET.has(text)) {
+        return this.parseDeclarationCommand();
+      } else if (UNSET_COMMAND_SET.has(text)) {
+        return this.parseUnsetCommand();
+      } else if (this.isFunctionDefinitionAhead()) {
+        return this.parseCompoundGuarded(() => this.parseFunctionDefinition());
+      }
+    }
+    return this.parseCommand();
+  }
+
+  /**
+   * Run a compound-statement parser under the scope-depth guard: beyond
+   * MAX_PARSE_DEPTH the keyword is parsed as a plain command word instead,
+   * which bounds the recursion chain (if → statements → if → …).
+   */
+  private parseCompoundGuarded(parse: () => Frame): Frame {
+    if (this.scopeDepth >= MAX_PARSE_DEPTH) {
+      this.hasError = true;
+      const fallback = this.parseCommand();
+      if (fallback !== null) return fallback;
+      const token = this.lexer.next();
+      return this.frame('ERROR', token.start, token.end);
+    }
+    this.scopeDepth++;
+    try {
+      return parse();
+    } finally {
+      this.scopeDepth--;
+    }
+  }
+
+  /** negated_command: `!` followed by a command, test command, assignment
+   *  or subshell (the grammar's full choice). */
+  private parseNegatedCommand(): Frame {
+    const bang = this.lexer.next();
+    const kids: Frame[] = [this.anon('!', bang.start, bang.end)];
+    let end = bang.end;
+    const token = this.lexer.peek();
+    if (token.type === 'op' && this.tokenText(token) === '(') {
+      const subshell = this.parseSubshell();
+      kids.push(subshell);
+      end = subshell.end;
+    } else if (token.type === 'word' && this.tokenText(token) === '[') {
+      const test = this.parseTestCommand();
+      kids.push(test);
+      end = test.end;
+    } else if (token.type === 'word' && this.tokenText(token).startsWith('((') && PAREN_TEST_RE.test(this.tokenText(token))) {
+      const test = this.parseParenTestCommand();
+      kids.push(test);
+      end = test.end;
+    } else {
+      const command = this.parseCommand();
+      if (command === null) {
+        this.hasError = true;
+      } else {
+        kids.push(command);
+        end = command.end;
+      }
+    }
+    return this.frame('negated_command', bang.start, end, kids);
+  }
+
+  /** subshell: `(` _statements `)` */
+  private parseSubshell(): Frame {
+    const open = this.lexer.next();
+    if (this.scopeDepth >= MAX_PARSE_DEPTH) {
+      // Absurd nesting: skip token-wise to the matching close paren. Word
+      // tokens hide their internal parens, so token-level counting is exact.
+      this.hasError = true;
+      let depth = 1;
+      let end = open.end;
+      for (;;) {
+        const token = this.lexer.next();
+        end = token.end;
+        if (token.type === 'eof') break;
+        this.completeHeredocs(token.heredocBodies);
+        if (token.type === 'op') {
+          const op = this.tokenText(token);
+          if (op === '(') depth++;
+          else if (op === ')') {
+            depth--;
+            if (depth === 0) break;
+          }
+        }
+      }
+      return this.frame('ERROR', open.start, end, [this.anon('(', open.start, open.end)]);
+    }
+    const kids: Frame[] = [this.anon('(', open.start, open.end)];
+    this.scopeDepth++;
+    const inner = this.parseStatementList({ stopAtParen: true });
+    this.scopeDepth--;
+    if (inner.length === 0) this.hasError = true; // empty subshell (grammar requires statements)
+    kids.push(...inner);
+    let end = this.endOf(inner, open.end);
+    const token = this.lexer.peek();
+    if (token.type === 'op' && this.tokenText(token) === ')') {
+      this.lexer.next();
+      kids.push(this.anon(')', token.start, token.end));
+      end = token.end;
+    } else {
+      this.hasError = true; // unterminated subshell
+    }
+    return this.frame('subshell', open.start, end, kids);
+  }
+
+  // ------------------------------------------------------ compound commands
+
+  /** compound_statement: `{` _terminated_statement `}` */
+  private parseCompoundStatement(): Frame {
+    const open = this.lexer.next();
+    const kids: Frame[] = [
+      this.anon('{', open.start, open.end),
+      ...this.parseStatementList({ stopWords: STOP_CLOSE_BRACE, terminatorRequired: true }),
+    ];
+    let end = this.endOf(kids, open.end);
+    const token = this.lexer.peek();
+    if (token.type === 'word' && this.tokenText(token) === '}') {
+      this.lexer.next();
+      kids.push(this.anon('}', token.start, token.end));
+      end = token.end;
+    } else {
+      this.hasError = true; // unterminated { …
+    }
+    return this.frame('compound_statement', open.start, end, kids);
+  }
+
+  /** if_statement: `if` condition `then` body elif* else? `fi` */
+  private parseIfStatement(): Frame {
+    const ifToken = this.lexer.next();
+    const kids: Frame[] = [
+      this.anon('if', ifToken.start, ifToken.end),
+      ...this.parseStatementList({ stopWords: STOP_THEN, terminatorRequired: true }),
+    ];
+    if (!this.consumeKeyword(kids, 'then')) {
+      this.hasError = true; // missing then
+      return this.frame('if_statement', ifToken.start, this.endOf(kids, ifToken.end), kids);
+    }
+    kids.push(...this.parseStatementList({ stopWords: STOP_IF_BODY, terminatorRequired: true }));
+    while (this.peekKeyword('elif')) {
+      kids.push(this.parseElifClause());
+    }
+    if (this.peekKeyword('else')) {
+      kids.push(this.parseElseClause());
+    }
+    let end = this.endOf(kids, ifToken.end);
+    if (this.consumeKeyword(kids, 'fi')) {
+      end = kids.at(-1)!.end;
+    } else {
+      this.hasError = true; // unterminated if
+    }
+    return this.frame('if_statement', ifToken.start, end, kids);
+  }
+
+  /** elif_clause: `elif` condition `then` body */
+  private parseElifClause(): Frame {
+    const elifToken = this.lexer.next();
+    const kids: Frame[] = [
+      this.anon('elif', elifToken.start, elifToken.end),
+      ...this.parseStatementList({ stopWords: STOP_THEN, terminatorRequired: true }),
+    ];
+    if (!this.consumeKeyword(kids, 'then')) {
+      this.hasError = true;
+      return this.frame('elif_clause', elifToken.start, this.endOf(kids, elifToken.end), kids);
+    }
+    kids.push(...this.parseStatementList({ stopWords: STOP_IF_BODY, terminatorRequired: true }));
+    // Like case_item, an elif/else clause extends over its trailing
+    // terminator (usually a newline) in the reference.
+    const end = Math.max(this.endOf(kids, elifToken.end), this.lastTerminatorEnd);
+    return this.frame('elif_clause', elifToken.start, end, kids);
+  }
+
+  /** else_clause: `else` body */
+  private parseElseClause(): Frame {
+    const elseToken = this.lexer.next();
+    const kids: Frame[] = [
+      this.anon('else', elseToken.start, elseToken.end),
+      ...this.parseStatementList({ stopWords: STOP_FI, terminatorRequired: true }),
+    ];
+    const end = Math.max(this.endOf(kids, elseToken.end), this.lastTerminatorEnd);
+    return this.frame('else_clause', elseToken.start, end, kids);
+  }
+
+  /** while_statement: (`while` | `until`) condition do_group */
+  private parseWhileStatement(): Frame {
+    const kwToken = this.lexer.next();
+    const keyword = this.tokenText(kwToken);
+    const kids: Frame[] = [
+      this.anon(keyword, kwToken.start, kwToken.end),
+      ...this.parseStatementList({ stopWords: STOP_DO, terminatorRequired: true }),
+    ];
+    if (this.peekKeyword('do')) {
+      kids.push(this.parseDoGroup());
+    } else {
+      this.hasError = true; // missing do
+    }
+    return this.frame('while_statement', kwToken.start, this.endOf(kids, kwToken.end), kids);
+  }
+
+  /** do_group: `do` body `done` */
+  private parseDoGroup(): Frame {
+    const doToken = this.lexer.next();
+    const kids: Frame[] = [
+      this.anon('do', doToken.start, doToken.end),
+      ...this.parseStatementList({ stopWords: STOP_DONE, terminatorRequired: true }),
+    ];
+    let end = this.endOf(kids, doToken.end);
+    if (this.consumeKeyword(kids, 'done')) {
+      end = kids.at(-1)!.end;
+    } else {
+      this.hasError = true; // unterminated do
+    }
+    return this.frame('do_group', doToken.start, end, kids);
+  }
+
+  /** Consume the terminator between a for-header and its body: `;`/`&`
+   *  become anonymous children, newlines are dropped, comments are kept. */
+  private consumeForTerminator(kids: Frame[]): void {
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'newline') {
+        this.lexer.next();
+        this.completeHeredocs(token.heredocBodies);
+        continue;
+      }
+      if (token.type === 'comment') {
+        this.lexer.next();
+        kids.push(this.frame('comment', token.start, token.end));
+        continue;
+      }
+      if (token.type === 'op') {
+        const op = this.tokenText(token);
+        if (op === ';' || op === '&') {
+          this.lexer.next();
+          kids.push(this.anon(op, token.start, token.end));
+        }
+      }
+      break;
+    }
+  }
+
+  /** for_statement: (`for` | `select`) variable (`in` values)? terminator
+   *  do_group — or, for `for ((…))`, a c_style_for_statement. */
+  private parseForStatement(): Frame {
+    const kwToken = this.lexer.next();
+    const keyword = this.tokenText(kwToken);
+    const kids: Frame[] = [this.anon(keyword, kwToken.start, kwToken.end)];
+    const next = this.lexer.peek();
+    if (keyword === 'for' && next.type === 'word' && this.tokenText(next).startsWith('((')) {
+      return this.parseCStyleForStatement(kwToken, kids);
+    }
+    const varToken = this.lexer.peek();
+    if (varToken.type === 'word' && /^\w+$/.test(this.tokenText(varToken))) {
+      this.lexer.next();
+      kids.push(this.frame('variable_name', varToken.start, varToken.end));
+    } else {
+      this.hasError = true; // missing loop variable
+    }
+    if (this.peekKeyword('in')) {
+      const inToken = this.lexer.next();
+      const values: Frame[] = [];
+      while (this.lexer.peek().type === 'word') {
+        values.push(this.parseWordArgument());
+      }
+      if (values.length === 0) {
+        // `for x in; do` — the reference wraps the bare `in` in ERROR.
+        this.hasError = true;
+        kids.push(this.frame('ERROR', inToken.start, inToken.end, [this.anon('in', inToken.start, inToken.end)]));
+      } else {
+        kids.push(this.anon('in', inToken.start, inToken.end));
+        kids.push(...values);
+      }
+    }
+    this.consumeForTerminator(kids);
+    if (this.peekKeyword('do')) {
+      kids.push(this.parseDoGroup());
+    } else {
+      this.hasError = true; // missing do
+    }
+    return this.frame('for_statement', kwToken.start, this.endOf(kids, kwToken.end), kids);
+  }
+
+  /** c_style_for_statement: `for` `((` init `;` cond `;` update `))`
+   *  terminator (do_group | compound_statement). The whole ((…)) header
+   *  arrived as one word token (see the lexer). */
+  private parseCStyleForStatement(kwToken: Token, kids: Frame[]): Frame {
+    const header = this.lexer.next();
+    const closed =
+      header.end - header.start >= 4 && this.source[header.end - 2] === ')' && this.source[header.end - 1] === ')';
+    if (!closed) this.hasError = true;
+    kids.push(this.anon('((', header.start, header.start + 2));
+    const innerEnd = closed ? header.end - 2 : header.end;
+    kids.push(...this.parseCForBody(header.start + 2, innerEnd));
+    if (closed) {
+      kids.push(this.anon('))', header.end - 2, header.end));
+    }
+    this.consumeForTerminator(kids);
+    if (this.peekKeyword('do')) {
+      kids.push(this.parseDoGroup());
+    } else if (this.peekKeyword('{')) {
+      kids.push(this.parseCompoundStatement());
+    } else {
+      this.hasError = true; // missing body
+    }
+    return this.frame('c_style_for_statement', kwToken.start, this.endOf(kids, kwToken.end), kids);
+  }
+
+  /** The init/condition/update parts of a c-style for header: three
+   *  comma-separated c-expression lists divided by `;`. */
+  private parseCForBody(start: number, end: number): Frame[] {
+    const st = this.newExprState(start, end, 'c');
+    const kids: Frame[] = [];
+    for (let part = 0; part < 3; part++) {
+      for (;;) {
+        const expression = this.parseExpression(st, 0);
+        if (expression !== null) kids.push(expression);
+        const token = this.exprPeek(st);
+        if (token.kind === 'op' && token.text === ',') {
+          this.exprNext(st);
+          kids.push(this.anon(',', token.start, token.end));
+          continue;
+        }
+        break;
+      }
+      if (part < 2) {
+        const token = this.exprPeek(st);
+        if (token.kind === 'op' && token.text === ';') {
+          this.exprNext(st);
+          kids.push(this.anon(';', token.start, token.end));
+        } else {
+          this.hasError = true; // missing ; in for-header
+          break;
+        }
+      }
+    }
+    const leftover = this.exprLeftover(st);
+    if (leftover !== null) kids.push(leftover);
+    return kids;
+  }
+
+  /** case_statement: `case` value `in` case_item* `esac` */
+  private parseCaseStatement(): Frame {
+    const caseToken = this.lexer.next();
+    const kids: Frame[] = [this.anon('case', caseToken.start, caseToken.end)];
+    if (this.lexer.peek().type === 'word') {
+      kids.push(this.parseWordArgument());
+    } else {
+      this.hasError = true; // missing value
+    }
+    this.skipCaseTerminators(kids);
+    if (!this.consumeKeyword(kids, 'in')) {
+      this.hasError = true;
+      return this.frame('case_statement', caseToken.start, this.endOf(kids, caseToken.end), kids);
+    }
+    this.skipCaseTerminators(kids);
+    for (;;) {
+      // Between items: newlines are dropped, comments kept as children.
+      for (;;) {
+        const token = this.lexer.peek();
+        if (token.type === 'newline') {
+          this.lexer.next();
+          this.completeHeredocs(token.heredocBodies);
+          continue;
+        }
+        if (token.type === 'comment') {
+          this.lexer.next();
+          kids.push(this.frame('comment', token.start, token.end));
+          continue;
+        }
+        break;
+      }
+      const token = this.lexer.peek();
+      if (token.type === 'eof') {
+        this.hasError = true; // missing esac
+        break;
+      }
+      if (token.type === 'word' && this.tokenText(token) === 'esac') break;
+      const before = this.lexer.pos;
+      kids.push(this.parseCaseItem());
+      if (this.lexer.pos === before) {
+        // Defensive: a case_item that consumed nothing would loop forever.
+        this.hasError = true;
+        this.lexer.next();
+      }
+      if (this.peekKeyword('esac')) {
+        // A fallthrough terminator on the LAST item is a grammar error in
+        // the reference (last_case_item only allows `;;`).
+        const last = kids.at(-1)!.children.at(-1);
+        if (last !== undefined && !last.isNamed && (last.type === ';&' || last.type === ';;&')) {
+          this.hasError = true;
+        }
+      }
+    }
+    let end = this.endOf(kids, caseToken.end);
+    if (this.consumeKeyword(kids, 'esac')) {
+      end = kids.at(-1)!.end;
+    }
+    return this.frame('case_statement', caseToken.start, end, kids);
+  }
+
+  /** Terminators around `in` / between case items: newlines dropped, `;`/`&`
+   *  kept as anonymous children, comments kept. */
+  private skipCaseTerminators(kids: Frame[]): void {
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'newline') {
+        this.lexer.next();
+        this.completeHeredocs(token.heredocBodies);
+        continue;
+      }
+      if (token.type === 'comment') {
+        this.lexer.next();
+        kids.push(this.frame('comment', token.start, token.end));
+        continue;
+      }
+      if (token.type === 'op') {
+        const op = this.tokenText(token);
+        if (op === ';' || op === '&') {
+          this.lexer.next();
+          kids.push(this.anon(op, token.start, token.end));
+          continue;
+        }
+      }
+      break;
+    }
+  }
+
+  /**
+   * case_item: `(`? pattern (`|` pattern)* `)` statements? terminator?
+   * Patterns are literals, except bare words that contain a glob character
+   * (* ? [) or that are followed by `|` — those are extglob_pattern nodes,
+   * matching the reference scanner.
+   */
+  /** case_item: `( pattern (| pattern)* ) statements? terminator?`.
+   *  Patterns are scanned character-wise (they may hold extglob groups with
+   *  parens the token stream cannot represent) and classified following the
+   *  reference scanner: a bare literal is a word (or concatenation pieces),
+   *  glob material is one extglob_pattern, and a pattern mixing glob text
+   *  with a quote/expansion splits into extglob_pattern pieces around the
+   *  construct (the grammar's _extglob_blob). */
+  private parseCaseItem(): Frame {
+    const kids: Frame[] = [];
+    let token = this.lexer.peek();
+    if (token.type === 'op' && this.tokenText(token) === '(') {
+      this.lexer.next();
+      kids.push(this.anon('(', token.start, token.end));
+    }
+    // Patterns.
+    let firstAlternative = true;
+    for (;;) {
+      token = this.lexer.peek();
+      if (token.type !== 'word') {
+        this.hasError = true; // missing pattern
+        break;
+      }
+      const alt = this.scanCasePatternEnd(token.start);
+      kids.push(...this.parseCasePattern(token.start, alt.end, firstAlternative));
+      firstAlternative = false;
+      this.lexer.reposition(alt.end);
+      token = this.lexer.peek();
+      if (token.type === 'op' && this.tokenText(token) === '|') {
+        this.lexer.next();
+        kids.push(this.anon('|', token.start, token.end));
+        continue;
+      }
+      break;
+    }
+    token = this.lexer.peek();
+    if (token.type === 'op' && this.tokenText(token) === ')') {
+      this.lexer.next();
+      kids.push(this.anon(')', token.start, token.end));
+    } else {
+      this.hasError = true; // missing )
+    }
+    kids.push(...this.parseCaseItemStatements());
+    token = this.lexer.peek();
+    if (token.type === 'op') {
+      const op = this.tokenText(token);
+      if (op === ';;' || op === ';&' || op === ';;&') {
+        this.lexer.next();
+        kids.push(this.anon(op, token.start, token.end));
+      }
+    }
+    const start = kids.length > 0 ? kids[0]!.start : token.start;
+    // An item closed by `esac` (no `;;`) extends over its trailing
+    // terminator (usually a newline) in the reference.
+    const end = Math.max(this.endOf(kids, token.start), this.lastTerminatorEnd);
+    return this.frame('case_item', start, end, kids);
+  }
+
+  /** End of one case pattern alternative: the position of the top-level `)`
+   *  or `|` (extglob group parens nest). Quote/escape/substitution aware. */
+  private scanCasePatternEnd(i: number): { end: number } {
+    const end = this.lexer.rangeEnd;
+    let depth = 0;
+    let j = i;
+    let sinceTick = 0;
+    while (j < end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[j]!;
+      if (ch === '\n' || ch === ';') break; // defensive: malformed item
+      if (ch === '\\') {
+        j += 2;
+        continue;
+      }
+      if (ch === '"') {
+        j = skipDoubleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === "'") {
+        j = skipSingleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === '`') {
+        j = skipBacktick(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === '$') {
+        j = this.skipDollarConstruct(j, end);
+        continue;
+      }
+      if (ch === '(') {
+        depth++;
+        j++;
+        continue;
+      }
+      if (ch === ')') {
+        if (depth === 0) return { end: j };
+        depth--;
+        j++;
+        continue;
+      }
+      if (ch === '|' && depth === 0) return { end: j };
+      if ((ch === ' ' || ch === '\t' || ch === '\r') && depth === 0) return { end: j };
+      j++;
+    }
+    return { end: j };
+  }
+
+  /** Build the frames for one case pattern alternative. The extglob_blob
+   *  piece split (glob text around a quote/expansion) only applies to the
+   *  FIRST alternative — after `|` the reference keeps such patterns as
+   *  plain concatenations. */
+  private parseCasePattern(start: number, end: number, allowBlob: boolean): Frame[] {
+    const raw = this.text(start, end);
+    // Extglob group present: accepted groups make the whole alternative one
+    // extglob_pattern; rejected ones degrade with hasError (the reference
+    // reparses the group as a new item — a recovery shape we do not mirror).
+    // `.(` is not a sigil, but the reference rejects it the same way.
+    const group = /[?*+@!.]\(/.exec(raw);
+    if (group !== null) {
+      if (this.extglobGroupAccepted(start, end)) return [this.frame('extglob_pattern', start, end)];
+      this.hasError = true;
+      return [this.frame('ERROR', start, end)];
+    }
+    if (!/["'$`]/.test(raw)) {
+      // Bare pattern.
+      if (this.isBareGlobPattern(start, end)) return [this.frame('extglob_pattern', start, end)];
+      return [this.parseLiteral(start, end)];
+    }
+    // Glob text mixed with a quote/expansion: extglob_pattern pieces around
+    // a single construct when the leading run qualifies as a glob.
+    if (allowBlob) {
+      const pieces = this.parseExtglobBlob(start, end);
+      if (pieces !== null) return pieces;
+    }
+    return [this.parseLiteral(start, end)];
+  }
+
+  /**
+   * Whether the extglob scanner accepts a group-containing pattern
+   * (scanner.c's first/second character rules): the first character must be
+   * a letter or one of `?*+@!-)\.[` (a leading escape consumes the escaped
+   * space/quote with it, plus one more character), and the character at the
+   * scanner's check position must be alphanumeric or one of `([?/\_*`.
+   */
+  private extglobGroupAccepted(start: number, end: number): boolean {
+    const isAlpha = (ch: string): boolean => /[A-Za-z]/.test(ch);
+    const isAlnum = (ch: string): boolean => /[A-Za-z0-9]/.test(ch);
+    const c0 = this.source[start]!;
+    if (!isAlpha(c0) && !'?*+@!-)\\.['.includes(c0)) return false;
+    if (c0 === '[') return true; // the scanner skips the second-char check
+    let check = start + 1;
+    if (c0 === '\\') {
+      const after = this.source[start + 1];
+      if (after === undefined || !(after === ' ' || after === '\t' || after === '\r' || after === '"')) return false;
+      check = start + 3; // escape pair + one unconditionally consumed char
+    }
+    if (check >= end) return false;
+    const ch = this.source[check]!;
+    return isAlnum(ch) || '([?/\\_*'.includes(ch);
+  }
+
+  /**
+   * Whether a bare case pattern (no quotes, substitutions or groups) is an
+   * extglob_pattern in the reference, derived from scanner.c's
+   * extglob_pattern scanner (the same source as isTestExtglob for `[[ ]]`
+   * right sides) and verified against the wasm build on an ~90-pattern
+   * matrix (foo_bar/word1/abc1d/v2.0/a1/a_ → glob; abc/a.b/1a/é/a-b → word):
+   *  1. the first character must be a (Unicode) letter or one of
+   *     `?*+@!-)\.[`, else the pattern is a literal;
+   *  2. a leading single dash makes the pattern a word when the rest is
+   *     plain alphanumerics (the scanner's special-variable branch eats
+   *     the `-` before the glob scan: `-o`, `-xy`, `-abc` → word, `-1` →
+   *     number); `--anything` is a glob, and so is a single dash with an
+   *     inner dash run (`-x-`, `-x-y`, `-x-y-z` → glob) unless the inner
+   *     dashes are doubled (`-x--` → word);
+   *  3. single-character patterns use the scanner's early exits: a letter
+   *     before `)` is a word, anything else (and anything before `|` or
+   *     whitespace) is a glob;
+   *  4. a dash in second position is consumed together with its
+   *     alphanumeric run BEFORE the validity check; the pattern is a word
+   *     when the run ends the pattern directly before `)` (`a-b)`, `z-)`)
+   *     or is followed by `.` / a character that is not valid pattern
+   *     material (`a-b-c`). Before `|` or whitespace the glob survives
+   *     (`a-b|`, `a-b `);
+   *  5. the second character must be alphanumeric or one of `([?/\_*`
+   *     (`*.txt`, `a.b`, `q+w`, `a:b` → word; skipped for a leading `[`);
+   *  6. ending quirk: a pattern directly followed by `|` is always a glob;
+   *  7. otherwise the pattern is a glob iff some character counts as glob
+   *     material: the first character when it is not a letter, or any later
+   *     character that is neither a letter nor `.` (digits and `_` count —
+   *     `foo_bar`, `abc1d` are globs; `abc`, `foo.txt` are words).
+   */
+  private isBareGlobPattern(start: number, end: number): boolean {
+    const raw = this.text(start, end);
+    if (raw.includes('\\')) return false;
+    if (NUMBER_RE.test(raw)) return false; // 12, -1, 0x1F → number
+    const isAlpha = (ch: string): boolean => /^\p{L}$/u.test(ch);
+    const isAlnum = (ch: string): boolean => /^[\p{L}\p{N}]$/u.test(ch);
+    const next = this.source[end] ?? '';
+    const c0 = raw[0]!;
+    // (1) first character
+    if (!isAlpha(c0) && !'?*+@!-)\\.['.includes(c0)) return false;
+    // (2) leading dash: the scanner's special-variable branch eats a
+    // leading `-` before the glob scan. A single dash followed by plain
+    // alphanumerics is a word/number (`-o`, `-xy`, `-x1` → word, `-1` →
+    // number); `--anything` is a glob; so is a single dash with exactly
+    // one INNER dash run (`-x-`, `-x-y`, `-x-y-z` → glob) — but a doubled
+    // inner dash falls back to a word (`-x--`).
+    if (c0 === '-') {
+      if (raw.length > 1 && raw[1] === '-') return true;
+      let innerDash = -1;
+      for (let i = 1; i < raw.length; i++) {
+        if (raw[i] === '-') {
+          innerDash = i;
+          break;
+        }
+      }
+      if (innerDash === -1) return false;
+      return raw[innerDash + 1] !== '-';
+    }
+    // (3) single-character patterns: the scanner's early exits — but only
+    // at a real pattern END (`)`, `|` or whitespace; when a quote or
+    // substitution follows, the run continues as a blob piece and the
+    // glob-material check below decides).
+    if (raw.length === 1) {
+      if (next === ')') return !isAlpha(c0);
+      if (next === '|' || next === ' ' || next === '\t' || next === '\r' || next === '\n' || next === '') return true;
+    }
+    // (4) a dash in second position is consumed together with its
+    // alphanumeric run BEFORE the validity check; the pattern is a word
+    // when the run ends the pattern directly before `)` (a-b), z-)) or is
+    // followed by `.` / a character that is not valid pattern material
+    // (a-b-c). Before `|` or whitespace the glob survives (a-b|, a-b␣).
+    if (raw.length > 1) {
+      let p = 1;
+      if (raw[1] === '-') {
+        let i = 2;
+        while (i < raw.length && isAlnum(raw[i]!)) i++;
+        if (i >= raw.length) {
+          if (next === ')') return false;
+          return true;
+        }
+        const after = raw[i]!;
+        if (after === '.' || after === '\\') return false;
+        if (!isAlnum(after) && !'([?/\\_*'.includes(after)) return false;
+        p = i;
+      }
+      // (5) second-character validity (skipped for a leading `[`)
+      if (c0 !== '[' && p === 1 && !isAlnum(raw[1]!) && !'([?/\\_*'.includes(raw[1]!)) return false;
+    }
+    // (6) ending quirk: directly before `|` the scanner accepts unconditionally
+    if (next === '|') return true;
+    // (7) glob material?
+    if (!isAlpha(c0)) return true;
+    for (let i = 1; i < raw.length; i++) {
+      const ch = raw[i]!;
+      if (ch !== '.' && !isAlpha(ch)) return true;
+    }
+    return false;
+  }
+
+  /**
+   * The _extglob_blob shape: extglob_pattern pieces around ONE construct
+   * (string / raw_string / expansion / command_substitution). Returns null
+   * when the leading bare run does not qualify as a glob (the pattern is a
+   * plain concatenation in the reference) or the shape does not fit.
+   */
+  private parseExtglobBlob(start: number, end: number, leadValidator?: (s: number, e: number) => boolean): Frame[] | null {
+    // Find the first construct.
+    let i = start;
+    while (i < end) {
+      const ch = this.source[i]!;
+      if (ch === '\\') {
+        i += 2;
+        continue;
+      }
+      if (ch === '"' || ch === "'" || ch === '`' || ch === '$') break;
+      i++;
+    }
+    if (i >= end || i === start) return null;
+    // The leading bare run must qualify as a glob. Runs with escapes count
+    // when they contain an explicit glob character (`*\ ${x}` → extglob
+    // "*\ "); unescaped runs go through the bare-pattern rule. A custom
+    // validator applies for group-containing runs (test patterns).
+    const leadText = this.text(start, i);
+    const leadIsGlob =
+      leadValidator !== undefined
+        ? leadValidator(start, i)
+        : leadText.includes('\\')
+          ? /[*?[]/.test(leadText)
+          : this.isBareGlobPattern(start, i);
+    if (!leadIsGlob) return null;
+    const lead = this.frame('extglob_pattern', start, i);
+    // The construct.
+    let construct: Frame;
+    let next: number;
+    const ch = this.source[i]!;
+    if (ch === '"') {
+      [construct, next] = this.parseString(i, end);
+    } else if (ch === "'") {
+      next = skipSingleQuoted(this.source, this.budget, i, end);
+      construct = this.frame('raw_string', i, next);
+    } else if (ch === '`') {
+      [construct, next] = this.parseBacktickSubstitution(i, end);
+    } else {
+      const dollar = this.parseDollar(i, end);
+      if (dollar === null) return null;
+      [construct, next] = dollar;
+    }
+    if (next >= end) return [lead, construct];
+    const trail = this.frame('extglob_pattern', next, end);
+    return [lead, construct, trail];
+  }
+
+  /** case_item statements with the esac-argument guard active (see
+   *  caseItemDepth). */
+  private parseCaseItemStatements(): Frame[] {
+    this.caseItemDepth++;
+    try {
+      return this.parseStatementList({ stopWords: STOP_ESAC, stopOps: CASE_TERMINATION_OPS });
+    } finally {
+      this.caseItemDepth--;
+    }
+  }
+
+  /** function_definition: (`function`)? name (`()`)? body redirect?
+   *  The body is a compound_statement, subshell, test_command or
+   *  if_statement (the grammar's full choice). */
+  private parseFunctionDefinition(): Frame {
+    const kids: Frame[] = [];
+    const first = this.lexer.next();
+    if (this.tokenText(first) === 'function') {
+      kids.push(this.anon('function', first.start, first.end));
+      const nameToken = this.lexer.peek();
+      if (nameToken.type === 'word') {
+        this.lexer.next();
+        kids.push(this.frame('word', nameToken.start, nameToken.end));
+      } else {
+        this.hasError = true; // missing function name
+      }
+      this.consumeParenPair(kids);
+    } else {
+      kids.push(this.frame('word', first.start, first.end));
+      this.consumeParenPair(kids);
+    }
+    // Blank lines/comments may sit between the header and the body.
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'newline') {
+        this.lexer.next();
+        this.completeHeredocs(token.heredocBodies);
+        continue;
+      }
+      if (token.type === 'comment') {
+        this.lexer.next();
+        kids.push(this.frame('comment', token.start, token.end));
+        continue;
+      }
+      break;
+    }
+    const token = this.lexer.peek();
+    if (token.type === 'word' && this.tokenText(token) === '{') {
+      kids.push(this.parseCompoundStatement());
+    } else if (token.type === 'op' && this.tokenText(token) === '(') {
+      kids.push(this.parseSubshell());
+    } else if (token.type === 'word' && this.tokenText(token) === '[') {
+      kids.push(this.parseTestCommand());
+    } else if (token.type === 'word' && this.tokenText(token) === 'if') {
+      kids.push(this.parseIfStatement());
+    } else {
+      this.hasError = true; // missing function body
+    }
+    // Optional single redirect (file_redirect or herestring).
+    const redirect = this.lexer.peek();
+    if (redirect.type === 'io_number') {
+      kids.push(this.parseRedirect());
+    } else if (redirect.type === 'op') {
+      const op = this.tokenText(redirect);
+      if (isFileRedirectOp(op) || op === '<<<') {
+        kids.push(this.parseRedirect());
+      }
+    }
+    return this.frame('function_definition', kids[0]!.start, this.endOf(kids, first.end), kids);
+  }
+
+  /** Consume a `(` `)` pair (blanks allowed between) as anonymous children
+   *  when present. */
+  private consumeParenPair(kids: Frame[]): void {
+    const open = this.lexer.peek();
+    if (open.type !== 'op' || this.tokenText(open) !== '(') return;
+    const close = this.lexer.peekAt(1);
+    if (close.type !== 'op' || this.tokenText(close) !== ')') return;
+    this.lexer.next();
+    this.lexer.next();
+    kids.push(this.anon('(', open.start, open.end));
+    kids.push(this.anon(')', close.start, close.end));
+  }
+
+  /** Lookahead for the `name() body` function form: a non-reserved word
+   *  followed by `(` `)` and a valid body start (`{`, `(`, `[`, `if`),
+   *  possibly across newlines. Without a valid body start the tokens belong
+   *  to a plain command (`foo () ls` is command + ERROR in the reference). */
+  private isFunctionDefinitionAhead(): boolean {
+    const name = this.lexer.peek();
+    const text = this.tokenText(name);
+    if (!FUNCTION_NAME_RE.test(text) || RESERVED_WORD_SET.has(text)) return false;
+    const open = this.lexer.peekAt(1);
+    if (open.type !== 'op' || this.tokenText(open) !== '(') return false;
+    const close = this.lexer.peekAt(2);
+    if (close.type !== 'op' || this.tokenText(close) !== ')') return false;
+    for (let n = 3; ; n++) {
+      const token = this.lexer.peekAt(n);
+      if (token.type === 'newline' || token.type === 'comment') continue;
+      if (token.type === 'op' && this.tokenText(token) === '(') return true;
+      if (token.type === 'word') {
+        const word = this.tokenText(token);
+        return word === '{' || word === '[' || word === 'if';
+      }
+      return false;
+    }
+  }
+
+  /** declaration_command: (`declare` | `typeset` | `export` | `readonly` |
+   *  `local`) (variable_assignment | literal | variable_name)* */
+  private parseDeclarationCommand(): Frame {
+    const kwToken = this.lexer.next();
+    const kids: Frame[] = [this.anon(this.tokenText(kwToken), kwToken.start, kwToken.end)];
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type !== 'word') break;
+      const text = this.tokenText(token);
+      if (ASSIGNMENT_RE.test(text)) {
+        this.lexer.next();
+        kids.push(this.parseVariableAssignment(token));
+      } else if (this.isSubscriptAssignmentAhead(token)) {
+        kids.push(this.parseSubscriptAssignment());
+      } else if (IDENTIFIER_RE.test(text)) {
+        this.lexer.next();
+        kids.push(this.frame('variable_name', token.start, token.end));
+      } else {
+        kids.push(this.parseWordArgument());
+      }
+    }
+    return this.frame('declaration_command', kwToken.start, this.endOf(kids, kwToken.end), kids);
+  }
+
+  /** unset_command: (`unset` | `unsetenv`) (literal | variable_name)* */
+  private parseUnsetCommand(): Frame {
+    const kwToken = this.lexer.next();
+    const kids: Frame[] = [this.anon(this.tokenText(kwToken), kwToken.start, kwToken.end)];
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type !== 'word') break;
+      const text = this.tokenText(token);
+      if (IDENTIFIER_RE.test(text)) {
+        this.lexer.next();
+        kids.push(this.frame('variable_name', token.start, token.end));
+      } else {
+        kids.push(this.parseWordArgument());
+      }
+    }
+    return this.frame('unset_command', kwToken.start, this.endOf(kids, kwToken.end), kids);
+  }
+
+  // --------------------------------------------------------------- commands
+
+  /**
+   * command: leading variable_assignment / redirect prefix, command_name,
+   * then arguments, inline herestring redirects and a trailing subshell
+   * (`foo (ls)` keeps the subshell inside the command, as the reference
+   * does). Returns null when no command name is present and nothing was
+   * consumed; a nameless prefix is assembled into variable_assignment(s) /
+   * redirected_statement here.
+   */
+  private parseCommand(): Frame | null {
+    const prefix: Frame[] = [];
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'word' && ASSIGNMENT_RE.test(this.tokenText(token))) {
+        this.lexer.next();
+        prefix.push(this.parseVariableAssignment(token));
+        continue;
+      }
+      if (token.type === 'word' && this.isSubscriptAssignmentAhead(token)) {
+        prefix.push(this.parseSubscriptAssignment());
+        continue;
+      }
+      // Prefix redirects take exactly one destination so a following word
+      // still becomes the command_name (`> a cmd x`). Heredoc operators are
+      // left for the trailing-redirect path.
+      if (token.type === 'io_number') {
+        prefix.push(this.parseRedirect(1));
+        continue;
+      }
+      if (token.type === 'op') {
+        const op = this.tokenText(token);
+        if (isFileRedirectOp(op) || op === '<<<') {
+          prefix.push(this.parseRedirect(1));
+          continue;
+        }
+      }
+      break;
+    }
+    if (this.lexer.peek().type !== 'word') {
+      if (prefix.length === 0) return null;
+      return this.assembleNamelessPrefix(prefix);
+    }
+    const start = prefix.length > 0 ? prefix[0]!.start : this.lexer.peek().start;
+    // The name merges adjacent word tokens (the lexer splits { } [ ] out as
+    // single-character tokens): `{1..3}` or `cmd{x}` is one name.
+    const [nameStart, nameEnd] = this.consumeWordRun();
+    const name = this.frame('command_name', nameStart, nameEnd, [this.parseLiteral(nameStart, nameEnd)]);
+    const command = this.frame('command', start, nameEnd, [...prefix, name]);
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'word') {
+        // Inside a case_item, `esac` closes the item rather than becoming
+        // an argument (see caseItemDepth).
+        if (this.caseItemDepth > 0 && this.tokenText(token) === 'esac') break;
+        for (const argument of this.parseCommandArgument()) {
+          this.addKid(command, argument);
+          command.end = argument.end;
+        }
+        continue;
+      }
+      if (token.type === 'op' && this.tokenText(token) === '<<<') {
+        const herestring = this.parseHerestringRedirect(null);
+        this.addKid(command, herestring);
+        command.end = herestring.end;
+        continue;
+      }
+      if (token.type === 'op' && this.tokenText(token) === '(') {
+        const subshell = this.parseSubshell();
+        this.addKid(command, subshell);
+        command.end = subshell.end;
+        continue;
+      }
+      break;
+    }
+    return command;
+  }
+
+  /** Consume a run of adjacent word tokens; returns [start, end). */
+  private consumeWordRun(): [number, number] {
+    const first = this.lexer.next();
+    let end = first.end;
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type !== 'word' || token.start !== end) break;
+      this.lexer.next();
+      end = token.end;
+    }
+    return [first.start, end];
+  }
+
+  /** One command argument: normally one literal, but `$"…"` (translated
+   *  string) is TWO arguments — an anonymous `$` plus the string — because
+   *  a bare dollar cannot start a concatenation in the reference grammar
+   *  (tree-sitter-bash 0.25.0 never produces translated_string). */
+  private parseCommandArgument(): Frame[] {
+    const [start, end] = this.consumeWordRun();
+    if (this.source[start] === '$' && this.source[start + 1] === '"' && start + 1 < end) {
+      return [this.anon('$', start, start + 1), this.parseLiteral(start + 1, end)];
+    }
+    return [this.parseLiteral(start, end)];
+  }
+
+  /** A prefix of assignments/redirects with no command name:
+   *  `FOO=bar`, `A=1 B=2`, `> out`, `> out FOO=bar`. */
+  private assembleNamelessPrefix(prefix: Frame[]): Frame {
+    const assignments = prefix.filter((f) => f.type === 'variable_assignment');
+    if (assignments.length === prefix.length) {
+      if (prefix.length === 1) return prefix[0]!;
+      return this.frame('variable_assignments', prefix[0]!.start, prefix.at(-1)!.end, prefix);
+    }
+    return this.frame('redirected_statement', prefix[0]!.start, prefix.at(-1)!.end, prefix);
+  }
+
+  /** True when a subscripted assignment (`arr[0]=x`, `a[i+1]+=2`) starts at
+   *  the peeked word token. */
+  private isSubscriptAssignmentAhead(token: Token): boolean {
+    if (!IDENTIFIER_RE.test(this.tokenText(token))) return false;
+    return SUBSCRIPT_ASSIGNMENT_RE.test(this.source.slice(token.start));
+  }
+
+  /** variable_assignment with a subscript name: `arr[0]=x`, `a[i+1]+=2`.
+   *  The tokens are the name, `[`, index pieces, `]` and a word starting
+   *  with the operator; ranges come from a regex over the source. */
+  private parseSubscriptAssignment(): Frame {
+    const start = this.lexer.peek().start;
+    const match = SUBSCRIPT_ASSIGNMENT_RE.exec(this.source.slice(start))!;
+    const nameEnd = start + match[1]!.length;
+    const indexStart = nameEnd + 1;
+    const indexEnd = indexStart + match[2]!.length;
+    const opStart = indexEnd + 1;
+    const opEnd = opStart + match[3]!.length;
+    let valueEnd = opEnd;
+    // Consume tokens covering [start, opEnd) — the last one may extend past
+    // the operator into the value (`=x"y"` is one word token).
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.start >= opEnd) break;
+      this.lexer.next();
+      valueEnd = Math.max(valueEnd, token.end);
+    }
+    // Merge following adjacent word tokens into the value.
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type !== 'word' || token.start !== valueEnd) break;
+      this.lexer.next();
+      valueEnd = token.end;
+    }
+    const subscriptKids: Frame[] = [this.frame('variable_name', start, nameEnd), this.anon('[', nameEnd, indexStart)];
+    if (indexEnd > indexStart) {
+      subscriptKids.push(this.parseLiteral(indexStart, indexEnd));
+    } else {
+      this.hasError = true; // empty subscript index
+    }
+    subscriptKids.push(this.anon(']', indexEnd, opStart));
+    const kids: Frame[] = [
+      this.frame('subscript', start, opStart, subscriptKids),
+      this.anon(match[3]!, opStart, opEnd),
+    ];
+    if (valueEnd > opEnd) {
+      kids.push(this.parseLiteral(opEnd, valueEnd));
+    }
+    return this.frame('variable_assignment', start, valueEnd, kids);
+  }
+
+  /** variable_assignment: NAME ( = | += ) value — the word token is split
+   *  into variable_name, the operator, and a value parsed from the rest of
+   *  the token plus any ADJACENT word tokens (the lexer splits { } [ ] out
+   *  as single-character tokens, so `x={1..5}` and `x=a{b}` are single
+   *  assignments). An empty value followed immediately by `(` is an array
+   *  literal (`arr=(a b)`). */
+  private parseVariableAssignment(token: Token): Frame {
+    const text = this.tokenText(token);
+    const match = ASSIGNMENT_SPLIT_RE.exec(text)!;
+    const nameEnd = token.start + match[1]!.length;
+    const opEnd = nameEnd + match[2]!.length;
+    const kids: Frame[] = [
+      this.frame('variable_name', token.start, nameEnd),
+      this.anon(match[2]!, nameEnd, opEnd),
+    ];
+    let valueEnd = token.end;
+    for (;;) {
+      const next = this.lexer.peek();
+      if (next.type !== 'word' || next.start !== valueEnd) break;
+      this.lexer.next();
+      valueEnd = next.end;
+    }
+    if (valueEnd > opEnd) {
+      kids.push(this.parseLiteral(opEnd, valueEnd));
+      return this.frame('variable_assignment', token.start, valueEnd, kids);
+    }
+    const next = this.lexer.peek();
+    if (next.type === 'op' && this.tokenText(next) === '(' && next.start === token.end) {
+      const array = this.parseArray();
+      kids.push(array);
+      return this.frame('variable_assignment', token.start, array.end, kids);
+    }
+    return this.frame('variable_assignment', token.start, token.end, kids);
+  }
+
+  /** array: `(` literal* `)` — the elements of an array assignment. */
+  private parseArray(): Frame {
+    const open = this.lexer.next();
+    const kids: Frame[] = [this.anon('(', open.start, open.end)];
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'word') {
+        kids.push(this.parseWordArgument());
+        continue;
+      }
+      if (token.type === 'newline') {
+        this.lexer.next();
+        this.completeHeredocs(token.heredocBodies);
+        continue;
+      }
+      if (token.type === 'comment') {
+        this.lexer.next();
+        kids.push(this.frame('comment', token.start, token.end));
+        continue;
+      }
+      if (token.type === 'op' && this.tokenText(token) === ')') {
+        this.lexer.next();
+        kids.push(this.anon(')', token.start, token.end));
+        return this.frame('array', open.start, token.end, kids);
+      }
+      this.hasError = true; // unterminated array
+      return this.frame('array', open.start, this.endOf(kids, open.end), kids);
+    }
+  }
+
+  // --------------------------------------------------------------- redirect
+
+  /** Parse one redirect with an optional io_number descriptor prefix already
+   *  peeked. Dispatches to file_redirect / herestring / heredoc handling.
+   *  `maxDestinations` limits greedy destination consumption (command-prefix
+   *  redirects take exactly one). */
+  private parseRedirect(maxDestinations = Number.POSITIVE_INFINITY): Frame {
+    let descriptor: Frame | null = null;
+    let token = this.lexer.peek();
+    if (token.type === 'io_number') {
+      this.lexer.next();
+      descriptor = this.frame('file_descriptor', token.start, token.end);
+      token = this.lexer.peek();
+    }
+    if (token.type !== 'op') {
+      // io_number not followed by an operator — recover.
+      this.hasError = true;
+      const stray = descriptor ?? this.frame('ERROR', token.start, token.end);
+      return this.frame('ERROR', stray.start, stray.end, descriptor === null ? [] : [stray]);
+    }
+    const op = this.tokenText(token);
+    if (op === '<<' || op === '<<-') {
+      return this.noHeredoc ? this.parseBrokenHeredoc(descriptor) : this.parseHeredocRedirect(descriptor);
+    }
+    if (op === '<<<') return this.parseHerestringRedirect(descriptor);
+    if (isFileRedirectOp(op)) return this.parseFileRedirect(descriptor, maxDestinations);
+    this.hasError = true;
+    const stray = descriptor ?? this.anon(op, token.start, token.end);
+    return this.frame('ERROR', stray.start, stray.end, [stray]);
+  }
+
+  /** file_redirect: [fd] op destination* — `>&-`/`<&-` take no destination.
+   *  In command-prefix position only one destination is taken (`> a cmd x`
+   *  makes `cmd` the command_name); after the command name the redirect is
+   *  greedy, matching tree-sitter-bash (`cmd > out arg` puts both words in
+   *  the redirect). */
+  private parseFileRedirect(descriptor: Frame | null, maxDestinations = Number.POSITIVE_INFINITY): Frame {
+    const opToken = this.lexer.next();
+    const op = this.tokenText(opToken);
+    const kids: Frame[] = [];
+    if (descriptor !== null) kids.push(descriptor);
+    kids.push(this.anon(op, opToken.start, opToken.end));
+    let end = opToken.end;
+    if (op === '>&-' || op === '<&-') {
+      // Close-fd operators take an OPTIONAL single destination in the
+      // reference (`exec {a} <&- {b}>&-` puts `{b}` in the `<&-` redirect).
+      if (this.lexer.peek().type === 'word') {
+        const destination = this.parseWordArgument();
+        kids.push(destination);
+        end = destination.end;
+      }
+    } else {
+      let destinations = 0;
+      while (destinations < maxDestinations && this.lexer.peek().type === 'word') {
+        const destination = this.parseWordArgument();
+        kids.push(destination);
+        end = destination.end;
+        destinations++;
+      }
+      if (destinations === 0) this.hasError = true; // missing destination
+    }
+    return this.frame('file_redirect', descriptor?.start ?? opToken.start, end, kids);
+  }
+
+  /** herestring_redirect: [fd] `<<<` literal */
+  private parseHerestringRedirect(descriptor: Frame | null): Frame {
+    const opToken = this.lexer.next();
+    const kids: Frame[] = [];
+    if (descriptor !== null) kids.push(descriptor);
+    kids.push(this.anon('<<<', opToken.start, opToken.end));
+    let end = opToken.end;
+    if (this.lexer.peek().type === 'word') {
+      const value = this.parseWordArgument();
+      kids.push(value);
+      end = value.end;
+    } else {
+      this.hasError = true;
+    }
+    return this.frame('herestring_redirect', descriptor?.start ?? opToken.start, end, kids);
+  }
+
+  /**
+   * heredoc_redirect: [fd] (`<<` | `<<-`) heredoc_start, then the rest of
+   * the line (arguments, redirects, a pipeline or &&/|| tail, and `;`/`&`
+   * separated follow-up statements) absorbed into the redirect — matching
+   * tree-sitter-bash's grammar, which swallows all of these into
+   * heredoc_redirect.
+   *
+   * Only ONE heredoc may be registered per line region: a second `<<`
+   * (or a `<<` inside a swallowed follow-up statement) cannot be
+   * represented as a tree — its body would have to sit inside a node whose
+   * range interleaves with the first heredoc's siblings. tree-sitter-bash
+   * 0.25.0 errors on `cat <<A <<B` for the same reason. The parser degrades
+   * such operators to ERROR nodes (see parseBrokenHeredoc) and leaves the
+   * "body" lines to be parsed as ordinary commands, mirroring the
+   * reference's recovery shape.
+   *
+   * The body/end nodes are attached later by completeHeredocs(), when the
+   * lexer reports the bodies scanned after the end of the line.
+   */
+  private parseHeredocRedirect(descriptor: Frame | null): Frame {
+    const opToken = this.lexer.next();
+    const op = this.tokenText(opToken);
+    const kids: Frame[] = [];
+    if (descriptor !== null) kids.push(descriptor);
+    kids.push(this.anon(op, opToken.start, opToken.end));
+    const redirect = this.frame('heredoc_redirect', descriptor?.start ?? opToken.start, opToken.end, kids);
+    const startToken = this.lexer.peek();
+    if (startToken.type !== 'word') {
+      this.hasError = true; // missing delimiter
+      return redirect;
+    }
+    this.lexer.next();
+    this.addKid(redirect, this.frame('heredoc_start', startToken.start, startToken.end));
+    redirect.end = startToken.end;
+    const spec = extractHeredocSpec(this.tokenText(startToken), op === '<<-');
+    this.lexer.queueHeredoc(spec);
+    this.heredocQueue.push({ frame: redirect, spec });
+    const saved = this.noHeredoc;
+    this.noHeredoc = true;
+    try {
+      this.swallowAfterHeredoc(redirect);
+    } finally {
+      this.noHeredoc = saved;
+    }
+    return redirect;
+  }
+
+  /** A `<<`/`<<-` where another heredoc is already open on this line (see
+   *  parseHeredocRedirect): recover with an ERROR node and do NOT queue a
+   *  body, so the following lines parse as ordinary commands. */
+  private parseBrokenHeredoc(descriptor: Frame | null): Frame {
+    this.hasError = true;
+    const opToken = this.lexer.next();
+    const kids: Frame[] = [];
+    if (descriptor !== null) kids.push(descriptor);
+    kids.push(this.anon(this.tokenText(opToken), opToken.start, opToken.end));
+    let end = opToken.end;
+    if (this.lexer.peek().type === 'word') {
+      const word = this.parseWordArgument();
+      kids.push(word);
+      end = word.end;
+    }
+    return this.frame('ERROR', descriptor?.start ?? opToken.start, end, kids);
+  }
+
+  /** Absorb the rest of the heredoc's line into the redirect frame. */
+  private swallowAfterHeredoc(redirect: Frame): void {
+    for (;;) {
+      const token = this.lexer.peek();
+      if (token.type === 'word') {
+        const argument = this.parseWordArgument();
+        this.addKid(redirect, argument);
+        redirect.end = argument.end;
+        continue;
+      }
+      if (token.type === 'io_number') {
+        const frame = this.parseRedirect();
+        this.addKid(redirect, frame);
+        redirect.end = frame.end;
+        continue;
+      }
+      if (token.type !== 'op') return;
+      const op = this.tokenText(token);
+      if (op === '<<' || op === '<<-') {
+        const broken = this.parseRedirect(); // second heredoc: ERROR recovery
+        this.addKid(redirect, broken);
+        redirect.end = broken.end;
+        continue;
+      }
+      if (isFileRedirectOp(op) || op === '<<<') {
+        const frame = this.parseRedirect();
+        this.addKid(redirect, frame);
+        redirect.end = frame.end;
+        continue;
+      }
+      if (op === '|' || op === '|&') {
+        this.lexer.next();
+        const kids: Frame[] = [this.anon(op, token.start, token.end)];
+        let end = token.end;
+        if (this.isStatementStart(this.lexer.peek())) {
+          const inner = this.parsePipelineTail(this.parseStatementNotPipeline());
+          kids.push(inner);
+          end = inner.end;
+        } else {
+          this.hasError = true;
+        }
+        const pipeline = this.frame('pipeline', token.start, end, kids);
+        this.addKid(redirect, pipeline);
+        redirect.end = pipeline.end;
+        continue;
+      }
+      if (op === '&&' || op === '||') {
+        this.lexer.next();
+        this.addKid(redirect, this.anon(op, token.start, token.end));
+        redirect.end = token.end;
+        if (this.isStatementStart(this.lexer.peek())) {
+          const right = this.parsePipeline();
+          this.addKid(redirect, right);
+          redirect.end = right.end;
+        } else {
+          this.hasError = true;
+        }
+        continue;
+      }
+      if (op === ';' || op === '&' || op === ';;') {
+        // Follow-up statements on the same line (e.g. `cat <<E; echo x`)
+        // are absorbed so the tree stays overlap-free once the body (which
+        // textually follows the whole line) is attached.
+        this.lexer.next();
+        this.addKid(redirect, this.anon(op, token.start, token.end));
+        redirect.end = token.end;
+        if (this.isStatementStart(this.lexer.peek())) {
+          const statement = this.parseList();
+          this.addKid(redirect, statement);
+          redirect.end = statement.end;
+        }
+        continue;
+      }
+      return;
+    }
+  }
+
+  /** Attach scanned bodies to pending heredoc redirects, in queue order. */
+  private completeHeredocs(bodies: HeredocBody[]): void {
+    for (const body of bodies) {
+      const pending = this.heredocQueue.shift();
+      if (pending === undefined) continue;
+      const { frame, spec } = pending;
+      const content = spec.quoted ? [] : this.parseHeredocContent(body.bodyStart, body.bodyEnd);
+      this.addKid(frame, this.frame('heredoc_body', body.bodyStart, body.bodyEnd, content));
+      let end = body.bodyEnd;
+      if (body.found) {
+        this.addKid(frame, this.frame('heredoc_end', body.endStart, body.endEnd));
+        end = body.endEnd;
+      } else {
+        this.hasError = true; // unterminated heredoc
+      }
+      frame.end = end;
+      // Ancestors were finalized before the body existed; extend them.
+      for (let parent = frame.parent; parent !== null; parent = parent.parent) {
+        if (parent.end < end) parent.end = end;
+      }
+    }
+  }
+
+  /** Pendings that never got a body (scope ended first): close them empty. */
+  private failOpenHeredocs(): void {
+    for (const pending of this.heredocQueue.splice(0)) {
+      this.hasError = true;
+      const at = pending.frame.end;
+      this.addKid(pending.frame, this.frame('heredoc_body', at, at));
+    }
+  }
+
+  // ------------------------------------------------------------------ words
+
+  /** Parse one argument/destination word. Adjacent word tokens (the lexer
+   *  splits { } [ ] out as single-character tokens) merge back into a single
+   *  literal, so `a{b}` is one concatenation argument. */
+  private parseWordArgument(): Frame {
+    const [start, end] = this.consumeWordRun();
+    return this.parseLiteral(start, end);
+  }
+
+  /**
+   * _literal over a source range: a single word/number/string/expansion/…,
+   * or a concatenation of adjacent pieces. The single characters { } [ ]
+   * become their own word pieces, matching tree-sitter-bash's
+   * _special_character alias — except `{N..M}` (digits only), which is a
+   * brace_expression, and a leading `((…))`, which is an arithmetic command.
+   *
+   * Guarded by literalDepth: parseLiteral ↔ parseString / parseExpansion
+   * recurse through parseDollar, and unlike command substitutions (which
+   * spawn depth-tracked sub-parsers) that chain stays inside one Parser, so
+   * it needs its own counter.
+   */
+  private parseLiteral(start: number, end: number): Frame {
+    if (this.literalDepth >= MAX_PARSE_DEPTH) {
+      this.hasError = true;
+      return this.frame('ERROR', start, end);
+    }
+    this.literalDepth++;
+    try {
+      return this.parseLiteralPieces(start, end);
+    } finally {
+      this.literalDepth--;
+    }
+  }
+
+  private parseLiteralPieces(start: number, end: number): Frame {
+    const pieces: Frame[] = [];
+    let i = start;
+    while (i < end) {
+      this.budget.progress();
+      const ch = this.source[i]!;
+      if (ch === '"') {
+        const [piece, next] = this.parseString(i, end);
+        pieces.push(piece);
+        i = next;
+        continue;
+      }
+      if (ch === "'") {
+        const close = skipSingleQuoted(this.source, this.budget, i, end);
+        if (close >= end && this.source[close - 1] !== "'") this.hasError = true; // unterminated raw string
+        pieces.push(this.frame('raw_string', i, close));
+        i = close;
+        continue;
+      }
+      if (ch === '`') {
+        const [piece, next] = this.parseBacktickSubstitution(i, end);
+        pieces.push(piece);
+        i = next;
+        continue;
+      }
+      if (ch === '$') {
+        // $"…" at the START of a literal is a translated_string (the
+        // reference produces it everywhere except command-argument
+        // position, where parseCommandArgument splits the bare `$` and the
+        // string into two arguments). Mid-literal it stays a bare `$` plus
+        // a string piece inside the concatenation.
+        if (this.source[i + 1] === '"' && i === start) {
+          const [translated, next] = this.parseString(i + 1, end);
+          pieces.push(this.frame('translated_string', i, next, [this.anon('$', i, i + 1), translated]));
+          i = next;
+          continue;
+        }
+        const dollar = this.parseDollar(i, end);
+        if (dollar !== null) {
+          pieces.push(dollar[0]);
+          i = dollar[1];
+        } else {
+          // Bare `$` (not followed by anything expandable), e.g. the `$` of
+          // a mid-literal $"…" (see above).
+          pieces.push(this.anon('$', i, i + 1));
+          i++;
+        }
+        continue;
+      }
+      if ((ch === '<' || ch === '>') && this.source[i + 1] === '(' && i + 1 < end) {
+        const [piece, next] = this.parseProcessSubstitution(i, end);
+        pieces.push(piece);
+        i = next;
+        continue;
+      }
+      if (ch === '(' && this.source[i + 1] === '(' && i + 1 < end) {
+        const [piece, next] = this.parseParenArithmetic(i, end);
+        pieces.push(piece);
+        i = next;
+        continue;
+      }
+      if (ch === '{') {
+        const brace = BRACE_EXPRESSION_RE.exec(this.source.slice(i, end));
+        if (brace !== null) {
+          const [full, low, high] = brace;
+          const lowStart = i + 1;
+          const highStart = lowStart + low!.length + 2;
+          const close = highStart + high!.length;
+          pieces.push(
+            this.frame('brace_expression', i, i + full.length, [
+              this.anon('{', i, lowStart),
+              this.frame('number', lowStart, lowStart + low!.length),
+              this.anon('..', lowStart + low!.length, highStart),
+              this.frame('number', highStart, close),
+              this.anon('}', close, close + 1),
+            ]),
+          );
+          i += full.length;
+          continue;
+        }
+        pieces.push(this.frame('word', i, i + 1));
+        i++;
+        continue;
+      }
+      if (ch === '}' || ch === '[' || ch === ']') {
+        pieces.push(this.frame('word', i, i + 1));
+        i++;
+        continue;
+      }
+      // Bare run of ordinary word characters (escapes included as-is).
+      let j = i;
+      let sinceTick = 0;
+      while (j < end) {
+        if (++sinceTick >= SCAN_TICK_INTERVAL) {
+          this.budget.progress();
+          sinceTick = 0;
+        }
+        const next = this.source[j]!;
+        if (next === '"' || next === "'" || next === '`' || next === '$') break;
+        if (next === '{' || next === '}' || next === '[' || next === ']') break;
+        if ((next === '<' || next === '>') && this.source[j + 1] === '(' && j + 1 < end) break;
+        if (next === '(' && this.source[j + 1] === '(' && j + 1 < end) break;
+        if (next === '\\' && j + 1 < end) {
+          j += 2;
+          continue;
+        }
+        j++;
+      }
+      if (j === i) j++; // defensive: never stall
+      // A bare run that is exactly a number is a number node even as one
+      // piece of a concatenation (`a[2]b` → word + number + word).
+      const numeric = NUMBER_RE.test(this.text(i, j));
+      pieces.push(this.frame(numeric ? 'number' : 'word', i, j));
+      i = j;
+    }
+    if (pieces.length === 1) {
+      const only = pieces[0]!;
+      if (only.type === 'word' && NUMBER_RE.test(this.text(only.start, only.end))) {
+        return this.frame('number', only.start, only.end);
+      }
+      return only;
+    }
+    return this.frame('concatenation', start, end, pieces);
+  }
+
+  /** string: " … " with string_content chunks and expansions inside. */
+  private parseString(start: number, rangeEnd: number): [Frame, number] {
+    const string = this.frame('string', start, start + 1, [this.anon('"', start, start + 1)]);
+    let chunkStart = start + 1;
+    let i = start + 1;
+    const flushChunk = (upto: number): void => {
+      if (upto > chunkStart) {
+        this.addKid(string, this.frame('string_content', chunkStart, upto));
+      }
+    };
+    /** A whitespace-only chunk is absorbed into the next construct's opener
+     *  (` " ${x}"` → (${) token " ${"), or into the closing quote after an
+     *  expansion (`"$x "` → (") token " \"") — the reference scanner never
+     *  emits a whitespace-only string_content at those positions. Leaf
+     *  constructs without an opener token (ansi_c_string) cannot absorb. */
+    const whitespaceOnlyChunk = (upto: number): boolean => {
+      if (upto <= chunkStart) return false;
+      return /^[ \t\r]+$/.test(this.source.slice(chunkStart, upto));
+    };
+    const absorbInto = (piece: Frame): boolean => {
+      const first = piece.children[0];
+      if (first === undefined) return false;
+      piece.start = chunkStart;
+      first.start = chunkStart;
+      return true;
+    };
+    let sinceTick = 0;
+    while (i < rangeEnd) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[i]!;
+      if (ch === '"') {
+        const last = string.children.at(-1);
+        if (whitespaceOnlyChunk(i) && last !== undefined && last.isNamed && last.type !== 'string_content') {
+          // Absorbed into the closing quote (see whitespaceOnlyChunk).
+          this.addKid(string, this.anon('"', chunkStart, i + 1));
+        } else {
+          flushChunk(i);
+          this.addKid(string, this.anon('"', i, i + 1));
+        }
+        string.end = i + 1;
+        return [string, i + 1];
+      }
+      if (ch === '\\') {
+        i += 2; // escaped characters stay inside the current content chunk
+        continue;
+      }
+      if (ch === '$') {
+        const dollar = this.parseDollar(i, rangeEnd);
+        if (dollar !== null) {
+          if (!(whitespaceOnlyChunk(i) && absorbInto(dollar[0]))) {
+            flushChunk(i);
+          }
+          this.addKid(string, dollar[0]);
+          i = dollar[1];
+          chunkStart = i;
+          continue;
+        }
+        // A `$` that starts no construct is an anonymous ($) token, splitting
+        // the content (`"s$"` → string_content "s", ($) "$").
+        flushChunk(i);
+        this.addKid(string, this.anon('$', i, i + 1));
+        i++;
+        chunkStart = i;
+        continue;
+      }
+      if (ch === '`') {
+        flushChunk(i);
+        const [piece, next] = this.parseBacktickSubstitution(i, rangeEnd);
+        this.addKid(string, piece);
+        i = next;
+        chunkStart = i;
+        continue;
+      }
+      i++;
+    }
+    // Unterminated string: keep the partial node, flag the error.
+    this.hasError = true;
+    flushChunk(rangeEnd);
+    string.end = rangeEnd;
+    return [string, rangeEnd];
+  }
+
+  /** Dispatch a $-construct at `i`: simple_expansion, expansion,
+   *  command_substitution, arithmetic_expansion, or ansi_c_string. Returns
+   *  null for a bare `$` (including the `$` of a mid-literal $"…" — see
+   *  parseLiteralPieces for the translated_string rule). */
+  private parseDollar(i: number, rangeEnd: number): [Frame, number] | null {
+    const next = this.source[i + 1];
+    if (next === '(' && i + 1 < rangeEnd) {
+      if (this.source[i + 2] === '(') return this.parseArithmeticExpansion(i, rangeEnd);
+      return this.parseCommandSubstitution(i, rangeEnd);
+    }
+    if (next === '{' && i + 1 < rangeEnd) return this.parseExpansion(i, rangeEnd);
+    if (next === '[' && i + 1 < rangeEnd) return this.parseBracketArithmetic(i, rangeEnd);
+    if (next === "'" && i + 1 < rangeEnd) return this.parseAnsiCString(i, rangeEnd);
+    if (next !== undefined && /[\w]/.test(next)) {
+      let j = i + 1;
+      while (j < rangeEnd && /[\w]/.test(this.source[j]!)) j++;
+      // `$0` is a special_variable_name; other digits are variable_name
+      // (`$1`, and `${10}` below) — matching the reference grammar, whose
+      // \w+ rule wins for everything except the standalone 0.
+      if (j === i + 2 && next === '0') {
+        return [
+          this.frame('simple_expansion', i, j, [this.anon('$', i, i + 1), this.frame('special_variable_name', i + 1, j)]),
+          j,
+        ];
+      }
+      const expansion = this.frame('simple_expansion', i, j, [
+        this.anon('$', i, i + 1),
+        this.frame('variable_name', i + 1, j),
+      ]);
+      return [expansion, j];
+    }
+    if (next !== undefined && i + 1 < rangeEnd && SPECIAL_VARIABLE_CHARS.includes(next)) {
+      const expansion = this.frame('simple_expansion', i, i + 2, [
+        this.anon('$', i, i + 1),
+        this.frame('special_variable_name', i + 1, i + 2),
+      ]);
+      return [expansion, i + 2];
+    }
+    return null;
+  }
+
+  /** ansi_c_string: $' … ' — one leaf node, \' is an escaped quote. */
+  private parseAnsiCString(i: number, rangeEnd: number): [Frame, number] {
+    let j = i + 2;
+    let sinceTick = 0;
+    while (j < rangeEnd) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[j]!;
+      if (ch === '\\') {
+        j += 2;
+        continue;
+      }
+      if (ch === "'") {
+        return [this.frame('ansi_c_string', i, j + 1), j + 1];
+      }
+      j++;
+    }
+    this.hasError = true; // unterminated ANSI-C string
+    return [this.frame('ansi_c_string', i, rangeEnd), rangeEnd];
+  }
+
+  /**
+   * expansion: ${ … } with optional prefix operators (! #), a variable_name
+   * / special_variable_name / subscript, and an optional infix operator
+   * whose value layout follows the reference:
+   *   # ## % %%     removal — the pattern is a regex node (single-quoted
+   *                   segments fold into it; double quotes stay strings)
+   *   / // /# /%    replacement — regex pattern, optional `/` + literal
+   *   ^ ^^ , ,,     case modification — optional regex
+   *   @X            transformation — two anonymous operator nodes
+   *   :             max length — arithmetic values around an optional `:`
+   *   :- := :+ :? - = + ?   default/assign/… — a word/concatenation value
+   *
+   * Guarded by literalDepth directly (not only via parseLiteral): the
+   * pattern → string → $ → expansion recursion cycle bypasses parseLiteral
+   * and would otherwise overflow the stack on pathological nesting.
+   */
+  private parseExpansion(i: number, rangeEnd: number): [Frame, number] {
+    if (this.literalDepth >= MAX_PARSE_DEPTH) {
+      this.hasError = true;
+      const scan = this.scanBalanced(i + 1, rangeEnd, '{', '}');
+      return [this.frame('ERROR', i, scan.end), scan.end];
+    }
+    this.literalDepth++;
+    try {
+      return this.parseExpansionInner(i, rangeEnd);
+    } finally {
+      this.literalDepth--;
+    }
+  }
+
+  /**
+   * Whether a ${a[…]} subscript index parses as an expression in the
+   * reference (binary/unary/parenthesized) rather than as a plain literal:
+   * true when the index contains a top-level arithmetic operator ADJACENT
+   * TO WHITESPACE (`${a[1 + 2]}` — a fused `i+1` stays a word) or starts
+   * with ++/-- (`${w[++counter]}`). Operators inside parens, brackets,
+   * quotes and $-constructs do not count.
+   */
+  private isArithmeticSubscriptIndex(start: number, end: number): boolean {
+    if (this.source.startsWith('++', start) || this.source.startsWith('--', start)) return true;
+    const isBlank = (ch: string | undefined): boolean => ch === ' ' || ch === '\t' || ch === '\r';
+    let i = start;
+    while (i < end) {
+      const ch = this.source[i]!;
+      if (ch === '\\') {
+        i += 2;
+        continue;
+      }
+      if (ch === '"' || ch === "'" || ch === '`') {
+        i++;
+        continue;
+      }
+      if (ch === '$') {
+        i = this.skipDollarConstruct(i, end);
+        continue;
+      }
+      if (ch === '(') {
+        i = this.scanBalanced(i, end, '(', ')').end;
+        continue;
+      }
+      if (ch === '[') {
+        i = this.scanBalanced(i, end, '[', ']').end;
+        continue;
+      }
+      if ('+-*/%<>=!&|^~?,'.includes(ch) && (isBlank(this.source[i - 1]) || isBlank(this.source[i + 1]))) return true;
+      i++;
+    }
+    return false;
+  }
+
+  /** Rename variable_name leaves to word (the reference uses plain _literal
+   *  operands — word, not variable_name — in subscript index expressions). */
+  private remapVariableNamesToWords(frame: Frame): void {
+    if (frame.type === 'variable_name') frame.type = 'word';
+    for (const child of frame.children) this.remapVariableNamesToWords(child);
+  }
+
+  private parseExpansionInner(i: number, rangeEnd: number): [Frame, number] {
+    const scan = this.scanBalanced(i + 1, rangeEnd, '{', '}');
+    const close = scan.end;
+    const terminated = scan.balanced;
+    if (!terminated) this.hasError = true;
+    const innerEnd = terminated ? close - 1 : close;
+    const expansion = this.frame('expansion', i, close, [this.anon('${', i, i + 2)]);
+    let j = i + 2;
+    // Prefix operators: ${#v} (length), ${!v} (indirect).
+    let bangPrefix = false;
+    while (j < innerEnd && (this.source[j] === '#' || this.source[j] === '!')) {
+      const after = this.source[j + 1];
+      if (after === undefined || j + 1 >= innerEnd || !/[\w@*#?$!-]/.test(after)) break;
+      if (this.source[j] === '!') bangPrefix = true;
+      this.addKid(expansion, this.anon(this.source[j]!, j, j + 1));
+      j++;
+    }
+    // The variable itself.
+    let hasName = false;
+    if (j < innerEnd && /[\w]/.test(this.source[j]!)) {
+      let nameEnd = j;
+      while (nameEnd < innerEnd && /[\w]/.test(this.source[nameEnd]!)) nameEnd++;
+      // ${0} is a special_variable_name; other digits are variable_name.
+      const special = nameEnd === j + 1 && this.source[j] === '0';
+      this.addKid(expansion, this.frame(special ? 'special_variable_name' : 'variable_name', j, nameEnd));
+      j = nameEnd;
+      hasName = true;
+    } else if (j < innerEnd && (this.source[j] === '#' || this.source[j] === '!') && j + 1 === innerEnd) {
+      // A lone # / ! in variable position is an anonymous token in the
+      // reference (`${#}`, `${!#}`, `${!##}`, `${#!}` — the length/indirect
+      // operator with no name), not a special_variable_name.
+      this.addKid(expansion, this.anon(this.source[j]!, j, j + 1));
+      j++;
+    } else if (j < innerEnd && SPECIAL_VARIABLE_CHARS.includes(this.source[j]!)) {
+      this.addKid(expansion, this.frame('special_variable_name', j, j + 1));
+      j++;
+      hasName = true;
+    }
+    // Subscript: ${a[0]} — replaces the bare variable_name child. A simple
+    // index is a literal; an index with a top-level arithmetic operator or a
+    // ++/-- prefix is an expression in the reference (`${a[1 + 2]}` →
+    // binary_expression, `${w[++c]}` → unary_expression).
+    if (j < innerEnd && this.source[j] === '[' && hasName) {
+      const sub = this.scanBalanced(j, rangeEnd, '[', ']');
+      const subEnd = sub.end;
+      if (!sub.balanced) this.hasError = true;
+      const indexEnd = sub.balanced ? subEnd - 1 : subEnd;
+      const variable = expansion.children.pop()!;
+      const subscript = this.frame('subscript', variable.start, subEnd, [
+        variable,
+        this.anon('[', j, j + 1),
+      ]);
+      if (indexEnd > j + 1) {
+        const indexText = this.text(j + 1, indexEnd);
+        if (indexText === '*' || indexText === '@') {
+          // ${a[*]} / ${a[@]} — all-indices subscripts are plain words.
+          this.addKid(subscript, this.parseLiteral(j + 1, indexEnd));
+        } else if (this.isArithmeticSubscriptIndex(j + 1, indexEnd)) {
+          const st = this.newExprState(j + 1, indexEnd, 'arith');
+          const index = this.parseExpression(st, 0);
+          if (index !== null) {
+            this.remapVariableNamesToWords(index);
+            this.addKid(subscript, index);
+          }
+          const leftover = this.exprLeftover(st);
+          if (leftover !== null) this.addKid(subscript, leftover);
+        } else {
+          this.addKid(subscript, this.parseLiteral(j + 1, indexEnd));
+        }
+      } else {
+        this.hasError = true;
+      }
+      if (sub.balanced) this.addKid(subscript, this.anon(']', subEnd - 1, subEnd));
+      this.addKid(expansion, subscript);
+      j = subEnd;
+    }
+    // ${!prefix*} / ${!name@}: a trailing * or @ after an indirect name.
+    if (bangPrefix && j < innerEnd && (this.source[j] === '*' || this.source[j] === '@')) {
+      this.addKid(expansion, this.anon(this.source[j]!, j, j + 1));
+      j++;
+    }
+    // Infix operator plus an optional value.
+    if (j < innerEnd) {
+      j = this.parseExpansionInfix(expansion, j, innerEnd);
+    }
+    if (terminated) this.addKid(expansion, this.anon('}', close - 1, close));
+    return [expansion, close];
+  }
+
+  /** The infix part of an expansion (see parseExpansion). Returns the new
+   *  scan position. */
+  private parseExpansionInfix(expansion: Frame, j: number, innerEnd: number): number {
+    const one = this.source[j]!;
+    const two = this.source.slice(j, j + 2);
+    if (two === '##' || two === '%%' || one === '#' || one === '%') {
+      const operator = two === '##' || two === '%%' ? two : one;
+      this.addKid(expansion, this.anon(operator, j, j + operator.length));
+      return this.parseExpansionPattern(expansion, j + operator.length, innerEnd);
+    }
+    if (two === '//' || two === '/#' || two === '/%' || one === '/') {
+      const operator = two === '//' || two === '/#' || two === '/%' ? two : one;
+      this.addKid(expansion, this.anon(operator, j, j + operator.length));
+      let pos = this.parseExpansionPattern(expansion, j + operator.length, innerEnd, '/');
+      if (pos < innerEnd && this.source[pos] === '/') {
+        this.addKid(expansion, this.anon('/', pos, pos + 1));
+        pos++;
+        if (pos < innerEnd) {
+          for (const piece of this.parseExpansionValue(pos, innerEnd)) this.addKid(expansion, piece);
+          pos = innerEnd;
+        }
+      }
+      return pos;
+    }
+    if (two === '^^' || two === ',,' || one === '^' || one === ',') {
+      const operator = two === '^^' || two === ',,' ? two : one;
+      this.addKid(expansion, this.anon(operator, j, j + operator.length));
+      return this.parseExpansionPattern(expansion, j + operator.length, innerEnd);
+    }
+    if (one === '@' && j + 1 < innerEnd) {
+      this.addKid(expansion, this.anon('@', j, j + 1));
+      this.addKid(expansion, this.anon(this.source[j + 1]!, j + 1, j + 2));
+      return j + 2;
+    }
+    if (one === ':' && two !== ':-' && two !== ':=' && two !== ':+' && two !== ':?') {
+      // Max length: ${v:offset:length} — arithmetic values.
+      this.addKid(expansion, this.anon(':', j, j + 1));
+      let pos = this.parseMaxLengthValue(expansion, j + 1, innerEnd);
+      if (pos < innerEnd && this.source[pos] === ':') {
+        this.addKid(expansion, this.anon(':', pos, pos + 1));
+        pos = this.parseMaxLengthValue(expansion, pos + 1, innerEnd);
+      }
+      if (pos < innerEnd) {
+        this.hasError = true; // extra content after ${v:o:l}
+        this.addKid(expansion, this.parseLiteral(pos, innerEnd));
+        pos = innerEnd;
+      }
+      return pos;
+    }
+    for (const operator of [':-', ':=', ':+', ':?', '-', '=', '+', '?']) {
+      if (this.source.startsWith(operator, j) && j + operator.length <= innerEnd) {
+        this.addKid(expansion, this.anon(operator, j, j + operator.length));
+        const pos = j + operator.length;
+        if (pos < innerEnd) {
+          for (const piece of this.parseExpansionValue(pos, innerEnd)) this.addKid(expansion, piece);
+        }
+        return innerEnd;
+      }
+    }
+    // Unknown infix content: keep it as a literal (and flag the error).
+    this.hasError = true;
+    this.addKid(expansion, this.parseLiteral(j, innerEnd));
+    return innerEnd;
+  }
+
+  /** A removal/replacement pattern inside ${…}: regex chunks interleaved
+   *  with string pieces. Rules from the reference scanner:
+   *  - leading whitespace after the operator drops out of the tree when
+   *    more pattern follows (${G%% *} → regex "*"); an all-whitespace
+   *    pattern stays a regex (${B[0]# } → regex " ");
+   *  - in the REPLACEMENT pattern (before the separator `/`), single-quoted
+   *    segments fold INTO the surrounding regex chunk (${M/'N'X/O} →
+   *    regex "'N'X"); in removal patterns (# ## % %% ^ ^ , ,,) they stay
+   *    separate raw_string nodes, and double quotes always stay strings;
+   *  - the replacement separator `/` ends the pattern even when escaped
+   *    (a pattern ending in backslash-slash → regex keeps the backslash,
+   *    then the `/` operator). */
+  private parseExpansionPattern(expansion: Frame, j: number, innerEnd: number, stop?: string): number {
+    let pos = j;
+    while (pos < innerEnd && (this.source[pos] === ' ' || this.source[pos] === '\t' || this.source[pos] === '\r')) {
+      pos++;
+    }
+    if (pos > j && (pos >= innerEnd || (stop !== undefined && this.source[pos] === stop))) {
+      // An all-whitespace pattern stays a regex node.
+      this.addKid(expansion, this.frame('regex', j, pos));
+      return pos;
+    }
+    const mergeQuotes = stop === '/';
+    while (pos < innerEnd) {
+      this.budget.progress();
+      const ch = this.source[pos]!;
+      if (ch === '"') {
+        const [piece, next] = this.parseString(pos, innerEnd);
+        this.addKid(expansion, piece);
+        pos = next;
+        continue;
+      }
+      if (ch === "'" && !mergeQuotes) {
+        const close = skipSingleQuoted(this.source, this.budget, pos, innerEnd);
+        this.addKid(expansion, this.frame('raw_string', pos, close));
+        pos = close;
+        continue;
+      }
+      // One regex chunk: bare text, plus single-quoted segments when they
+      // fold into the regex (replacement patterns).
+      let k = pos;
+      let sinceTick = 0;
+      while (k < innerEnd && this.source[k] !== '"' && this.source[k] !== stop) {
+        if (++sinceTick >= SCAN_TICK_INTERVAL) {
+          this.budget.progress();
+          sinceTick = 0;
+        }
+        const ck = this.source[k]!;
+        if (ck === "'") {
+          if (!mergeQuotes) break;
+          k = skipSingleQuoted(this.source, this.budget, k, innerEnd);
+          continue;
+        }
+        if (ck === '\\' && k + 1 < innerEnd) {
+          if (stop !== undefined && this.source[k + 1] === stop) {
+            k++; // the escape backslash stays in the chunk; the stop char ends it
+            break;
+          }
+          k += 2;
+          continue;
+        }
+        k++;
+      }
+      if (k === pos) break; // defensive: only the stop character remains
+      this.addKid(expansion, this.frame('regex', pos, k));
+      pos = k;
+    }
+    return pos;
+  }
+
+  /** A default/assign value inside ${…} (`${v:-word}`), as a list of frames
+   *  (usually one — the reference sometimes attaches the value as several
+   *  direct children of the expansion). Layout rules from the reference:
+   *  - a bare value is a word (never a number — `${v:-1}` is a word). An
+   *    interior space splits it into word + word (the second keeps the
+   *    leading space: `${v:-d e f}` → word "d" + word " e f"); trailing-only
+   *    whitespace stays inside the single word (`${v:-) }` → word ") "), and
+   *    a leading `? ` (Gentoo atom syntax) keeps the whole value as one word;
+   *  - bare word pieces directly before a quote/construct are right-trimmed
+   *    (`${v:-command '$1' x}` → word "command", raw_string, word " x") and
+   *    a whitespace-only piece before a construct drops out entirely
+   *    (`${2+ ${2}}` → just the expansion);
+   *  - a command_substitution followed by a bare remainder attaches as two
+   *    children without a concatenation wrapper. */
+  private parseExpansionValue(start: number, end: number): Frame[] {
+    const raw = this.text(start, end);
+    if (!/["'$`\\]/.test(raw)) {
+      // Bare value. Rules from the reference's _expansion_word scanner:
+      // - an interior space splits it into word + word (the second keeps
+      //   the leading space: `${v:-d e f}` → word "d" + word " e f");
+      // - leading or trailing whitespace stays inside the single word
+      //   (`${v:- -quiet}` → word " -quiet", `${v:-) }` → word ") "), and a
+      //   leading `? ` (Gentoo atom syntax) keeps the whole value as one
+      //   word. Whitespace only drops out directly before a quote or a
+      //   construct (handled by the piece path below).
+      if (this.source.startsWith('? ', start)) return [this.frame('word', start, end)];
+      for (let k = start + 1; k < end; k++) {
+        const ch = this.source[k]!;
+        if (ch !== ' ' && ch !== '\t' && ch !== '\r') continue;
+        // Split at the first interior space (non-space text follows it).
+        if (/[^ \t\r]/.test(this.source.slice(k + 1, end))) {
+          return [
+            this.frame('concatenation', start, end, [
+              this.frame('word', start, k),
+              this.frame('word', k, end),
+            ]),
+          ];
+        }
+        break;
+      }
+      return [this.frame('word', start, end)];
+    }
+    const value = this.parseLiteral(start, end);
+    const pieces = value.type === 'concatenation' ? [...value.children] : [value];
+    // Right-trim bare word pieces that directly precede another piece (the
+    // whitespace drops out of the tree); drop pieces that become empty.
+    // Pieces starting with `(` keep their spaces (the reference's
+    // expansion_word scanner treats them via its paren branch).
+    const trimmed: Frame[] = [];
+    for (let p = 0; p < pieces.length; p++) {
+      const piece = pieces[p]!;
+      if (piece.type === 'word' && p + 1 < pieces.length && this.source[piece.start] !== '(') {
+        let newEnd = piece.end;
+        while (newEnd > piece.start && (this.source[newEnd - 1] === ' ' || this.source[newEnd - 1] === '\t' || this.source[newEnd - 1] === '\r')) {
+          newEnd--;
+        }
+        if (newEnd === piece.start) continue;
+        trimmed.push(newEnd === piece.end ? piece : this.frame('word', piece.start, newEnd));
+      } else {
+        trimmed.push(piece);
+      }
+    }
+    // command_substitution + bare remainder: two direct children.
+    if (trimmed.length === 2 && trimmed[0]!.type === 'command_substitution') return trimmed;
+    if (trimmed.length === 1) return trimmed;
+    if (trimmed.length === 0) return [];
+    return [this.frame('concatenation', start, end, trimmed)];
+  }
+
+  /** One arithmetic value of ${v:offset:length} (up to `:` or innerEnd). */
+  private parseMaxLengthValue(expansion: Frame, j: number, innerEnd: number): number {
+    let end = j;
+    let sinceTick = 0;
+    while (end < innerEnd && this.source[end] !== ':') {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[end]!;
+      if (ch === '\\') {
+        end += 2;
+        continue;
+      }
+      if (ch === '$') {
+        end = this.skipDollarConstruct(end, innerEnd);
+        continue;
+      }
+      if (ch === '(') {
+        end = this.scanBalanced(end, innerEnd, '(', ')').end;
+        continue;
+      }
+      end++;
+    }
+    if (end > j) {
+      // In max-length context a negative offset is a number literal, not a
+      // unary expression (`${x: -5}` → number "-5" in the reference).
+      const negative = /^\s*(-\d+)\s*$/.exec(this.text(j, end));
+      if (negative !== null) {
+        const start = j + this.text(j, end).indexOf('-');
+        this.addKid(expansion, this.frame('number', start, start + negative[1]!.length));
+        return end;
+      }
+      const st = this.newExprState(j, end, 'arith');
+      const value = this.parseExpression(st, 0);
+      if (value !== null) {
+        this.addKid(expansion, value);
+      }
+      const leftover = this.exprLeftover(st);
+      if (leftover !== null) this.addKid(expansion, leftover);
+    }
+    return end;
+  }
+
+  /** Skip a $-construct without building nodes (max-length scanning). */
+  private skipDollarConstruct(i: number, end: number): number {
+    const next = this.source[i + 1];
+    if (next === '(') return this.scanBalanced(i + 1, end, '(', ')').end;
+    if (next === '{') return this.scanBalanced(i + 1, end, '{', '}').end;
+    if (next !== undefined && /[\w]/.test(next)) {
+      let j = i + 1;
+      while (j < end && /[\w]/.test(this.source[j]!)) j++;
+      return j;
+    }
+    return i + 1 < end ? i + 2 : i + 1;
+  }
+
+  /** command_substitution: $( _statements ) — or the special $( redirect )
+   *  form (`$(< file)`), which holds a bare file_redirect without the
+   *  redirected_statement wrapper in the reference. The close-paren scan is
+   *  case-aware (see scanBalancedStatements): a case_item pattern `)` must
+   *  not close the substitution early. */
+  private parseCommandSubstitution(i: number, rangeEnd: number): [Frame, number] {
+    const scan = scanBalancedStatements(this.source, this.budget, i + 1, rangeEnd);
+    const close = scan.end;
+    if (!scan.balanced) this.hasError = true;
+    const innerEnd = scan.balanced ? close - 1 : close;
+    const substitution = this.frame('command_substitution', i, close, [this.anon('$(', i, i + 2)]);
+    let children = this.parseScopedStatements(i + 2, innerEnd);
+    if (
+      children.length === 1 &&
+      children[0]!.type === 'redirected_statement' &&
+      children[0]!.children.length > 0 &&
+      children[0]!.children.every((child) => child.type === 'file_redirect' || child.type === 'herestring_redirect')
+    ) {
+      children = children[0]!.children;
+    }
+    for (const child of children) {
+      this.addKid(substitution, child);
+    }
+    if (scan.balanced) this.addKid(substitution, this.anon(')', close - 1, close));
+    return [substitution, close];
+  }
+
+  /** command_substitution: ` _statements ` (legacy backtick form). */
+  private parseBacktickSubstitution(i: number, rangeEnd: number): [Frame, number] {
+    const close = skipBacktick(this.source, this.budget, i, rangeEnd);
+    const terminated = close > i + 1 && this.source[close - 1] === '`';
+    if (!terminated) this.hasError = true;
+    const innerEnd = terminated ? close - 1 : close;
+    const substitution = this.frame('command_substitution', i, close, [this.anon('`', i, i + 1)]);
+    if (innerEnd > i + 1) {
+      for (const child of this.parseScopedStatements(i + 1, innerEnd)) {
+        this.addKid(substitution, child);
+      }
+    }
+    if (terminated) this.addKid(substitution, this.anon('`', close - 1, close));
+    return [substitution, close];
+  }
+
+  /** process_substitution: ( <( | >( ) _statements ) — case-aware close
+   *  scan, like command_substitution. */
+  private parseProcessSubstitution(i: number, rangeEnd: number): [Frame, number] {
+    const scan = scanBalancedStatements(this.source, this.budget, i + 1, rangeEnd);
+    const close = scan.end;
+    if (!scan.balanced) this.hasError = true;
+    const innerEnd = scan.balanced ? close - 1 : close;
+    const opener = this.source[i]!;
+    const substitution = this.frame('process_substitution', i, close, [this.anon(`${opener}(`, i, i + 2)]);
+    for (const child of this.parseScopedStatements(i + 2, innerEnd)) {
+      this.addKid(substitution, child);
+    }
+    if (scan.balanced) this.addKid(substitution, this.anon(')', close - 1, close));
+    return [substitution, close];
+  }
+
+  // ------------------------------------------------------ expression engine
+
+  private newExprState(start: number, end: number, mode: ExprMode): ExprState {
+    return { pos: start, end, mode, lookahead: null, parenDepth: 0, expectOperator: false };
+  }
+
+  private exprPeek(st: ExprState): ExprToken {
+    st.lookahead ??= this.scanExprToken(st);
+    return st.lookahead;
+  }
+
+  /** When unconsumed input remains in the expression state, wrap it in an
+   *  ERROR node (covering the rest of the region, so no source text
+   *  silently drops out of the tree) and flag the error. */
+  private exprLeftover(st: ExprState): Frame | null {
+    const token = this.exprPeek(st);
+    if (token.kind === 'end') return null;
+    this.hasError = true;
+    return this.frame('ERROR', token.start, st.end);
+  }
+
+  private exprNext(st: ExprState): ExprToken {
+    const token = this.exprPeek(st);
+    st.lookahead = null;
+    return token;
+  }
+
+  /** arithmetic_expansion: $(( expr (, expr)* )) */
+  private parseArithmeticExpansion(i: number, rangeEnd: number): [Frame, number] {
+    const scan = this.scanBalanced(i + 1, rangeEnd, '(', ')');
+    const close = scan.end;
+    // The content sits between `$((` and the final `))`.
+    const closed = scan.balanced && close - 2 >= i + 3 && this.source[close - 2] === ')';
+    if (!closed) this.hasError = true;
+    const innerStart = Math.min(i + 3, close);
+    const innerEnd = closed ? close - 2 : close;
+    const expansion = this.frame('arithmetic_expansion', i, close, [this.anon('$((', i, innerStart)]);
+    this.addArithmeticChildren(expansion, innerStart, innerEnd);
+    if (closed) {
+      this.addKid(expansion, this.anon('))', innerEnd, close));
+    }
+    return [expansion, close];
+  }
+
+  /** arithmetic_expansion without the `$`: (( expr )) at word position. */
+  private parseParenArithmetic(i: number, rangeEnd: number): [Frame, number] {
+    const scan = this.scanBalanced(i, rangeEnd, '(', ')');
+    const close = scan.end;
+    const closed = scan.balanced && close - 2 >= i + 2 && this.source[close - 2] === ')';
+    if (!closed) this.hasError = true;
+    const innerStart = Math.min(i + 2, close);
+    const innerEnd = closed ? close - 2 : close;
+    const expansion = this.frame('arithmetic_expansion', i, close, [this.anon('((', i, innerStart)]);
+    this.addArithmeticChildren(expansion, innerStart, innerEnd);
+    if (closed) {
+      this.addKid(expansion, this.anon('))', innerEnd, close));
+    }
+    return [expansion, close];
+  }
+
+  /** arithmetic_expansion: $[ expr ] (legacy form). */
+  private parseBracketArithmetic(i: number, rangeEnd: number): [Frame, number] {
+    const scan = this.scanBalanced(i + 1, rangeEnd, '[', ']');
+    const close = scan.end;
+    if (!scan.balanced) this.hasError = true;
+    const innerEnd = scan.balanced ? close - 1 : close;
+    const expansion = this.frame('arithmetic_expansion', i, close, [this.anon('$[', i, i + 2)]);
+    this.addArithmeticChildren(expansion, i + 2, innerEnd);
+    if (scan.balanced) {
+      this.addKid(expansion, this.anon(']', close - 1, close));
+    }
+    return [expansion, close];
+  }
+
+  /** The comma-separated expression list inside an arithmetic_expansion. */
+  private addArithmeticChildren(expansion: Frame, start: number, end: number): void {
+    const st = this.newExprState(start, end, 'arith');
+    for (;;) {
+      const expression = this.parseExpression(st, 0);
+      if (expression !== null) this.addKid(expansion, expression);
+      const token = this.exprPeek(st);
+      if (token.kind === 'op' && token.text === ',') {
+        this.exprNext(st);
+        this.addKid(expansion, this.anon(',', token.start, token.end));
+        continue;
+      }
+      break;
+    }
+    const leftover = this.exprLeftover(st);
+    if (leftover !== null) this.addKid(expansion, leftover);
+  }
+
+  /**
+   * Pratt parser for arithmetic / c-style-for / test expressions. Mirrors
+   * grammar.js: left-associative binaries parse their right side at
+   * level+1, `**` is right-associative only in test mode (left in
+   * arithmetic), prefix -/+/~/! take their operand at level 11 (so
+   * `-x + ~y` is -(x + ~y)) and prefix ++/-- at level 17, postfix ++/-- sit
+   * at level 18, ternary at level 2.
+   */
+  private parseExpression(st: ExprState, minPrecedence: number): Frame | null {
+    let left: Frame | null;
+    const head = this.exprPeek(st);
+    // `;` and `,` are separators (c-style for parts, comma lists), handled
+    // by the caller — never operands.
+    if (head.kind === 'op' && (head.text === ';' || head.text === ',')) return null;
+    if (head.kind === 'op' && (head.text === '++' || head.text === '--')) {
+      this.exprNext(st);
+      const kids: Frame[] = [this.anon(head.text, head.start, head.end)];
+      const operand = this.parseExpression(st, PREC_PREFIX);
+      if (operand === null) this.hasError = true;
+      else kids.push(operand);
+      left = this.frame('unary_expression', head.start, this.endOf(kids, head.end), kids);
+    } else if (head.kind === 'op' && (head.text === '!' || head.text === '~' || head.text === '+' || head.text === '-')) {
+      this.exprNext(st);
+      const kids: Frame[] = [this.anon(head.text, head.start, head.end)];
+      const operand = this.parseExpression(st, PREC_UNARY);
+      if (operand === null) this.hasError = true;
+      else kids.push(operand);
+      left = this.frame('unary_expression', head.start, this.endOf(kids, head.end), kids);
+    } else if (head.kind === 'testop') {
+      this.exprNext(st);
+      // A test_operator with no usable operand ahead is just a word
+      // (`[[ $a == -foo ]]` → word "-foo", `[[ -foo == x ]]` → word,
+      // `[[ -f ]]` → word — the reference's fallback when the operator
+      // reading fails).
+      const after = this.exprPeek(st);
+      const demote =
+        after.kind === 'end' ||
+        after.kind === 'rparen' ||
+        (after.kind === 'op' &&
+          (after.text === '=' || after.text === '==' || after.text === '!=' || after.text === '=~' || after.text === '&&' || after.text === '||'));
+      if (demote) {
+        left = this.frame('word', head.start, head.end);
+      } else {
+        const kids: Frame[] = [this.frame('test_operator', head.start, head.end)];
+        const operand = this.parseExpression(st, PREC_TEST);
+        if (operand === null) this.hasError = true;
+        else kids.push(operand);
+        left = this.frame('unary_expression', head.start, this.endOf(kids, head.end), kids);
+      }
+    } else {
+      left = this.parseExprPrimary(st);
+    }
+    if (left === null) return null;
+    // A complete operand: what follows is operator position (test mode).
+    st.expectOperator = true;
+    for (;;) {
+      const token = this.exprPeek(st);
+      if (token.kind === 'op' && (token.text === '++' || token.text === '--') && PREC_POSTFIX >= minPrecedence) {
+        this.exprNext(st);
+        left = this.frame('postfix_expression', left.start, token.end, [
+          left,
+          this.anon(token.text, token.start, token.end),
+        ]);
+        continue;
+      }
+      if (token.kind === 'op' && token.text === '?' && PREC_TERNARY >= minPrecedence) {
+        this.exprNext(st);
+        st.expectOperator = false;
+        const kids: Frame[] = [left, this.anon('?', token.start, token.end)];
+        const consequence = this.parseExpression(st, 0);
+        if (consequence === null) this.hasError = true;
+        else kids.push(consequence);
+        const colon = this.exprPeek(st);
+        if (colon.kind === 'op' && colon.text === ':') {
+          this.exprNext(st);
+          st.expectOperator = false;
+          kids.push(this.anon(':', colon.start, colon.end));
+        } else {
+          this.hasError = true; // missing : in ternary
+        }
+        const alternative = this.parseExpression(st, PREC_TERNARY + 1);
+        if (alternative === null) this.hasError = true;
+        else kids.push(alternative);
+        left = this.frame('ternary_expression', left.start, this.endOf(kids, token.end), kids);
+        st.expectOperator = true;
+        continue;
+      }
+      const isTestOp = token.kind === 'testop';
+      const precedence = isTestOp ? PREC_TEST : token.kind === 'op' ? EXPRESSION_PRECEDENCE[token.text] : undefined;
+      if (precedence === undefined || precedence < minPrecedence) break;
+      this.exprNext(st);
+      st.expectOperator = false;
+      // `=~` in a test command takes a raw regex (or string/expansion) as
+      // its right side instead of a full expression.
+      if (st.mode === 'test' && token.kind === 'op' && token.text === '=~') {
+        const right = this.parseTestRegex(st);
+        const kids: Frame[] = [left, this.anon('=~', token.start, token.end)];
+        if (right === null) this.hasError = true;
+        else kids.push(...right);
+        left = this.frame('binary_expression', left.start, this.endOf(kids, token.end), kids);
+        st.expectOperator = true;
+        continue;
+      }
+      const rightPrecedence = token.text === '**' && st.mode === 'test' ? precedence : precedence + 1;
+      let right: Frame | null;
+      // A pattern-shaped right side after ==/!= (extglob group, or glob
+      // text mixed with a quote/expansion) — see tryParseTestPattern.
+      if (st.mode === 'test' && (token.text === '==' || token.text === '!=')) {
+        const pattern = this.tryParseTestPattern(st);
+        if (pattern !== null) {
+          const kids: Frame[] = [left, this.anon(token.text, token.start, token.end), ...pattern.frames];
+          left = this.frame('binary_expression', left.start, pattern.end, kids);
+          st.expectOperator = true;
+          continue;
+        }
+      }
+      right = this.parseExpression(st, rightPrecedence);
+      if (right === null) {
+        this.hasError = true; // missing right operand
+      } else if (st.mode === 'test' && token.kind === 'op') {
+        right = this.convertTestRightSide(token.text, right, st);
+      }
+      const operator =
+        token.kind === 'testop'
+          ? this.frame('test_operator', token.start, token.end)
+          : this.anon(token.text, token.start, token.end);
+      const kids: Frame[] = right === null ? [left, operator] : [left, operator, right];
+      left = this.frame('binary_expression', left.start, this.endOf(kids, token.end), kids);
+      st.expectOperator = true;
+    }
+    return left;
+  }
+
+  private parseExprPrimary(st: ExprState): Frame | null {
+    const token = this.exprPeek(st);
+    switch (token.kind) {
+      case 'end':
+      case 'rparen':
+        return null;
+      case 'number':
+        this.exprNext(st);
+        return this.frame('number', token.start, token.end);
+      case 'ident': {
+        this.exprNext(st);
+        if (st.mode === 'c') {
+          // `name = value` in a c-style for header is a variable_assignment
+          // (all other assignment operators are binary_expressions).
+          const next = this.exprPeek(st);
+          if (next.kind === 'op' && next.text === '=') {
+            this.exprNext(st);
+            const kids: Frame[] = [
+              this.frame('variable_name', token.start, token.end),
+              this.anon('=', next.start, next.end),
+            ];
+            const value = this.parseExpression(st, 0);
+            if (value === null) this.hasError = true;
+            else kids.push(value);
+            return this.frame('variable_assignment', token.start, this.endOf(kids, next.end), kids);
+          }
+          return this.frame('word', token.start, token.end);
+        }
+        return this.frame('variable_name', token.start, token.end);
+      }
+      case 'word': {
+        this.exprNext(st);
+        // In a test command a `-digits` operand is unary minus, not a
+        // number literal (`[[ $a == -1 ]]` → unary(-, number 1) — the
+        // opposite of the max-length rule in ${v: -1}).
+        if (st.mode === 'test' && /^-\d+$/.test(token.text)) {
+          return this.frame('unary_expression', token.start, token.end, [
+            this.anon('-', token.start, token.start + 1),
+            this.frame('number', token.start + 1, token.end),
+          ]);
+        }
+        return this.parseLiteral(token.start, token.end);
+      }
+      case 'subst':
+      case 'string':
+        this.exprNext(st);
+        return token.frame!;
+      case 'lparen': {
+        this.exprNext(st);
+        if (this.exprDepth >= MAX_PARSE_DEPTH) {
+          this.hasError = true;
+          st.pos = st.end;
+          st.lookahead = null;
+          return this.frame('ERROR', token.start, st.end);
+        }
+        this.exprDepth++;
+        st.parenDepth++;
+        const kids: Frame[] = [this.anon('(', token.start, token.end)];
+        const inner = this.parseExpression(st, 0);
+        if (inner !== null) kids.push(inner);
+        if (st.mode === 'c') {
+          // _c_parenthesized_expression: comma-separated expressions.
+          for (;;) {
+            const comma = this.exprPeek(st);
+            if (comma.kind !== 'op' || comma.text !== ',') break;
+            this.exprNext(st);
+            kids.push(this.anon(',', comma.start, comma.end));
+            const next = this.parseExpression(st, 0);
+            if (next === null) {
+              this.hasError = true;
+              break;
+            }
+            kids.push(next);
+          }
+        }
+        let end = this.endOf(kids, token.end);
+        const close = this.exprPeek(st);
+        if (close.kind === 'rparen') {
+          this.exprNext(st);
+          kids.push(this.anon(')', close.start, close.end));
+          end = close.end;
+        } else {
+          this.hasError = true; // unterminated parenthesized expression
+        }
+        st.parenDepth--;
+        this.exprDepth--;
+        return this.frame('parenthesized_expression', token.start, end, kids);
+      }
+      case 'op':
+      case 'testop':
+      case 'unknown': {
+        // An operator where an operand was expected: recover.
+        this.exprNext(st);
+        this.hasError = true;
+        return this.frame('ERROR', token.start, token.end);
+      }
+    }
+  }
+
+  /** The right side of `=~` in a test command, as a list of frames. Quote
+   *  rules from the reference scanner: a regex STARTING with a single quote
+   *  folds the quoted segments into the regex token (`' boop '(.*)$` is one
+   *  regex) unless the whole thing is just a quoted string plus plain word
+   *  characters (`'a'b` → raw_string + word); a quote appearing mid-run
+   *  makes the whole thing a plain concatenation (`a'b'c`). Whitespace
+   *  inside (…) stays inside the regex token. */
+  private parseTestRegex(st: ExprState): Frame[] | null {
+    let i = st.pos;
+    while (i < st.end && (this.source[i] === ' ' || this.source[i] === '\t' || this.source[i] === '\r')) i++;
+    st.pos = i;
+    st.lookahead = null;
+    if (i >= st.end) return null;
+    const ch = this.source[i]!;
+    if (ch === '"') {
+      const [piece, next] = this.parseString(i, st.end);
+      st.pos = next;
+      return [piece];
+    }
+    if (ch === '$') {
+      // An expansion holds the pattern: parse a normal operand.
+      const operand = this.parseExpression(st, PREC_TEST + 1);
+      return operand === null ? null : [operand];
+    }
+    // Scan the run as one regex token: it ends at whitespace that is not
+    // protected by an open paren (the reference scanner includes whitespace
+    // inside (…) in the token), at an unbalanced closer, or at the region
+    // end. Single quotes toggle an in-quote state; `$`-constructs are raw
+    // text (the scanner does not split them out here).
+    let j = i;
+    let inQuote = false;
+    let hasQuote = false;
+    let parenDepth = 0;
+    let sinceTick = 0;
+    while (j < st.end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const c = this.source[j]!;
+      if (!inQuote && parenDepth === 0 && (c === ' ' || c === '\t' || c === '\r' || c === '\n')) break;
+      if (c === "'") {
+        hasQuote = true;
+        inQuote = !inQuote;
+        j++;
+        continue;
+      }
+      if (c === '\\') {
+        j += 2;
+        continue;
+      }
+      if (!inQuote) {
+        if (c === '(') parenDepth++;
+        else if (c === ')') {
+          if (parenDepth === 0) break;
+          parenDepth--;
+        }
+      }
+      j++;
+    }
+    st.pos = j;
+    if (ch === "'") {
+      // Quoted string plus plain word characters → a plain literal.
+      if (/^'[^']*'\w*$/.test(this.text(i, j))) return [this.parseLiteral(i, j)];
+      return [this.frame('regex', i, j)];
+    }
+    if (hasQuote) return [this.parseLiteral(i, j)];
+    return [this.frame('regex', i, j)];
+  }
+
+  /**
+   * Pattern-shaped right side of ==/!= in a test command (the grammar's
+   * _extglob_blob). Handles what a plain expression parse cannot:
+   *  - a whole extglob group pattern (+(a|b), X/@(default).vim with X a
+   *    glob, *([0-9])([0-9])) as ONE extglob_pattern, when the scanner's
+   *    first/second character rules accept it;
+   *  - glob text mixed with one quote/expansion (*${var2}*, *"7f 4c"*),
+   *    split into extglob_pattern pieces around the construct.
+   * Returns null (no state touched) for plain operands — those go through
+   * the normal expression parse plus convertTestRightSide.
+   */
+  private tryParseTestPattern(st: ExprState): { frames: Frame[]; end: number } | null {
+    let i = st.pos;
+    while (i < st.end && (this.source[i] === ' ' || this.source[i] === '\t' || this.source[i] === '\r')) i++;
+    if (i >= st.end) return null;
+    // Find the end of the pattern: unquoted whitespace at paren depth 0.
+    let depth = 0;
+    let j = i;
+    let sawGroup = false;
+    let sawConstruct = false;
+    while (j < st.end) {
+      const ch = this.source[j]!;
+      if (ch === '\\') {
+        j += 2;
+        continue;
+      }
+      if (ch === '"') {
+        sawConstruct = true;
+        j = skipDoubleQuoted(this.source, this.budget, j, st.end);
+        continue;
+      }
+      if (ch === "'") {
+        sawConstruct = true;
+        j = skipSingleQuoted(this.source, this.budget, j, st.end);
+        continue;
+      }
+      if (ch === '`') {
+        sawConstruct = true;
+        j = skipBacktick(this.source, this.budget, j, st.end);
+        continue;
+      }
+      if (ch === '$') {
+        sawConstruct = true;
+        j = skipDollar(this.source, this.budget, j, st.end);
+        continue;
+      }
+      if ((ch === '?' || ch === '*' || ch === '+' || ch === '@' || ch === '!') && this.source[j + 1] === '(') {
+        sawGroup = true;
+      }
+      if (ch === '(') {
+        depth++;
+        j++;
+        continue;
+      }
+      if (ch === ')') {
+        if (depth > 0) {
+          depth--;
+          j++;
+          continue;
+        }
+        break;
+      }
+      if (depth === 0 && (ch === ' ' || ch === '\t' || ch === '\r' || ch === '\n')) break;
+      j++;
+    }
+    const hasGroup = sawGroup;
+    const hasConstruct = sawConstruct;
+    // A pattern containing an extglob group but no construct: one
+    // extglob_pattern node when the scanner accepts it; rejected groups
+    // keep the normal (error) path.
+    if (hasGroup && !hasConstruct) {
+      if (!this.extglobGroupAccepted(i, j)) return null;
+      st.pos = j;
+      st.lookahead = null;
+      return { frames: [this.frame('extglob_pattern', i, j)], end: j };
+    }
+    if (!hasConstruct) return null;
+    // Glob text (possibly with a group) mixed with one quote/expansion:
+    // extglob_pattern pieces around the construct (`@(*${x}.tar|*.sig)` →
+    // extglob "@(*", expansion, extglob ".tar|*.sig)").
+    const pieces = this.parseExtglobBlob(i, j, hasGroup ? (s, e) => this.extglobGroupAccepted(s, e) : undefined);
+    if (pieces === null) return null;
+    st.pos = j;
+    st.lookahead = null;
+    return { frames: pieces, end: j };
+  }
+
+  /** Reference-shaped right-hand sides for test comparisons:
+   *  - after `=`, a bare word is a regex when it contains a glob character
+   *    or `=` — but only outside parentheses (`[[ a = b*c ]]` → regex,
+   *    `[[ a = b=c ]]` → regex, `[[ ( $a = x* ) ]]` → word);
+   *  - after `==`/`!=`, a bare word follows the scanner.c extglob_pattern
+   *    rule — see isTestExtglob.
+   *  The right side may arrive as a concatenation of word/number fragments
+   *  (the lexer splits [ ] into their own pieces, so `[0-9]*` is not one
+   *  word); bare fragment runs are judged — and converted — as a whole. */
+  private convertTestRightSide(operator: string, right: Frame, st: ExprState): Frame {
+    if (right.type !== 'word' && !(right.type === 'concatenation' && this.isBarePieces(right))) {
+      return right;
+    }
+    const text = this.text(right.start, right.end);
+    if (operator === '=' && st.parenDepth === 0 && /[*?[\]=]/.test(text)) {
+      return this.frame('regex', right.start, right.end);
+    }
+    if ((operator === '==' || operator === '!=') && this.isTestExtglob(right)) {
+      return this.frame('extglob_pattern', right.start, right.end);
+    }
+    return right;
+  }
+
+  /** True when a concatenation consists only of word/number fragments, so
+   *  it can be treated as one bare word in test comparisons. */
+  private isBarePieces(frame: Frame): boolean {
+    return frame.children.every((child) => child.type === 'word' || child.type === 'number');
+  }
+
+  /**
+   * The extglob_pattern scan rule from tree-sitter-bash's scanner.c
+   * (the `extglob_pattern:` label), reduced to what a bare test word can
+   * reach (test words never contain blanks, quotes, `$`, `|` or parens):
+   *   - the first character must be a letter or one of `?*+@!-)\.[`;
+   *   - a leading `-letter` is just a word (`[[ $a == -foo ]]` → word);
+   *   - a single character becomes a glob only when directly followed by
+   *     whitespace (`[[ $a == b ]]` → glob, `[[ (a == b) ]]` → word);
+   *   - a `-letters` second run followed by `)`, `\` or `.` stays a word;
+   *   - otherwise the second character must be alphanumeric or one of
+   *     `[?/\_*` (skipped when the word starts with `[`), and the word is
+   *     a glob iff any character is neither a letter nor `.` — escapes
+   *     count only the character after the backslash when it is not a
+   *     space/quote (`foo\ bar` → word, `b\*c` → glob).
+   */
+  private isTestExtglob(right: Frame): boolean {
+    const text = this.text(right.start, right.end);
+    const nextChar = this.source[right.end];
+    const isSpace = (ch: string | undefined): boolean => ch === ' ' || ch === '\t' || ch === '\r';
+    const isAlpha = (ch: string): boolean => /[A-Za-z]/.test(ch);
+    const isAlnum = (ch: string): boolean => /[A-Za-z0-9]/.test(ch);
+    const c0 = text[0]!;
+    if (!isAlpha(c0) && !'?*+@!-)\\.['.includes(c0)) return false;
+    let sawNonAlphaDot = !isAlpha(c0);
+    let i = 1;
+    if (i >= text.length) {
+      if (isSpace(nextChar) || nextChar === '|') return true;
+      if ((c0 === ')' || c0 === '*') && nextChar === ')' && isSpace(this.source[right.end + 1])) {
+        return sawNonAlphaDot;
+      }
+      return false;
+    }
+    // "-\w" is just a word after ==/!= in the reference.
+    if (c0 === '-' && isAlpha(text[i]!)) return false;
+    if (text[i] === '-') {
+      // "-\w" in second position: just a word, unless something special
+      // follows the run.
+      i++;
+      while (i < text.length && isAlnum(text[i]!)) i++;
+      const after = i < text.length ? text[i]! : nextChar;
+      if (after === ')' || after === '\\' || after === '.') return false;
+      if (i >= text.length) return true;
+    }
+    // The second-character check is skipped for a leading `[` (the scanner
+    // does not advance past it before checking).
+    if (c0 !== '[' && !isAlnum(text[i]!) && !'[?/\\_*'.includes(text[i]!)) return false;
+    for (; i < text.length; i++) {
+      const ch = text[i]!;
+      if (ch === '\\') {
+        // A backslash never counts itself; an escaped space/quote is
+        // consumed silently (anything else is judged on its own).
+        const after = text[i + 1];
+        if (after === ' ' || after === '\t' || after === '\r' || after === '"') i++;
+        continue;
+      }
+      if (!isAlpha(ch) && ch !== '.') sawNonAlphaDot = true;
+    }
+    return sawNonAlphaDot;
+  }
+
+  /** Character-level tokenizer for the expression engine over [st.pos,
+   *  st.end). Produces one token per call and advances st.pos. */
+  private scanExprToken(st: ExprState): ExprToken {
+    const end = st.end;
+    let i = st.pos;
+    let sinceTick = 0;
+    for (;;) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      if (i >= end) break;
+      const ch = this.source[i]!;
+      if (ch === ' ' || ch === '\t' || ch === '\r' || ch === '\n') {
+        i++;
+        continue;
+      }
+      if (ch === '\\' && this.source[i + 1] === '\n') {
+        i += 2; // line continuation
+        continue;
+      }
+      break;
+    }
+    st.pos = i;
+    if (i >= end) return { kind: 'end', start: i, end: i, text: '' };
+    const ch = this.source[i]!;
+    if (st.mode === 'test') return this.scanTestToken(st, i, ch);
+
+    // Arithmetic / c-style-for tokens.
+    if (ch >= '0' && ch <= '9') {
+      let j = i;
+      if (ch === '0' && (this.source[i + 1] === 'x' || this.source[i + 1] === 'X')) {
+        j = i + 2;
+        while (j < end && /[0-9a-fA-F]/.test(this.source[j]!)) j++;
+      } else {
+        while (j < end && this.source[j]! >= '0' && this.source[j]! <= '9') j++;
+        if (this.source[j] === '#') {
+          j++;
+          while (j < end && /[0-9A-Za-z@_]/.test(this.source[j]!)) j++;
+        }
+      }
+      st.pos = j;
+      return { kind: 'number', start: i, end: j, text: this.text(i, j) };
+    }
+    if (/[A-Za-z_]/.test(ch)) {
+      let j = i + 1;
+      while (j < end && /\w/.test(this.source[j]!)) j++;
+      if (this.source[j] === '[' && j < end) {
+        // Subscript: name[index] — the index is a literal.
+        const sub = this.scanBalanced(j, end, '[', ']');
+        if (!sub.balanced) this.hasError = true;
+        const indexEnd = sub.balanced ? sub.end - 1 : sub.end;
+        const kids: Frame[] = [this.frame('variable_name', i, j), this.anon('[', j, j + 1)];
+        if (indexEnd > j + 1) {
+          kids.push(this.parseLiteral(j + 1, indexEnd));
+        } else {
+          this.hasError = true; // empty subscript index
+        }
+        if (sub.balanced) kids.push(this.anon(']', sub.end - 1, sub.end));
+        st.pos = sub.end;
+        return { kind: 'subst', start: i, end: sub.end, text: this.text(i, sub.end), frame: this.frame('subscript', i, sub.end, kids) };
+      }
+      st.pos = j;
+      return { kind: 'ident', start: i, end: j, text: this.text(i, j) };
+    }
+    if (ch === '$') {
+      const dollar = this.parseDollar(i, end);
+      if (dollar !== null) {
+        st.pos = dollar[1];
+        return { kind: 'subst', start: i, end: dollar[1], text: this.text(i, dollar[1]), frame: dollar[0] };
+      }
+      this.hasError = true; // bare $ inside arithmetic
+      st.pos = i + 1;
+      return { kind: 'subst', start: i, end: i + 1, text: '$', frame: this.frame('ERROR', i, i + 1) };
+    }
+    if (ch === '"') {
+      const [piece, next] = this.parseString(i, end);
+      st.pos = next;
+      return { kind: 'string', start: i, end: next, text: this.text(i, next), frame: piece };
+    }
+    if (ch === '`') {
+      const [piece, next] = this.parseBacktickSubstitution(i, end);
+      st.pos = next;
+      return { kind: 'subst', start: i, end: next, text: this.text(i, next), frame: piece };
+    }
+    if (ch === '(') {
+      st.pos = i + 1;
+      return { kind: 'lparen', start: i, end: i + 1, text: '(' };
+    }
+    if (ch === ')') {
+      st.pos = i + 1;
+      return { kind: 'rparen', start: i, end: i + 1, text: ')' };
+    }
+    for (const operator of EXPRESSION_OPERATORS) {
+      if (i + operator.length <= end && this.source.startsWith(operator, i)) {
+        st.pos = i + operator.length;
+        return { kind: 'op', start: i, end: st.pos, text: operator };
+      }
+    }
+    // Unknown character: recover one char at a time.
+    st.pos = i + 1;
+    return { kind: 'unknown', start: i, end: i + 1, text: ch };
+  }
+
+  /** Tokenizer for test-command expressions ([[ … ]] / [ … ]). Words stop
+   *  at whitespace and ( ) < > & | only — `=` and `!` are word characters
+   *  unless they start a token (`a=b` is one word in the reference, while
+   *  spaced `=`/`==`/`!=`/`=~` are operators). */
+  private scanTestToken(st: ExprState, i: number, ch: string): ExprToken {
+    const end = st.end;
+    if (ch === '(') {
+      // `((…))` inside a test is an arithmetic_expansion when the balanced
+      // region closes with `))` (`[[ ((a)) == x ]]` in the reference);
+      // otherwise the parens nest as parenthesized_expressions.
+      if (this.source[i + 1] === '(') {
+        const scan = this.scanBalanced(i, end, '(', ')');
+        if (scan.balanced && scan.end - 2 > i + 1 && this.source[scan.end - 2] === ')') {
+          const [piece, next] = this.parseParenArithmetic(i, end);
+          st.pos = next;
+          return { kind: 'subst', start: i, end: next, text: this.text(i, next), frame: piece };
+        }
+      }
+      st.pos = i + 1;
+      return { kind: 'lparen', start: i, end: i + 1, text: '(' };
+    }
+    if (ch === ')') {
+      st.pos = i + 1;
+      return { kind: 'rparen', start: i, end: i + 1, text: ')' };
+    }
+    if (ch === '&' && this.source[i + 1] === '&') {
+      st.pos = i + 2;
+      return { kind: 'op', start: i, end: i + 2, text: '&&' };
+    }
+    if (ch === '|' && this.source[i + 1] === '|') {
+      st.pos = i + 2;
+      return { kind: 'op', start: i, end: i + 2, text: '||' };
+    }
+    if (ch === '<' || ch === '>') {
+      const wide = this.source[i + 1] === '=' ? 2 : 1;
+      st.pos = i + wide;
+      return { kind: 'op', start: i, end: i + wide, text: this.text(i, i + wide) };
+    }
+    if (ch === '!' || ch === '=') {
+      if (st.expectOperator) {
+        // After an operand these are comparison operators even when the
+        // next word is attached (`=~^x`, `==b`, `=b` all split).
+        const two = this.source.slice(i, i + 2);
+        const wide = two === '==' || two === '=~' || two === '!=' ? 2 : 1;
+        st.pos = i + wide;
+        return { kind: 'op', start: i, end: i + wide, text: this.text(i, i + wide) };
+      }
+      // At operand position only a standalone `!` is the prefix operator;
+      // everything else (=b, !x, !=b) is word material in the reference.
+      const after = this.source[i + 1];
+      if (ch === '!' && (after === undefined || after === ' ' || after === '\t' || after === '\r' || i + 1 >= end)) {
+        st.pos = i + 1;
+        return { kind: 'op', start: i, end: i + 1, text: '!' };
+      }
+      // Fall through to the word region below.
+    }
+    if (ch === '-' && /[A-Za-z]/.test(this.source[i + 1] ?? '')) {
+      // test_operator: -letters followed by whitespace AND by a real
+      // operand (scanner.c's rule: -\w followed by an operator, the closer
+      // or nothing is just a word — `[[ -foo == x ]]`, `[[ $a == -foo ]]`,
+      // `[[ -f ]]` all keep `-foo`/`-f` as words in the reference).
+      let j = i + 1;
+      while (j < end && /[A-Za-z]/.test(this.source[j]!)) j++;
+      const after = this.source[j];
+      if (j < end && (after === ' ' || after === '\t' || after === '\r')) {
+        let k = j;
+        while (k < end && (this.source[k] === ' ' || this.source[k] === '\t' || this.source[k] === '\r')) k++;
+        const next = this.source[k];
+        if (k < end && next !== '=' && next !== ']' && next !== '&' && next !== '|' && next !== ')') {
+          st.pos = j;
+          return { kind: 'testop', start: i, end: j, text: this.text(i, j) };
+        }
+      }
+    }
+    // A word region: quote- and substitution-aware, stopping at blanks and
+    // the operator characters above.
+    let j = i;
+    let sinceTick = 0;
+    while (j < end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const c = this.source[j]!;
+      if (c === ' ' || c === '\t' || c === '\r' || c === '\n') break;
+      if (c === '(' || c === ')' || c === '<' || c === '>' || c === '&' || c === '|') break;
+      if (c === '"') {
+        j = skipDoubleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (c === "'") {
+        j = skipSingleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (c === '`') {
+        j = skipBacktick(this.source, this.budget, j, end);
+        continue;
+      }
+      if (c === '$') {
+        j = this.skipDollarConstruct(j, end);
+        continue;
+      }
+      if (c === '\\') {
+        j += 2;
+        continue;
+      }
+      j++;
+    }
+    if (j === i) j++; // defensive: never stall
+    st.pos = j;
+    // A lone arithmetic operator character is an operator, not a word.
+    if (j === i + 1 && (ch === '+' || ch === '*' || ch === '/' || ch === '%')) {
+      return { kind: 'op', start: i, end: j, text: ch };
+    }
+    return { kind: 'word', start: i, end: j, text: this.text(i, j) };
+  }
+
+  // ------------------------------------------------------------ test command
+
+  /**
+   * test_command in the `(( expression ))` form — only reached for
+   * statement-position `((word++…))` / `((word--…))` (see PAREN_TEST_RE);
+   * every other `((…))` is an arithmetic_expansion. The header arrived as
+   * one word token (see the lexer), so no repositioning is needed.
+   */
+  private parseParenTestCommand(): Frame {
+    const token = this.lexer.next();
+    const closed =
+      token.end - token.start >= 4 && this.source[token.end - 2] === ')' && this.source[token.end - 1] === ')';
+    if (!closed) this.hasError = true;
+    const kids: Frame[] = [this.anon('((', token.start, token.start + 2)];
+    const innerEnd = closed ? token.end - 2 : token.end;
+    if (innerEnd > token.start + 2) {
+      const st = this.newExprState(token.start + 2, innerEnd, 'test');
+      const expression = this.parseExpression(st, 0);
+      if (expression !== null) kids.push(expression);
+      const leftover = this.exprLeftover(st);
+      if (leftover !== null) kids.push(leftover);
+    }
+    if (closed) {
+      kids.push(this.anon('))', token.end - 2, token.end));
+    }
+    return this.frame('test_command', token.start, this.endOf(kids, token.end), kids);
+  }
+
+  /**
+   * test_command: `[[ expression ]]` or `[ expression ]`. The peeked token
+   * is the single-character word `[`; `[[` is detected by the adjacent
+   * second `[` in the source. The expression range is scanned
+   * character-wise (quote/substitution aware) up to the closer or the end
+   * of the line, parsed by the expression engine in test mode, and the main
+   * lexer is then repositioned past the closer.
+   */
+  private parseTestCommand(): Frame {
+    const openToken = this.lexer.next();
+    const double = this.source[openToken.end] === '[';
+    const opener = double ? '[[' : '[';
+    const closer = double ? ']]' : ']';
+    const exprStart = openToken.end + (double ? 1 : 0);
+    const kids: Frame[] = [this.anon(opener, openToken.start, exprStart)];
+    const scan = this.scanTestCloser(exprStart, double);
+    // A single-bracket test whose body contains a top-level redirect is a
+    // redirected statement in the reference, not a test expression
+    // (`[ ! command -v go &>/dev/null ]` → redirected_statement with a
+    // negated_command inside the test_command).
+    if (!double && scan.closerStart > exprStart && this.rangeHasTopLevelRedirect(exprStart, scan.closerStart)) {
+      const statements = this.parseScopedStatements(exprStart, scan.closerStart);
+      kids.push(...statements);
+      let end = scan.closerStart;
+      if (scan.found) {
+        kids.push(this.anon(closer, scan.closerStart, scan.afterCloser));
+        end = scan.afterCloser;
+      } else {
+        this.hasError = true;
+      }
+      this.lexer.reposition(end);
+      return this.frame('test_command', openToken.start, end, kids);
+    }
+    if (scan.closerStart > exprStart) {
+      const st = this.newExprState(exprStart, scan.closerStart, 'test');
+      const expression = this.parseExpression(st, 0);
+      if (expression !== null) kids.push(expression);
+      const leftover = this.exprLeftover(st);
+      if (leftover !== null) kids.push(leftover);
+    }
+    let end = scan.closerStart;
+    if (scan.found) {
+      if (double && kids.length === 1) {
+        // `[[ ]]` (empty expression): the reference parses the closer text
+        // as two word pieces inside a concatenation and inserts a
+        // zero-width `]]` after them.
+        this.hasError = true;
+        kids.push(
+          this.frame('concatenation', scan.closerStart, scan.afterCloser, [
+            this.frame('word', scan.closerStart, scan.closerStart + 1),
+            this.frame('word', scan.closerStart + 1, scan.afterCloser),
+          ]),
+        );
+        kids.push(this.anon(closer, scan.afterCloser, scan.afterCloser));
+      } else {
+        kids.push(this.anon(closer, scan.closerStart, scan.afterCloser));
+      }
+      end = scan.afterCloser;
+    } else {
+      // Unterminated: keep a zero-width closer and flag the error (our
+      // documented recovery policy); trailing blanks stay outside the node,
+      // as in the reference. The reference inserts a zero-width closer only
+      // for an unterminated `[` or for `[[` ending in a complete unary
+      // test — other unterminated `[[` forms become an ERROR node there.
+      this.hasError = true;
+      let closeAt = scan.closerStart;
+      while (closeAt > exprStart && (this.source[closeAt - 1] === ' ' || this.source[closeAt - 1] === '\t' || this.source[closeAt - 1] === '\r')) {
+        closeAt--;
+      }
+      kids.push(this.anon(closer, closeAt, closeAt));
+      end = closeAt;
+    }
+    this.lexer.reposition(end);
+    return this.frame('test_command', openToken.start, end, kids);
+  }
+
+  /** Whether [start, end) contains a top-level redirect operator (`<`/`>`
+   *  outside quotes, substitutions and `<(`/`>(`). Used to recognize the
+   *  redirected-statement form of a single-bracket test command. */
+  private rangeHasTopLevelRedirect(start: number, end: number): boolean {
+    let j = start;
+    let sinceTick = 0;
+    while (j < end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[j]!;
+      if (ch === '\\') {
+        j += 2;
+        continue;
+      }
+      if (ch === '"') {
+        j = skipDoubleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === "'") {
+        j = skipSingleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === '`') {
+        j = skipBacktick(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === '$') {
+        j = this.skipDollarConstruct(j, end);
+        continue;
+      }
+      if ((ch === '<' || ch === '>') && this.source[j + 1] !== '(') return true;
+      j++;
+    }
+    return false;
+  }
+
+  /** Find the closer of a test command, skipping quotes and substitutions.
+   *  Bounded by the end of the line and the lexer's range. */
+  private scanTestCloser(start: number, double: boolean): { closerStart: number; afterCloser: number; found: boolean } {
+    const end = this.lexer.rangeEnd;
+    let j = start;
+    let sinceTick = 0;
+    while (j < end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[j]!;
+      if (ch === '\n') break;
+      if (ch === '\\') {
+        j += 2;
+        continue;
+      }
+      if (ch === '"') {
+        j = skipDoubleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === "'") {
+        j = skipSingleQuoted(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === '`') {
+        j = skipBacktick(this.source, this.budget, j, end);
+        continue;
+      }
+      if (ch === '$') {
+        j = this.skipDollarConstruct(j, end);
+        continue;
+      }
+      if ((ch === '<' || ch === '>') && this.source[j + 1] === '(') {
+        j = this.scanBalanced(j + 1, end, '(', ')').end;
+        continue;
+      }
+      if (ch === ']') {
+        if (double) {
+          if (this.source[j + 1] === ']') return { closerStart: j, afterCloser: j + 2, found: true };
+        } else {
+          return { closerStart: j, afterCloser: j + 1, found: true };
+        }
+      }
+      j++;
+    }
+    return { closerStart: j, afterCloser: j, found: false };
+  }
+
+  /** heredoc_body content: heredoc_content chunks interleaved with
+   *  expansions and command substitutions (unquoted delimiters only).
+   *  Unlike tree-sitter-bash's scanner, every plain chunk — including the
+   *  leading one — becomes a heredoc_content node. */
+  private parseHeredocContent(start: number, end: number): Frame[] {
+    const pieces: Frame[] = [];
+    let chunkStart = start;
+    let i = start;
+    const flush = (upto: number): void => {
+      if (upto > chunkStart) {
+        pieces.push(this.frame('heredoc_content', chunkStart, upto));
+      }
+    };
+    let sinceTick = 0;
+    while (i < end) {
+      if (++sinceTick >= SCAN_TICK_INTERVAL) {
+        this.budget.progress();
+        sinceTick = 0;
+      }
+      const ch = this.source[i]!;
+      if (ch === '\\') {
+        i += 2; // escaped characters stay literal (and suppress expansion)
+        continue;
+      }
+      if (ch === '$') {
+        // `$'` is NOT an ANSI-C string inside a heredoc body — bash does not
+        // expand those there and the reference keeps the text as plain
+        // content too (except for its line-start quirk, see README).
+        if (this.source[i + 1] === "'") {
+          i++;
+          continue;
+        }
+        const dollar = this.parseDollar(i, end);
+        if (dollar !== null) {
+          flush(i);
+          pieces.push(dollar[0]);
+          i = dollar[1];
+          chunkStart = i;
+          continue;
+        }
+        i++;
+        continue;
+      }
+      if (ch === '`') {
+        flush(i);
+        const [piece, next] = this.parseBacktickSubstitution(i, end);
+        pieces.push(piece);
+        i = next;
+        chunkStart = i;
+        continue;
+      }
+      i++;
+    }
+    flush(end);
+    return pieces;
+  }
+
+  /** Balanced-scan wrapper returning whether the close was actually found. */
+  private scanBalanced(i: number, end: number, open: string, close: string): BalancedScan {
+    return scanBalanced(this.source, this.budget, i, end, open, close);
+  }
+}
+
+/**
+ * Convert a finished frame tree into SyntaxNodeBuilders. Iterative
+ * (explicit stack) so deep trees cannot overflow the call stack; node
+ * creation here is not budget-ticked because every frame already ticked the
+ * budget when it was created (one frame ↔ one node).
+ */
+export function materialize(root: Frame, source: string): SyntaxNodeBuilder {
+  const nodes = new Map<Frame, SyntaxNodeBuilder>();
+  const order: Frame[] = [];
+  const stack: Frame[] = [root];
+  while (stack.length > 0) {
+    const frame = stack.pop()!;
+    order.push(frame);
+    for (const child of frame.children) stack.push(child);
+  }
+  for (const frame of order) {
+    nodes.set(
+      frame,
+      new SyntaxNodeBuilder({
+        type: frame.type,
+        source,
+        startIndex: frame.start,
+        endIndex: frame.end,
+        isNamed: frame.isNamed,
+      }),
+    );
+  }
+  for (const frame of order) {
+    const node = nodes.get(frame)!;
+    for (const child of frame.children) {
+      node.addChild(nodes.get(child)!);
+    }
+  }
+  return nodes.get(root)!;
+}
diff --git a/packages/tree-sitter-bash/test/differential.test.ts b/packages/tree-sitter-bash/test/differential.test.ts
new file mode 100644
index 0000000000..0c185e5125
--- /dev/null
+++ b/packages/tree-sitter-bash/test/differential.test.ts
@@ -0,0 +1,169 @@
+// test/differential.test.ts
+//
+// Differential suite: every sample in test/fixtures/differential/ and every
+// official tree-sitter-bash 0.25.0 corpus case in test/fixtures/corpus/ is
+// parsed by BOTH this package and the reference (web-tree-sitter + the
+// official wasm build) and the normalized trees are compared byte-for-byte.
+//
+//   - @match samples must produce a tree identical to the reference;
+//   - @known-diff samples must (a) still differ from the reference and
+//     (b) match the stored dump of our parser exactly — so a documented
+//     deviation cannot silently drift, and a fixed deviation fails loudly
+//     ("remove it from the known list").
+//
+// The known-difference registry (test/helpers/known-differences.ts) is the
+// single source of truth and is cross-checked against README.md's Known
+// differences section here.
+
+import { readFileSync, readdirSync } from 'node:fs';
+import path from 'node:path';
+
+import { beforeAll, describe, expect, it } from 'vitest';
+
+import { compareSource, diffDumps, ourDump, parseCorpusFile, parseFixtureFile, referenceDump } from './helpers/differential';
+import type { FixtureSample } from './helpers/differential';
+import { KNOWN_DIFFERENCES, isKnownDifferenceId } from './helpers/known-differences';
+
+const PKG_ROOT = path.resolve(import.meta.dirname, '..');
+const DIFF_DIR = path.join(PKG_ROOT, 'test/fixtures/differential');
+const CORPUS_DIR = path.join(PKG_ROOT, 'test/fixtures/corpus');
+
+interface LocatedSample extends FixtureSample {
+  file: string;
+}
+
+function loadDifferentialFixtures(): LocatedSample[] {
+  const out: LocatedSample[] = [];
+  for (const file of readdirSync(DIFF_DIR).filter((f) => f.endsWith('.txt')).toSorted()) {
+    const content = readFileSync(path.join(DIFF_DIR, file), 'utf8');
+    for (const sample of parseFixtureFile(`${DIFF_DIR}/${file}`, content)) {
+      out.push({ ...sample, file });
+    }
+  }
+  return out;
+}
+
+function loadCorpusKnownDiffs(): Map<string, LocatedSample> {
+  const file = 'known-diffs.txt';
+  const content = readFileSync(path.join(CORPUS_DIR, file), 'utf8');
+  const map = new Map<string, LocatedSample>();
+  for (const sample of parseFixtureFile(`${CORPUS_DIR}/${file}`, content)) {
+    // description: "corpus <file>::<case name>"
+    const key = sample.description.replace(/^corpus /, '');
+    map.set(key, { ...sample, file });
+  }
+  return map;
+}
+
+function label(sample: LocatedSample): string {
+  return `${sample.file}:${sample.line} ${sample.kind === 'match' ? 'match' : `known-diff ${sample.id}`} — ${sample.description.slice(0, 60)}`;
+}
+
+const fixtures = loadDifferentialFixtures();
+const corpusKnownDiffs = loadCorpusKnownDiffs();
+const corpusCases = readdirSync(CORPUS_DIR)
+  .filter((f) => f.endsWith('.txt') && f !== 'known-diffs.txt')
+  .toSorted()
+  .flatMap((file) => parseCorpusFile(readFileSync(path.join(CORPUS_DIR, file), 'utf8')).map((c) => ({ ...c, file })));
+
+beforeAll(async () => {
+  // Fails with a descriptive error when the wasm reference cannot be loaded.
+  await compareSource('echo sanity');
+});
+
+describe('differential fixtures', () => {
+  for (const sample of fixtures) registerFixtureTest(sample);
+});
+
+/** Register one fixture test (lookup table instead of conditionals — the
+ *  vitest lint rules reject conditionals around `it`). */
+function registerFixtureTest(sample: LocatedSample): void {
+  const register: Record<FixtureSample['kind'], () => void> = {
+    match: () => {
+      it(label(sample), async () => {
+        const cmp = await compareSource(sample.source);
+        expect(cmp.diff).toBe('');
+      });
+    },
+    'known-diff': () => {
+      it(label(sample), async () => {
+        const [ours, reference] = [ourDump(sample.source), await referenceDump(sample.source)];
+        expect(
+          ours !== reference,
+          'this sample now matches the reference — remove it from the known-difference list (fixture + registry + README)',
+        ).toBe(true);
+        expect(ours === sample.expectedOurs, `deviation shape drifted:\n${diffDumps(ours, sample.expectedOurs!)}`).toBe(
+          true,
+        );
+      });
+    },
+  };
+  register[sample.kind]();
+}
+
+describe('official tree-sitter-bash 0.25.0 corpus', () => {
+  for (const corpusCase of corpusCases) registerCorpusTest(corpusCase);
+});
+
+/** Register one corpus case test. */
+function registerCorpusTest(corpusCase: { file: string; name: string; input: string }): void {
+  const key = `${corpusCase.file}::${corpusCase.name}`;
+  const known = corpusKnownDiffs.get(key) ?? null;
+  const register: Record<'match' | 'known-diff', () => void> = {
+    match: () => {
+      it(`${key} (match)`, async () => {
+        const cmp = await compareSource(corpusCase.input);
+        expect(cmp.diff).toBe('');
+      });
+    },
+    'known-diff': () => {
+      it(`${key} (known-diff ${known!.id})`, async () => {
+        const [ours, reference] = [ourDump(corpusCase.input), await referenceDump(corpusCase.input)];
+        expect(
+          ours !== reference,
+          'this corpus case now matches the reference — remove it from known-diffs.txt, the registry and the README',
+        ).toBe(true);
+        expect(ours === known!.expectedOurs, `deviation shape drifted:\n${diffDumps(ours, known!.expectedOurs!)}`).toBe(
+          true,
+        );
+      });
+    },
+  };
+  register[known === null ? 'match' : 'known-diff']();
+}
+
+describe('known-difference registry consistency', () => {
+  const usedIds = new Set<string>();
+  for (const sample of fixtures) {
+    if (sample.id !== undefined) usedIds.add(sample.id);
+  }
+  for (const sample of corpusKnownDiffs.values()) {
+    if (sample.id !== undefined) usedIds.add(sample.id);
+  }
+
+  it('every @known-diff id used in fixtures exists in the registry', () => {
+    for (const id of usedIds) {
+      expect(isKnownDifferenceId(id), `fixture uses unregistered id '${id}'`).toBe(true);
+    }
+  });
+
+  it('every registry id is exercised by at least one fixture or corpus case', () => {
+    for (const entry of KNOWN_DIFFERENCES) {
+      expect(usedIds.has(entry.id), `registry id '${entry.id}' has no fixture sample`).toBe(true);
+    }
+  });
+
+  it('every registry entry is anchored in README.md Known differences', () => {
+    const readme = readFileSync(path.join(PKG_ROOT, 'README.md'), 'utf8');
+    const sectionMatch = /## Known differences[^\n]*\n([\s\S]*?)(?=\n## |\s*$)/.exec(readme);
+    expect(sectionMatch, 'README.md has no "## Known differences" section').not.toBeNull();
+    // Whitespace-normalized so anchors may span line breaks in the README.
+    const section = sectionMatch![1]!.replaceAll(/\s+/g, ' ');
+    for (const entry of KNOWN_DIFFERENCES) {
+      expect(
+        section.includes(entry.readmeAnchor.replaceAll(/\s+/g, ' ')),
+        `README.md Known differences does not contain the anchor for '${entry.id}': ${entry.readmeAnchor}`,
+      ).toBe(true);
+    }
+  });
+});
diff --git a/packages/tree-sitter-bash/test/fixtures/README.md b/packages/tree-sitter-bash/test/fixtures/README.md
new file mode 100644
index 0000000000..bcb4403b06
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/README.md
@@ -0,0 +1,42 @@
+# Differential fixtures
+
+Fixtures for the differential test suite (`test/differential.test.ts`):
+every sample is parsed by this package and by the reference tree-sitter-bash
+0.25.0 parser (web-tree-sitter + wasm), and the normalized trees are
+compared.
+
+## Format
+
+Blocks are separated by a line containing exactly `===`. Each block starts
+with a directive line:
+
+```
+@match: <description>
+<source>
+===
+@known-diff <registry-id>: <description>
+<source>
+---
+<expected normalized dump of THIS package's parser>
+===
+```
+
+- `@match`: the two trees must be identical.
+- `@known-diff <id>`: the sample documents a known deviation. `<id>` must
+  exist in `test/helpers/known-differences.ts` (the registry is
+  cross-checked against the README's Known differences section). The block
+  must still deviate from the reference, and our tree must match the stored
+  dump exactly — so deviations cannot silently drift, and a fixed deviation
+  fails the test with a reminder to remove it from the list.
+
+`differential/` holds curated samples grouped by theme (statements,
+redirects, heredoc, expansions, test-command, arithmetic, case, recovery).
+`corpus/` holds the official tree-sitter-bash corpus (see its README);
+corpus cases that deviate are listed in `corpus/known-diffs.txt` in the same
+format.
+
+The normalized dump is a pre-order walk of every node (named and
+anonymous): one line per node — `type [start,end] "text preview"`,
+anonymous types in parentheses — preceded by a `hasError:` line. Offsets
+are UTF-16 code units on both sides (web-tree-sitter reports UTF-16
+offsets for string input, same as this parser).
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/README.md b/packages/tree-sitter-bash/test/fixtures/corpus/README.md
new file mode 100644
index 0000000000..12524df4ed
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/README.md
@@ -0,0 +1,21 @@
+# Official tree-sitter-bash corpus
+
+The five `*.txt` files in this directory are an unmodified copy of the
+official tree-sitter-bash test corpus:
+
+- Source: https://github.com/tree-sitter/tree-sitter-bash/tree/v0.25.0/test/corpus
+- Version: tag `v0.25.0`
+- Retrieved: 2026-07-21 (via the GitHub release tarball; only `test/corpus/`
+  was taken)
+
+They are inputs to `test/differential.test.ts`: every corpus case is parsed
+by both this package and the live reference parser (web-tree-sitter +
+tree-sitter-bash.wasm v0.25.0) and the trees are compared byte-for-byte.
+The expected S-expressions inside the corpus files are not used — the live
+wasm build is the comparison target.
+
+`known-diffs.txt` lists the corpus cases whose trees deliberately deviate
+(see the "Known differences" section of the package README), in the fixture
+format described in `test/fixtures/README.md`. A corpus case listed there
+must still deviate from the reference AND keep the exact stored deviation
+shape; every other corpus case must match the reference byte-for-byte.
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/commands.txt b/packages/tree-sitter-bash/test/fixtures/corpus/commands.txt
new file mode 100644
index 0000000000..63e95ce2d9
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/commands.txt
@@ -0,0 +1,725 @@
+===============================
+Commands
+===============================
+
+whoami
+
+---
+
+(program
+  (command (command_name (word))))
+
+===============================
+Commands with arguments
+===============================
+
+cat file1.txt
+git diff --word-diff=color -- file1.txt file2.txt
+echo $sing\
+levar
+
+---
+
+(program
+  (command (command_name (word)) (word))
+  (command (command_name (word)) (word) (word) (word) (word) (word))
+  (command (command_name (word)) (simple_expansion (variable_name)) (word)))
+
+===============================
+Quoted command names
+===============================
+
+"$a/$b" c
+
+---
+
+(program
+  (command
+    (command_name (string (simple_expansion (variable_name)) (string_content) (simple_expansion (variable_name))))
+    (word)))
+
+===============================
+Commands with numeric arguments
+===============================
+
+exit 1
+
+---
+
+(program
+  (command (command_name (word)) (number)))
+
+===================================
+Commands with environment variables
+===================================
+
+VAR1=1 ./script/test
+VAR1=a VAR2="ok" git diff --word-diff=color
+
+---
+
+(program
+  (command
+    (variable_assignment (variable_name) (number))
+    (command_name (word)))
+  (command
+    (variable_assignment (variable_name) (word))
+    (variable_assignment (variable_name) (string (string_content)))
+    (command_name (word))
+    (word)
+    (word)))
+
+===================================
+Empty environment variables
+===================================
+
+VAR1=
+VAR2= echo
+
+---
+
+(program
+  (variable_assignment (variable_name))
+  (command (variable_assignment (variable_name)) (command_name (word))))
+
+===============================
+File redirects
+===============================
+
+whoami > /dev/null
+cat a b > /dev/null
+2>&1 whoami
+echo "foobar" >&2
+[ ! command -v go &>/dev/null ] && return
+
+if [ ]; then
+	>aa >bb
+fi
+
+exec {VIRTWL[0]} {VIRTWL[1]} <&- >&-
+exec {VIRTWL[0]}<&- {VIRTWL[1]}>&-
+
+grep 2>/dev/null -q "^/usr/bin/scponly$" /etc/shells
+
+x <x a b c
+
+---
+
+(program
+  (redirected_statement
+    (command (command_name (word)))
+    (file_redirect (word)))
+  (redirected_statement
+    (command (command_name (word)) (word) (word))
+    (file_redirect (word)))
+  (command
+    (file_redirect (file_descriptor) (number))
+    (command_name (word)))
+  (redirected_statement
+    (command (command_name (word)) (string (string_content)))
+    (file_redirect (number)))
+  (list
+    (test_command
+      (redirected_statement
+        (negated_command
+          (command (command_name (word)) (word) (word)))
+        (file_redirect (word))))
+    (command (command_name (word))))
+  (if_statement
+    (test_command)
+    (redirected_statement
+      (file_redirect (word))
+      (file_redirect (word))))
+      (redirected_statement
+    (command
+      (command_name (word))
+      (concatenation (word) (word) (word) (number) (word) (word))
+      (concatenation (word) (word) (word) (number) (word) (word)))
+    (file_redirect)
+    (file_redirect))
+  (redirected_statement
+    (command (command_name (word))
+      (concatenation (word) (word) (word) (number) (word) (word)))
+    (file_redirect
+      (concatenation
+        (word)
+        (word)
+        (word)
+        (number)
+        (word)
+        (word)))
+    (file_redirect))
+  (redirected_statement
+    (command (command_name (word)))
+    (file_redirect (file_descriptor) (word)
+      (word)
+      (string
+        (string_content))
+      (word)))
+  (redirected_statement
+    (command (command_name (word)))
+    (file_redirect (word)
+      (word)
+      (word)
+      (word))))
+
+===============================
+File redirects (noclobber override)
+===============================
+
+whoami >| /dev/null
+cat a b >| /dev/null
+
+---
+
+(program
+  (redirected_statement
+    (command (command_name (word)))
+    (file_redirect (word)))
+  (redirected_statement
+    (command (command_name (word)) (word) (word))
+    (file_redirect (word))))
+
+===============================
+Heredoc redirects
+===============================
+
+node <<JS
+console.log("hi")
+JS
+
+bash -c <<JS
+echo hi
+JS
+
+newins <<-EOF - org.freedesktop.Notifications.service
+	[D-BUS Service]
+	Name=org.freedesktop.Notifications
+	Exec=/usr/libexec/notification-daemon
+EOF
+
+---
+
+(program
+  (redirected_statement
+    (command (command_name (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body)
+      (heredoc_end)))
+  (redirected_statement
+    (command (command_name (word)) (word))
+    (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body)
+      (heredoc_end)))
+  (redirected_statement
+    (command (command_name (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (word)
+      (word)
+      (heredoc_body)
+      (heredoc_end))))
+
+===============================
+Heredocs with variables
+===============================
+
+node <<JS
+a $B ${C}
+JS
+
+exit
+
+---
+
+(program
+  (redirected_statement
+    (command
+      (command_name
+        (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body
+        (simple_expansion
+          (variable_name))
+        (heredoc_content)
+        (expansion
+          (variable_name))
+        (heredoc_content))
+      (heredoc_end)))
+  (command
+    (command_name
+      (word))))
+
+=================================
+Heredocs with file redirects
+=================================
+
+cat <<EOF > $tmpfile
+a $B ${C}
+EOF
+
+wc -l $tmpfile
+
+---
+
+(program
+  (redirected_statement
+    (command
+      (command_name
+        (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (file_redirect
+        (simple_expansion
+          (variable_name)))
+      (heredoc_body
+        (simple_expansion
+          (variable_name))
+        (heredoc_content)
+        (expansion
+          (variable_name))
+        (heredoc_content))
+      (heredoc_end)))
+  (command
+    (command_name
+      (word))
+    (word)
+    (simple_expansion
+      (variable_name))))
+
+=================================
+Heredocs with many file redirects
+=================================
+
+FOO=bar echo <<EOF 2> err.txt > hello.txt
+hello
+EOF
+
+---
+
+(program
+  (redirected_statement
+    body: (command
+      (variable_assignment
+        name: (variable_name)
+        value: (word))
+      name: (command_name
+        (word)))
+    redirect: (heredoc_redirect
+      (heredoc_start)
+      redirect: (file_redirect
+        descriptor: (file_descriptor)
+        destination: (word))
+      redirect: (file_redirect
+        destination: (word))
+      (heredoc_body)
+      (heredoc_end))))
+
+=================================
+Heredocs with pipes
+=================================
+
+one <<EOF | grep two
+three
+EOF
+
+---
+
+(program
+  (redirected_statement
+    (command
+      (command_name
+        (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (pipeline
+        (command
+          (command_name
+            (word))
+          (word)))
+      (heredoc_body)
+      (heredoc_end))))
+
+======================================
+Heredocs with escaped expansions
+======================================
+
+cat  << EOF
+DEV_NAME=\$(lsblk)
+EOF
+
+---
+
+(program (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end))))
+
+======================================
+Quoted Heredocs
+======================================
+
+cat << 'EOF'
+a=$b
+EOF
+
+cat << "EOF"
+a=$b
+EOF
+
+cat <<"END OF FILE"
+hello,
+world
+END OF FILE
+
+cat << \EOF
+EOF
+
+---
+
+(program
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end))))
+
+==========================================
+Heredocs with indented closing delimiters
+==========================================
+
+usage() {
+	cat <<-EOF
+		Usage: ${0##*/} FOO BAR
+	EOF
+}
+
+---
+
+(program
+  (function_definition
+    (word)
+    (compound_statement
+      (redirected_statement
+        (command (command_name (word)))
+        (heredoc_redirect
+          (heredoc_start)
+          (heredoc_body (expansion (special_variable_name) (regex)) (heredoc_content))
+          (heredoc_end))))))
+
+==========================================
+Heredocs with empty bodies
+==========================================
+
+node <<JS
+JS
+
+node << 'SJ'
+SJ
+
+usage() {
+	cat <<-EOF
+	EOF
+}
+
+node << 'EOF' > temp
+EOF
+
+---
+
+(program
+  (redirected_statement
+    body: (command
+      name: (command_name
+        (word)))
+    redirect: (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body)
+      (heredoc_end)))
+  (redirected_statement
+    body: (command
+      name: (command_name
+        (word)))
+    redirect: (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body)
+      (heredoc_end)))
+  (function_definition
+    name: (word)
+    body: (compound_statement
+      (redirected_statement
+        body: (command
+          name: (command_name
+            (word)))
+        redirect: (heredoc_redirect
+          (heredoc_start)
+          (heredoc_body)
+          (heredoc_end)))))
+  (redirected_statement
+    body: (command
+      name: (command_name
+        (word)))
+    redirect: (heredoc_redirect
+      (heredoc_start)
+      redirect: (file_redirect
+        destination: (word))
+      (heredoc_body)
+      (heredoc_end))))
+
+==========================================
+Heredocs with weird characters
+==========================================
+
+node <<_DELIMITER_WITH_UNDERSCORES_
+Hello.
+_DELIMITER_WITH_UNDERSCORES_
+
+node <<'```'
+Hello.
+```
+
+node <<!HEREDOC!
+Hello.
+!HEREDOC!
+
+node <<\'
+Hello.
+'
+
+node <<\\
+Hello.
+\
+
+---
+
+(program
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end)))
+  (redirected_statement (command (command_name (word))) (heredoc_redirect (heredoc_start) (heredoc_body) (heredoc_end))))
+
+==========================================
+Heredocs with a rhs statement
+==========================================
+
+cat <<-_EOF_ || die "cat EOF failed"
+	#!/bin/sh
+	echo hello
+_EOF_
+
+---
+
+(program
+  (redirected_statement
+    (command (command_name (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (command (command_name (word)) (string (string_content)))
+      (heredoc_body)
+      (heredoc_end))))
+
+==========================================
+Heredocs with a $ that is not an expansion
+==========================================
+
+cat <<EOF
+# check out this regex '^EOF$'
+EOF
+
+---
+
+(program
+  (redirected_statement
+    (command (command_name (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body)
+      (heredoc_end))))
+
+==========================================
+Nested Heredocs
+==========================================
+
+cat <<OUTER
+Outer Heredoc Start
+$(cat <<INNER
+Inner Heredoc Content
+$(cat <<INNERMOST
+Innermost Heredoc Content
+INNERMOST
+)
+INNER)
+Outer Heredoc End
+OUTER
+
+---
+
+(program
+  (redirected_statement
+    (command (command_name (word)))
+    (heredoc_redirect
+      (heredoc_start)
+      (heredoc_body
+        (command_substitution
+          (redirected_statement
+            (command (command_name (word)))
+            (heredoc_redirect
+              (heredoc_start)
+              (heredoc_body
+                (command_substitution
+                  (redirected_statement
+                    (command (command_name (word)))
+                    (heredoc_redirect
+                      (heredoc_start)
+                      (heredoc_body)
+                      (heredoc_end))))
+                (heredoc_content))
+              (heredoc_end))))
+        (heredoc_content))
+      (heredoc_end))))
+
+==========================================
+Herestrings
+==========================================
+
+node <<< foo
+
+while read -u 3 entry; do
+  echo $entry
+done 3<<<"$ENTRIES"
+
+$(tc-getCC) -Werror -Wl,-l:libobjc.so.${ver} -x objective-c \
+		- <<<$'int main() {}' -o /dev/null 2> /dev/null;
+
+<<<string cmd arg
+
+cmd arg <<<string
+
+cmd <<<string arg
+
+<<<string
+
+---
+
+(program
+  (command (command_name (word)) (herestring_redirect (word)))
+  (redirected_statement
+    (while_statement
+      (command (command_name (word)) (word) (number) (word))
+      (do_group
+        (command (command_name (word)) (simple_expansion (variable_name)))))
+    (herestring_redirect
+      (file_descriptor)
+      (string (simple_expansion (variable_name)))))
+  (redirected_statement
+    (command
+      (command_name (command_substitution (command (command_name (word)))))
+      (word)
+      (concatenation (word) (expansion (variable_name)))
+      (word)
+      (word)
+      (word)
+      (herestring_redirect (ansi_c_string))
+      (word)
+      (word))
+    (file_redirect (file_descriptor) (word)))
+  (command (herestring_redirect (word)) (command_name (word)) (word))
+  (command (command_name (word)) (word) (herestring_redirect (word)))
+  (command (command_name (word)) (herestring_redirect (word)) (word))
+  (redirected_statement (herestring_redirect (word))))
+
+==========================================
+Subscripts
+==========================================
+
+echo ${a[1 + 2]}
+
+echo ${b[1234 % 2]}
+
+${words[++counter]}
+
+${array[(($number+1))]}
+
+${array[((number+1))]}
+
+---
+
+(program
+  (command
+    (command_name (word))
+    (expansion
+      (subscript (variable_name) (binary_expression (number) (number)))))
+  (command
+    (command_name (word))
+    (expansion
+      (subscript (variable_name) (binary_expression (number) (number)))))
+  (command
+    (command_name
+      (expansion
+        (subscript (variable_name) (unary_expression (word))))))
+  (command
+    (command_name
+      (expansion
+        (subscript
+          (variable_name)
+          (arithmetic_expansion (binary_expression (simple_expansion (variable_name)) (number)))))))
+  (command
+    (command_name
+      (expansion
+        (subscript
+          (variable_name)
+          (arithmetic_expansion (binary_expression (variable_name) (number))))))))
+
+==========================================
+Bare $
+==========================================
+
+echo $
+echo "${module}"$
+echo $$
+
+---
+
+(program
+  (command (command_name (word)))
+  (command
+    (command_name (word))
+    (concatenation (string (expansion (variable_name)))))
+  (command
+    (command_name (word))
+    (simple_expansion (special_variable_name))))
+
+==========================================
+Arithmetic with command substitution
+==========================================
+
+$(( $( ver_cut 2 ) - 1 ))
+
+---
+
+(program
+  (command
+    (command_name
+      (arithmetic_expansion
+        (binary_expression
+          (command_substitution
+            (command
+              (command_name
+                (word))
+              (number)))
+          (number))))))
+
+
+==========================================
+Ralative path without dots
+==========================================
+
+bin/ls /usr/bin
+
+---
+
+(program
+  (command
+    (command_name
+      (word))
+    (word)))
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/crlf.txt b/packages/tree-sitter-bash/test/fixtures/corpus/crlf.txt
new file mode 100644
index 0000000000..874634f043
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/crlf.txt
@@ -0,0 +1,13 @@
+================================
+Variables with CRLF line endings
+================================
+
+A=one
+
+B=two
+
+---
+
+(program
+  (variable_assignment (variable_name) (word))
+  (variable_assignment (variable_name) (word)))
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/known-diffs.txt b/packages/tree-sitter-bash/test/fixtures/corpus/known-diffs.txt
new file mode 100644
index 0000000000..57622d11ca
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/known-diffs.txt
@@ -0,0 +1,2009 @@
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredoc redirects
+node <<JS
+console.log("hi")
+JS
+
+bash -c <<JS
+echo hi
+JS
+
+newins <<-EOF - org.freedesktop.Notifications.service
+	[D-BUS Service]
+	Name=org.freedesktop.Notifications
+	Exec=/usr/libexec/notification-daemon
+EOF
+---
+hasError: false
+program [0,206] "node <<JS\nconsole.log(\"hi\")\nJS\n\nbash ..."
+  redirected_statement [0,30] "node <<JS\nconsole.log(\"hi\")\nJS"
+    command [0,4] "node"
+      command_name [0,4] "node"
+        word [0,4] "node"
+    heredoc_redirect [5,30] "<<JS\nconsole.log(\"hi\")\nJS"
+      (<<) [5,7] "<<"
+      heredoc_start [7,9] "JS"
+      heredoc_body [10,28] "console.log(\"hi\")\n"
+        heredoc_content [10,28] "console.log(\"hi\")\n"
+      heredoc_end [28,30] "JS"
+  redirected_statement [32,55] "bash -c <<JS\necho hi\nJS"
+    command [32,39] "bash -c"
+      command_name [32,36] "bash"
+        word [32,36] "bash"
+      word [37,39] "-c"
+    heredoc_redirect [40,55] "<<JS\necho hi\nJS"
+      (<<) [40,42] "<<"
+      heredoc_start [42,44] "JS"
+      heredoc_body [45,53] "echo hi\n"
+        heredoc_content [45,53] "echo hi\n"
+      heredoc_end [53,55] "JS"
+  redirected_statement [57,206] "newins <<-EOF - org.freedesktop.Notif..."
+    command [57,63] "newins"
+      command_name [57,63] "newins"
+        word [57,63] "newins"
+    heredoc_redirect [64,206] "<<-EOF - org.freedesktop.Notification..."
+      (<<-) [64,67] "<<-"
+      heredoc_start [67,70] "EOF"
+      word [71,72] "-"
+      word [73,110] "org.freedesktop.Notifications.service"
+      heredoc_body [112,203] "[D-BUS Service]\n\tName=org.freedesktop..."
+        heredoc_content [112,203] "[D-BUS Service]\n\tName=org.freedesktop..."
+      heredoc_end [203,206] "EOF"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with variables
+node <<JS
+a $B ${C}
+JS
+
+exit
+---
+hasError: false
+program [0,28] "node <<JS\na $B ${C}\nJS\n\nexit"
+  redirected_statement [0,22] "node <<JS\na $B ${C}\nJS"
+    command [0,4] "node"
+      command_name [0,4] "node"
+        word [0,4] "node"
+    heredoc_redirect [5,22] "<<JS\na $B ${C}\nJS"
+      (<<) [5,7] "<<"
+      heredoc_start [7,9] "JS"
+      heredoc_body [10,20] "a $B ${C}\n"
+        heredoc_content [10,12] "a "
+        simple_expansion [12,14] "$B"
+          ($) [12,13] "$"
+          variable_name [13,14] "B"
+        heredoc_content [14,15] " "
+        expansion [15,19] "${C}"
+          (${) [15,17] "${"
+          variable_name [17,18] "C"
+          (}) [18,19] "}"
+        heredoc_content [19,20] "\n"
+      heredoc_end [20,22] "JS"
+  command [24,28] "exit"
+    command_name [24,28] "exit"
+      word [24,28] "exit"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with file redirects
+cat <<EOF > $tmpfile
+a $B ${C}
+EOF
+
+wc -l $tmpfile
+---
+hasError: false
+program [0,50] "cat <<EOF > $tmpfile\na $B ${C}\nEOF\n\nw..."
+  redirected_statement [0,34] "cat <<EOF > $tmpfile\na $B ${C}\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,34] "<<EOF > $tmpfile\na $B ${C}\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      file_redirect [10,20] "> $tmpfile"
+        (>) [10,11] ">"
+        simple_expansion [12,20] "$tmpfile"
+          ($) [12,13] "$"
+          variable_name [13,20] "tmpfile"
+      heredoc_body [21,31] "a $B ${C}\n"
+        heredoc_content [21,23] "a "
+        simple_expansion [23,25] "$B"
+          ($) [23,24] "$"
+          variable_name [24,25] "B"
+        heredoc_content [25,26] " "
+        expansion [26,30] "${C}"
+          (${) [26,28] "${"
+          variable_name [28,29] "C"
+          (}) [29,30] "}"
+        heredoc_content [30,31] "\n"
+      heredoc_end [31,34] "EOF"
+  command [36,50] "wc -l $tmpfile"
+    command_name [36,38] "wc"
+      word [36,38] "wc"
+    word [39,41] "-l"
+    simple_expansion [42,50] "$tmpfile"
+      ($) [42,43] "$"
+      variable_name [43,50] "tmpfile"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with many file redirects
+FOO=bar echo <<EOF 2> err.txt > hello.txt
+hello
+EOF
+---
+hasError: false
+program [0,51] "FOO=bar echo <<EOF 2> err.txt > hello..."
+  redirected_statement [0,51] "FOO=bar echo <<EOF 2> err.txt > hello..."
+    command [0,12] "FOO=bar echo"
+      variable_assignment [0,7] "FOO=bar"
+        variable_name [0,3] "FOO"
+        (=) [3,4] "="
+        word [4,7] "bar"
+      command_name [8,12] "echo"
+        word [8,12] "echo"
+    heredoc_redirect [13,51] "<<EOF 2> err.txt > hello.txt\nhello\nEOF"
+      (<<) [13,15] "<<"
+      heredoc_start [15,18] "EOF"
+      file_redirect [19,29] "2> err.txt"
+        file_descriptor [19,20] "2"
+        (>) [20,21] ">"
+        word [22,29] "err.txt"
+      file_redirect [30,41] "> hello.txt"
+        (>) [30,31] ">"
+        word [32,41] "hello.txt"
+      heredoc_body [42,48] "hello\n"
+        heredoc_content [42,48] "hello\n"
+      heredoc_end [48,51] "EOF"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with pipes
+one <<EOF | grep two
+three
+EOF
+---
+hasError: false
+program [0,30] "one <<EOF | grep two\nthree\nEOF"
+  redirected_statement [0,30] "one <<EOF | grep two\nthree\nEOF"
+    command [0,3] "one"
+      command_name [0,3] "one"
+        word [0,3] "one"
+    heredoc_redirect [4,30] "<<EOF | grep two\nthree\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      pipeline [10,20] "| grep two"
+        (|) [10,11] "|"
+        command [12,20] "grep two"
+          command_name [12,16] "grep"
+            word [12,16] "grep"
+          word [17,20] "two"
+      heredoc_body [21,27] "three\n"
+        heredoc_content [21,27] "three\n"
+      heredoc_end [27,30] "EOF"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with escaped expansions
+cat  << EOF
+DEV_NAME=\$(lsblk)
+EOF
+---
+hasError: false
+program [0,34] "cat  << EOF\nDEV_NAME=\\$(lsblk)\nEOF"
+  redirected_statement [0,34] "cat  << EOF\nDEV_NAME=\\$(lsblk)\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [5,34] "<< EOF\nDEV_NAME=\\$(lsblk)\nEOF"
+      (<<) [5,7] "<<"
+      heredoc_start [8,11] "EOF"
+      heredoc_body [12,31] "DEV_NAME=\\$(lsblk)\n"
+        heredoc_content [12,31] "DEV_NAME=\\$(lsblk)\n"
+      heredoc_end [31,34] "EOF"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with indented closing delimiters
+usage() {
+	cat <<-EOF
+		Usage: ${0##*/} FOO BAR
+	EOF
+}
+---
+hasError: false
+program [0,54] "usage() {\n\tcat <<-EOF\n\t\tUsage: ${0##*..."
+  function_definition [0,54] "usage() {\n\tcat <<-EOF\n\t\tUsage: ${0##*..."
+    word [0,5] "usage"
+    (() [5,6] "("
+    ()) [6,7] ")"
+    compound_statement [8,54] "{\n\tcat <<-EOF\n\t\tUsage: ${0##*/} FOO B..."
+      ({) [8,9] "{"
+      redirected_statement [11,52] "cat <<-EOF\n\t\tUsage: ${0##*/} FOO BAR\n..."
+        command [11,14] "cat"
+          command_name [11,14] "cat"
+            word [11,14] "cat"
+        heredoc_redirect [15,52] "<<-EOF\n\t\tUsage: ${0##*/} FOO BAR\n\tEOF"
+          (<<-) [15,18] "<<-"
+          heredoc_start [18,21] "EOF"
+          heredoc_body [24,49] "Usage: ${0##*/} FOO BAR\n\t"
+            heredoc_content [24,31] "Usage: "
+            expansion [31,39] "${0##*/}"
+              (${) [31,33] "${"
+              special_variable_name [33,34] "0"
+              (##) [34,36] "##"
+              regex [36,38] "*/"
+              (}) [38,39] "}"
+            heredoc_content [39,49] " FOO BAR\n\t"
+          heredoc_end [49,52] "EOF"
+      (}) [53,54] "}"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with weird characters
+node <<_DELIMITER_WITH_UNDERSCORES_
+Hello.
+_DELIMITER_WITH_UNDERSCORES_
+
+node <<'```'
+Hello.
+```
+
+node <<!HEREDOC!
+Hello.
+!HEREDOC!
+
+node <<\'
+Hello.
+'
+
+node <<\\
+Hello.
+\
+---
+hasError: false
+program [0,171] "node <<_DELIMITER_WITH_UNDERSCORES_\nH..."
+  redirected_statement [0,71] "node <<_DELIMITER_WITH_UNDERSCORES_\nH..."
+    command [0,4] "node"
+      command_name [0,4] "node"
+        word [0,4] "node"
+    heredoc_redirect [5,71] "<<_DELIMITER_WITH_UNDERSCORES_\nHello...."
+      (<<) [5,7] "<<"
+      heredoc_start [7,35] "_DELIMITER_WITH_UNDERSCORES_"
+      heredoc_body [36,43] "Hello.\n"
+        heredoc_content [36,43] "Hello.\n"
+      heredoc_end [43,71] "_DELIMITER_WITH_UNDERSCORES_"
+  redirected_statement [73,96] "node <<'```'\nHello.\n```"
+    command [73,77] "node"
+      command_name [73,77] "node"
+        word [73,77] "node"
+    heredoc_redirect [78,96] "<<'```'\nHello.\n```"
+      (<<) [78,80] "<<"
+      heredoc_start [80,85] "'```'"
+      heredoc_body [86,93] "Hello.\n"
+      heredoc_end [93,96] "```"
+  redirected_statement [98,131] "node <<!HEREDOC!\nHello.\n!HEREDOC!"
+    command [98,102] "node"
+      command_name [98,102] "node"
+        word [98,102] "node"
+    heredoc_redirect [103,131] "<<!HEREDOC!\nHello.\n!HEREDOC!"
+      (<<) [103,105] "<<"
+      heredoc_start [105,114] "!HEREDOC!"
+      heredoc_body [115,122] "Hello.\n"
+        heredoc_content [115,122] "Hello.\n"
+      heredoc_end [122,131] "!HEREDOC!"
+  redirected_statement [133,151] "node <<\\'\nHello.\n'"
+    command [133,137] "node"
+      command_name [133,137] "node"
+        word [133,137] "node"
+    heredoc_redirect [138,151] "<<\\'\nHello.\n'"
+      (<<) [138,140] "<<"
+      heredoc_start [140,142] "\\'"
+      heredoc_body [143,150] "Hello.\n"
+      heredoc_end [150,151] "'"
+  redirected_statement [153,171] "node <<\\\\\nHello.\n\\"
+    command [153,157] "node"
+      command_name [153,157] "node"
+        word [153,157] "node"
+    heredoc_redirect [158,171] "<<\\\\\nHello.\n\\"
+      (<<) [158,160] "<<"
+      heredoc_start [160,162] "\\\\"
+      heredoc_body [163,170] "Hello.\n"
+      heredoc_end [170,171] "\\"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with a rhs statement
+cat <<-_EOF_ || die "cat EOF failed"
+	#!/bin/sh
+	echo hello
+_EOF_
+---
+hasError: false
+program [0,65] "cat <<-_EOF_ || die \"cat EOF failed\"\n..."
+  redirected_statement [0,65] "cat <<-_EOF_ || die \"cat EOF failed\"\n..."
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,65] "<<-_EOF_ || die \"cat EOF failed\"\n\t#!/..."
+      (<<-) [4,7] "<<-"
+      heredoc_start [7,12] "_EOF_"
+      (||) [13,15] "||"
+      command [16,36] "die \"cat EOF failed\""
+        command_name [16,19] "die"
+          word [16,19] "die"
+        string [20,36] "\"cat EOF failed\""
+          (") [20,21] "\""
+          string_content [21,35] "cat EOF failed"
+          (") [35,36] "\""
+      heredoc_body [38,60] "#!/bin/sh\n\techo hello\n"
+        heredoc_content [38,60] "#!/bin/sh\n\techo hello\n"
+      heredoc_end [60,65] "_EOF_"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Heredocs with a $ that is not an expansion
+cat <<EOF
+# check out this regex '^EOF$'
+EOF
+---
+hasError: false
+program [0,44] "cat <<EOF\n# check out this regex '^EO..."
+  redirected_statement [0,44] "cat <<EOF\n# check out this regex '^EO..."
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,44] "<<EOF\n# check out this regex '^EOF$'\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,41] "# check out this regex '^EOF$'\n"
+        heredoc_content [10,41] "# check out this regex '^EOF$'\n"
+      heredoc_end [41,44] "EOF"
+===
+===
+@known-diff heredoc-content-chunks: corpus commands.txt::Nested Heredocs
+cat <<OUTER
+Outer Heredoc Start
+$(cat <<INNER
+Inner Heredoc Content
+$(cat <<INNERMOST
+Innermost Heredoc Content
+INNERMOST
+)
+INNER)
+Outer Heredoc End
+OUTER
+---
+hasError: false
+program [0,154] "cat <<OUTER\nOuter Heredoc Start\n$(cat..."
+  redirected_statement [0,154] "cat <<OUTER\nOuter Heredoc Start\n$(cat..."
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,154] "<<OUTER\nOuter Heredoc Start\n$(cat <<I..."
+      (<<) [4,6] "<<"
+      heredoc_start [6,11] "OUTER"
+      heredoc_body [12,149] "Outer Heredoc Start\n$(cat <<INNER\nInn..."
+        heredoc_content [12,32] "Outer Heredoc Start\n"
+        command_substitution [32,130] "$(cat <<INNER\nInner Heredoc Content\n$..."
+          ($() [32,34] "$("
+          redirected_statement [34,129] "cat <<INNER\nInner Heredoc Content\n$(c..."
+            command [34,37] "cat"
+              command_name [34,37] "cat"
+                word [34,37] "cat"
+            heredoc_redirect [38,129] "<<INNER\nInner Heredoc Content\n$(cat <..."
+              (<<) [38,40] "<<"
+              heredoc_start [40,45] "INNER"
+              heredoc_body [46,124] "Inner Heredoc Content\n$(cat <<INNERMO..."
+                heredoc_content [46,68] "Inner Heredoc Content\n"
+                command_substitution [68,123] "$(cat <<INNERMOST\nInnermost Heredoc C..."
+                  ($() [68,70] "$("
+                  redirected_statement [70,121] "cat <<INNERMOST\nInnermost Heredoc Con..."
+                    command [70,73] "cat"
+                      command_name [70,73] "cat"
+                        word [70,73] "cat"
+                    heredoc_redirect [74,121] "<<INNERMOST\nInnermost Heredoc Content..."
+                      (<<) [74,76] "<<"
+                      heredoc_start [76,85] "INNERMOST"
+                      heredoc_body [86,112] "Innermost Heredoc Content\n"
+                        heredoc_content [86,112] "Innermost Heredoc Content\n"
+                      heredoc_end [112,121] "INNERMOST"
+                  ()) [122,123] ")"
+                heredoc_content [123,124] "\n"
+              heredoc_end [124,129] "INNER"
+          ()) [129,130] ")"
+        heredoc_content [130,149] "\nOuter Heredoc End\n"
+      heredoc_end [149,154] "OUTER"
+===
+===
+@known-diff expansion-replacement-escaped-backslash: corpus literals.txt::Variable expansions with regexes
+A=${B//:;;/$'\n'}
+
+# escaped space
+C=${D/;\ *;|}
+MOFILES=${LINGUAS// /.po }.po
+MY_P="${PN/aspell/aspell"${ASPELL_VERSION}"}"
+pyc=${pyc//*\/}
+${pv/\.}
+${new_test_cp//"${old_ver_cp}"/}
+${tests_to_run//"${classes}"\/}
+${allarchives// /\\|}
+---
+hasError: false
+program [0,236] "A=${B//:;;/$'\\n'}\n\n# escaped space\nC=..."
+  variable_assignment [0,17] "A=${B//:;;/$'\\n'}"
+    variable_name [0,1] "A"
+    (=) [1,2] "="
+    expansion [2,17] "${B//:;;/$'\\n'}"
+      (${) [2,4] "${"
+      variable_name [4,5] "B"
+      (//) [5,7] "//"
+      regex [7,10] ":;;"
+      (/) [10,11] "/"
+      ansi_c_string [11,16] "$'\\n'"
+      (}) [16,17] "}"
+  comment [19,34] "# escaped space"
+  variable_assignment [35,48] "C=${D/;\\ *;|}"
+    variable_name [35,36] "C"
+    (=) [36,37] "="
+    expansion [37,48] "${D/;\\ *;|}"
+      (${) [37,39] "${"
+      variable_name [39,40] "D"
+      (/) [40,41] "/"
+      regex [41,47] ";\\ *;|"
+      (}) [47,48] "}"
+  variable_assignment [49,78] "MOFILES=${LINGUAS// /.po }.po"
+    variable_name [49,56] "MOFILES"
+    (=) [56,57] "="
+    concatenation [57,78] "${LINGUAS// /.po }.po"
+      expansion [57,75] "${LINGUAS// /.po }"
+        (${) [57,59] "${"
+        variable_name [59,66] "LINGUAS"
+        (//) [66,68] "//"
+        regex [68,69] " "
+        (/) [69,70] "/"
+        word [70,74] ".po "
+        (}) [74,75] "}"
+      word [75,78] ".po"
+  variable_assignment [79,124] "MY_P=\"${PN/aspell/aspell\"${ASPELL_VER..."
+    variable_name [79,83] "MY_P"
+    (=) [83,84] "="
+    string [84,124] "\"${PN/aspell/aspell\"${ASPELL_VERSION}\"}\""
+      (") [84,85] "\""
+      expansion [85,123] "${PN/aspell/aspell\"${ASPELL_VERSION}\"}"
+        (${) [85,87] "${"
+        variable_name [87,89] "PN"
+        (/) [89,90] "/"
+        regex [90,96] "aspell"
+        (/) [96,97] "/"
+        concatenation [97,122] "aspell\"${ASPELL_VERSION}\""
+          word [97,103] "aspell"
+          string [103,122] "\"${ASPELL_VERSION}\""
+            (") [103,104] "\""
+            expansion [104,121] "${ASPELL_VERSION}"
+              (${) [104,106] "${"
+              variable_name [106,120] "ASPELL_VERSION"
+              (}) [120,121] "}"
+            (") [121,122] "\""
+        (}) [122,123] "}"
+      (") [123,124] "\""
+  variable_assignment [125,140] "pyc=${pyc//*\\/}"
+    variable_name [125,128] "pyc"
+    (=) [128,129] "="
+    expansion [129,140] "${pyc//*\\/}"
+      (${) [129,131] "${"
+      variable_name [131,134] "pyc"
+      (//) [134,136] "//"
+      regex [136,138] "*\\"
+      (/) [138,139] "/"
+      (}) [139,140] "}"
+  command [141,149] "${pv/\\.}"
+    command_name [141,149] "${pv/\\.}"
+      expansion [141,149] "${pv/\\.}"
+        (${) [141,143] "${"
+        variable_name [143,145] "pv"
+        (/) [145,146] "/"
+        regex [146,148] "\\."
+        (}) [148,149] "}"
+  command [150,182] "${new_test_cp//\"${old_ver_cp}\"/}"
+    command_name [150,182] "${new_test_cp//\"${old_ver_cp}\"/}"
+      expansion [150,182] "${new_test_cp//\"${old_ver_cp}\"/}"
+        (${) [150,152] "${"
+        variable_name [152,163] "new_test_cp"
+        (//) [163,165] "//"
+        string [165,180] "\"${old_ver_cp}\""
+          (") [165,166] "\""
+          expansion [166,179] "${old_ver_cp}"
+            (${) [166,168] "${"
+            variable_name [168,178] "old_ver_cp"
+            (}) [178,179] "}"
+          (") [179,180] "\""
+        (/) [180,181] "/"
+        (}) [181,182] "}"
+  command [183,214] "${tests_to_run//\"${classes}\"\\/}"
+    command_name [183,214] "${tests_to_run//\"${classes}\"\\/}"
+      expansion [183,214] "${tests_to_run//\"${classes}\"\\/}"
+        (${) [183,185] "${"
+        variable_name [185,197] "tests_to_run"
+        (//) [197,199] "//"
+        string [199,211] "\"${classes}\""
+          (") [199,200] "\""
+          expansion [200,210] "${classes}"
+            (${) [200,202] "${"
+            variable_name [202,209] "classes"
+            (}) [209,210] "}"
+          (") [210,211] "\""
+        regex [211,212] "\\"
+        (/) [212,213] "/"
+        (}) [213,214] "}"
+  command [215,236] "${allarchives// /\\\\|}"
+    command_name [215,236] "${allarchives// /\\\\|}"
+      expansion [215,236] "${allarchives// /\\\\|}"
+        (${) [215,217] "${"
+        variable_name [217,228] "allarchives"
+        (//) [228,230] "//"
+        regex [230,231] " "
+        (/) [231,232] "/"
+        word [232,235] "\\\\|"
+        (}) [235,236] "}"
+===
+===
+@known-diff expansion-bang-hash-special: corpus literals.txt::Variable Expansions: Bizarre Cases
+${!#}
+${!# }
+${!##}
+${!## }
+${!##/}
+# here be dragons
+echo "${kw}? ( ${cond:+${cond}? (} ${baseuri}-${ver}-${kw}.${suff} ${cond:+) })"
+---
+hasError: true
+program [0,134] "${!#}\n${!# }\n${!##}\n${!## }\n${!##/}\n#..."
+  command [0,5] "${!#}"
+    command_name [0,5] "${!#}"
+      expansion [0,5] "${!#}"
+        (${) [0,2] "${"
+        (!) [2,3] "!"
+        (#) [3,4] "#"
+        (}) [4,5] "}"
+  command [6,12] "${!# }"
+    command_name [6,12] "${!# }"
+      expansion [6,12] "${!# }"
+        (${) [6,8] "${"
+        (!) [8,9] "!"
+        special_variable_name [9,10] "#"
+        word [10,11] " "
+        (}) [11,12] "}"
+  command [13,19] "${!##}"
+    command_name [13,19] "${!##}"
+      expansion [13,19] "${!##}"
+        (${) [13,15] "${"
+        (!) [15,16] "!"
+        (#) [16,17] "#"
+        (#) [17,18] "#"
+        (}) [18,19] "}"
+  command [20,27] "${!## }"
+    command_name [20,27] "${!## }"
+      expansion [20,27] "${!## }"
+        (${) [20,22] "${"
+        (!) [22,23] "!"
+        (#) [23,24] "#"
+        special_variable_name [24,25] "#"
+        word [25,26] " "
+        (}) [26,27] "}"
+  command [28,35] "${!##/}"
+    command_name [28,35] "${!##/}"
+      expansion [28,35] "${!##/}"
+        (${) [28,30] "${"
+        (!) [30,31] "!"
+        (#) [31,32] "#"
+        special_variable_name [32,33] "#"
+        (/) [33,34] "/"
+        (}) [34,35] "}"
+  comment [36,53] "# here be dragons"
+  command [54,134] "echo \"${kw}? ( ${cond:+${cond}? (} ${..."
+    command_name [54,58] "echo"
+      word [54,58] "echo"
+    string [59,134] "\"${kw}? ( ${cond:+${cond}? (} ${baseu..."
+      (") [59,60] "\""
+      expansion [60,65] "${kw}"
+        (${) [60,62] "${"
+        variable_name [62,64] "kw"
+        (}) [64,65] "}"
+      string_content [65,69] "? ( "
+      expansion [69,88] "${cond:+${cond}? (}"
+        (${) [69,71] "${"
+        variable_name [71,75] "cond"
+        (:+) [75,77] ":+"
+        concatenation [77,87] "${cond}? ("
+          expansion [77,84] "${cond}"
+            (${) [77,79] "${"
+            variable_name [79,83] "cond"
+            (}) [83,84] "}"
+          word [84,87] "? ("
+        (}) [87,88] "}"
+      expansion [88,99] " ${baseuri}"
+        (${) [88,91] " ${"
+        variable_name [91,98] "baseuri"
+        (}) [98,99] "}"
+      string_content [99,100] "-"
+      expansion [100,106] "${ver}"
+        (${) [100,102] "${"
+        variable_name [102,105] "ver"
+        (}) [105,106] "}"
+      string_content [106,107] "-"
+      expansion [107,112] "${kw}"
+        (${) [107,109] "${"
+        variable_name [109,111] "kw"
+        (}) [111,112] "}"
+      string_content [112,113] "."
+      expansion [113,120] "${suff}"
+        (${) [113,115] "${"
+        variable_name [115,119] "suff"
+        (}) [119,120] "}"
+      expansion [120,132] " ${cond:+) }"
+        (${) [120,123] " ${"
+        variable_name [123,127] "cond"
+        (:+) [127,129] ":+"
+        word [129,131] ") "
+        (}) [131,132] "}"
+      string_content [132,133] ")"
+      (") [133,134] "\""
+===
+===
+@known-diff expansion-equals-operator: corpus literals.txt::Variable Expansions: Weird Cases
+${completions[*]}
+${=1}
+${2?}
+${p_key#*=}
+${abc:- }
+${B[0]# }
+${to_enables[0]##*/}
+exec "${0#-}" --rcfile "${BASH_IT_BASHRC:-${HOME?}/.bashrc}"
+recho "TDEFAULTS = ${selvecs:+-DSELECT_VECS=\"$selvecs\"}"
+local msg="${2:-command '$1' does not exist}"
+${cdir:+#}
+${dict_langs:+;}
+${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/util-linux )}
+${id}${2+ ${2}}
+${BRANDING_GCC_PKGVERSION/(/(Gentoo ${PVR}${extvers}, } # look at that parenthesis!
+some-command ${foo:+--arg <(printf '%s\n' "$foo")}
+---
+hasError: false
+program [0,477] "${completions[*]}\n${=1}\n${2?}\n${p_key..."
+  command [0,17] "${completions[*]}"
+    command_name [0,17] "${completions[*]}"
+      expansion [0,17] "${completions[*]}"
+        (${) [0,2] "${"
+        subscript [2,16] "completions[*]"
+          variable_name [2,13] "completions"
+          ([) [13,14] "["
+          word [14,15] "*"
+          (]) [15,16] "]"
+        (}) [16,17] "}"
+  command [18,23] "${=1}"
+    command_name [18,23] "${=1}"
+      expansion [18,23] "${=1}"
+        (${) [18,20] "${"
+        (=) [20,21] "="
+        word [21,22] "1"
+        (}) [22,23] "}"
+  command [24,29] "${2?}"
+    command_name [24,29] "${2?}"
+      expansion [24,29] "${2?}"
+        (${) [24,26] "${"
+        variable_name [26,27] "2"
+        (?) [27,28] "?"
+        (}) [28,29] "}"
+  command [30,41] "${p_key#*=}"
+    command_name [30,41] "${p_key#*=}"
+      expansion [30,41] "${p_key#*=}"
+        (${) [30,32] "${"
+        variable_name [32,37] "p_key"
+        (#) [37,38] "#"
+        regex [38,40] "*="
+        (}) [40,41] "}"
+  command [42,51] "${abc:- }"
+    command_name [42,51] "${abc:- }"
+      expansion [42,51] "${abc:- }"
+        (${) [42,44] "${"
+        variable_name [44,47] "abc"
+        (:-) [47,49] ":-"
+        word [49,50] " "
+        (}) [50,51] "}"
+  command [52,61] "${B[0]# }"
+    command_name [52,61] "${B[0]# }"
+      expansion [52,61] "${B[0]# }"
+        (${) [52,54] "${"
+        subscript [54,58] "B[0]"
+          variable_name [54,55] "B"
+          ([) [55,56] "["
+          number [56,57] "0"
+          (]) [57,58] "]"
+        (#) [58,59] "#"
+        regex [59,60] " "
+        (}) [60,61] "}"
+  command [62,82] "${to_enables[0]##*/}"
+    command_name [62,82] "${to_enables[0]##*/}"
+      expansion [62,82] "${to_enables[0]##*/}"
+        (${) [62,64] "${"
+        subscript [64,77] "to_enables[0]"
+          variable_name [64,74] "to_enables"
+          ([) [74,75] "["
+          number [75,76] "0"
+          (]) [76,77] "]"
+        (##) [77,79] "##"
+        regex [79,81] "*/"
+        (}) [81,82] "}"
+  command [83,143] "exec \"${0#-}\" --rcfile \"${BASH_IT_BAS..."
+    command_name [83,87] "exec"
+      word [83,87] "exec"
+    string [88,96] "\"${0#-}\""
+      (") [88,89] "\""
+      expansion [89,95] "${0#-}"
+        (${) [89,91] "${"
+        special_variable_name [91,92] "0"
+        (#) [92,93] "#"
+        regex [93,94] "-"
+        (}) [94,95] "}"
+      (") [95,96] "\""
+    word [97,105] "--rcfile"
+    string [106,143] "\"${BASH_IT_BASHRC:-${HOME?}/.bashrc}\""
+      (") [106,107] "\""
+      expansion [107,142] "${BASH_IT_BASHRC:-${HOME?}/.bashrc}"
+        (${) [107,109] "${"
+        variable_name [109,123] "BASH_IT_BASHRC"
+        (:-) [123,125] ":-"
+        concatenation [125,141] "${HOME?}/.bashrc"
+          expansion [125,133] "${HOME?}"
+            (${) [125,127] "${"
+            variable_name [127,131] "HOME"
+            (?) [131,132] "?"
+            (}) [132,133] "}"
+          word [133,141] "/.bashrc"
+        (}) [141,142] "}"
+      (") [142,143] "\""
+  command [144,202] "recho \"TDEFAULTS = ${selvecs:+-DSELEC..."
+    command_name [144,149] "recho"
+      word [144,149] "recho"
+    string [150,202] "\"TDEFAULTS = ${selvecs:+-DSELECT_VECS..."
+      (") [150,151] "\""
+      string_content [151,163] "TDEFAULTS = "
+      expansion [163,201] "${selvecs:+-DSELECT_VECS=\\\"$selvecs\\\"}"
+        (${) [163,165] "${"
+        variable_name [165,172] "selvecs"
+        (:+) [172,174] ":+"
+        concatenation [174,200] "-DSELECT_VECS=\\\"$selvecs\\\""
+          word [174,190] "-DSELECT_VECS=\\\""
+          simple_expansion [190,198] "$selvecs"
+            ($) [190,191] "$"
+            variable_name [191,198] "selvecs"
+          word [198,200] "\\\""
+        (}) [200,201] "}"
+      (") [201,202] "\""
+  declaration_command [203,248] "local msg=\"${2:-command '$1' does not..."
+    (local) [203,208] "local"
+    variable_assignment [209,248] "msg=\"${2:-command '$1' does not exist}\""
+      variable_name [209,212] "msg"
+      (=) [212,213] "="
+      string [213,248] "\"${2:-command '$1' does not exist}\""
+        (") [213,214] "\""
+        expansion [214,247] "${2:-command '$1' does not exist}"
+          (${) [214,216] "${"
+          variable_name [216,217] "2"
+          (:-) [217,219] ":-"
+          concatenation [219,246] "command '$1' does not exist"
+            word [219,226] "command"
+            raw_string [227,231] "'$1'"
+            word [231,246] " does not exist"
+          (}) [246,247] "}"
+        (") [247,248] "\""
+  command [249,259] "${cdir:+#}"
+    command_name [249,259] "${cdir:+#}"
+      expansion [249,259] "${cdir:+#}"
+        (${) [249,251] "${"
+        variable_name [251,255] "cdir"
+        (:+) [255,257] ":+"
+        word [257,258] "#"
+        (}) [258,259] "}"
+  command [260,276] "${dict_langs:+;}"
+    command_name [260,276] "${dict_langs:+;}"
+      expansion [260,276] "${dict_langs:+;}"
+        (${) [260,262] "${"
+        variable_name [262,272] "dict_langs"
+        (:+) [272,274] ":+"
+        word [274,275] ";"
+        (}) [275,276] "}"
+  command [277,326] "${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/u..."
+    command_name [277,326] "${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/u..."
+      expansion [277,326] "${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/u..."
+        (${) [277,279] "${"
+        subscript [279,297] "UTIL_LINUX_LIBC[@]"
+          variable_name [279,294] "UTIL_LINUX_LIBC"
+          ([) [294,295] "["
+          word [295,296] "@"
+          (]) [296,297] "]"
+        (/%) [297,299] "/%"
+        (/) [299,300] "/"
+        word [300,325] "? ( sys-apps/util-linux )"
+        (}) [325,326] "}"
+  command [327,342] "${id}${2+ ${2}}"
+    command_name [327,342] "${id}${2+ ${2}}"
+      concatenation [327,342] "${id}${2+ ${2}}"
+        expansion [327,332] "${id}"
+          (${) [327,329] "${"
+          variable_name [329,331] "id"
+          (}) [331,332] "}"
+        expansion [332,342] "${2+ ${2}}"
+          (${) [332,334] "${"
+          variable_name [334,335] "2"
+          (+) [335,336] "+"
+          expansion [337,341] "${2}"
+            (${) [337,339] "${"
+            variable_name [339,340] "2"
+            (}) [340,341] "}"
+          (}) [341,342] "}"
+  command [343,398] "${BRANDING_GCC_PKGVERSION/(/(Gentoo $..."
+    command_name [343,398] "${BRANDING_GCC_PKGVERSION/(/(Gentoo $..."
+      expansion [343,398] "${BRANDING_GCC_PKGVERSION/(/(Gentoo $..."
+        (${) [343,345] "${"
+        variable_name [345,368] "BRANDING_GCC_PKGVERSION"
+        (/) [368,369] "/"
+        regex [369,370] "("
+        (/) [370,371] "/"
+        concatenation [371,397] "(Gentoo ${PVR}${extvers}, "
+          word [371,379] "(Gentoo "
+          expansion [379,385] "${PVR}"
+            (${) [379,381] "${"
+            variable_name [381,384] "PVR"
+            (}) [384,385] "}"
+          expansion [385,395] "${extvers}"
+            (${) [385,387] "${"
+            variable_name [387,394] "extvers"
+            (}) [394,395] "}"
+          word [395,397] ", "
+        (}) [397,398] "}"
+  comment [399,426] "# look at that parenthesis!"
+  command [427,477] "some-command ${foo:+--arg <(printf '%..."
+    command_name [427,439] "some-command"
+      word [427,439] "some-command"
+    expansion [440,477] "${foo:+--arg <(printf '%s\\n' \"$foo\")}"
+      (${) [440,442] "${"
+      variable_name [442,445] "foo"
+      (:+) [445,447] ":+"
+      concatenation [447,476] "--arg <(printf '%s\\n' \"$foo\")"
+        word [447,452] "--arg"
+        process_substitution [453,476] "<(printf '%s\\n' \"$foo\")"
+          (<() [453,455] "<("
+          command [455,475] "printf '%s\\n' \"$foo\""
+            command_name [455,461] "printf"
+              word [455,461] "printf"
+            raw_string [462,468] "'%s\\n'"
+            string [469,475] "\"$foo\""
+              (") [469,470] "\""
+              simple_expansion [470,474] "$foo"
+                ($) [470,471] "$"
+                variable_name [471,474] "foo"
+              (") [474,475] "\""
+          ()) [475,476] ")"
+      (}) [476,477] "}"
+===
+===
+@known-diff number-base-with-expansion: corpus literals.txt::Variable Expansions: Regex
+A=${B//:;;/$'\n'}
+C="${D/#* -E /}"
+BASH_IT_GIT_URL="${BASH_IT_GIT_URL/git@/https://}"
+10#${command_start##*.}
+echo ${LIB_DEPEND//\[static-libs(+)]}
+${ALL_LLVM_TARGETS[@]/%/(-)?}
+filterdiff -p1 ${paths[@]/#/-i }
+${cflags//-O? /$(get-flag O) }
+curf="${f%'-roff2html'*}.html"
+reff="${f/'-roff2html'*/'-ref'}.html"
+---
+hasError: false
+program [0,310] "A=${B//:;;/$'\\n'}\nC=\"${D/#* -E /}\"\nBA..."
+  variable_assignment [0,17] "A=${B//:;;/$'\\n'}"
+    variable_name [0,1] "A"
+    (=) [1,2] "="
+    expansion [2,17] "${B//:;;/$'\\n'}"
+      (${) [2,4] "${"
+      variable_name [4,5] "B"
+      (//) [5,7] "//"
+      regex [7,10] ":;;"
+      (/) [10,11] "/"
+      ansi_c_string [11,16] "$'\\n'"
+      (}) [16,17] "}"
+  variable_assignment [18,34] "C=\"${D/#* -E /}\""
+    variable_name [18,19] "C"
+    (=) [19,20] "="
+    string [20,34] "\"${D/#* -E /}\""
+      (") [20,21] "\""
+      expansion [21,33] "${D/#* -E /}"
+        (${) [21,23] "${"
+        variable_name [23,24] "D"
+        (/#) [24,26] "/#"
+        regex [26,31] "* -E "
+        (/) [31,32] "/"
+        (}) [32,33] "}"
+      (") [33,34] "\""
+  variable_assignment [35,85] "BASH_IT_GIT_URL=\"${BASH_IT_GIT_URL/gi..."
+    variable_name [35,50] "BASH_IT_GIT_URL"
+    (=) [50,51] "="
+    string [51,85] "\"${BASH_IT_GIT_URL/git@/https://}\""
+      (") [51,52] "\""
+      expansion [52,84] "${BASH_IT_GIT_URL/git@/https://}"
+        (${) [52,54] "${"
+        variable_name [54,69] "BASH_IT_GIT_URL"
+        (/) [69,70] "/"
+        regex [70,74] "git@"
+        (/) [74,75] "/"
+        word [75,83] "https://"
+        (}) [83,84] "}"
+      (") [84,85] "\""
+  command [86,109] "10#${command_start##*.}"
+    command_name [86,109] "10#${command_start##*.}"
+      concatenation [86,109] "10#${command_start##*.}"
+        word [86,89] "10#"
+        expansion [89,109] "${command_start##*.}"
+          (${) [89,91] "${"
+          variable_name [91,104] "command_start"
+          (##) [104,106] "##"
+          regex [106,108] "*."
+          (}) [108,109] "}"
+  command [110,147] "echo ${LIB_DEPEND//\\[static-libs(+)]}"
+    command_name [110,114] "echo"
+      word [110,114] "echo"
+    expansion [115,147] "${LIB_DEPEND//\\[static-libs(+)]}"
+      (${) [115,117] "${"
+      variable_name [117,127] "LIB_DEPEND"
+      (//) [127,129] "//"
+      regex [129,146] "\\[static-libs(+)]"
+      (}) [146,147] "}"
+  command [148,177] "${ALL_LLVM_TARGETS[@]/%/(-)?}"
+    command_name [148,177] "${ALL_LLVM_TARGETS[@]/%/(-)?}"
+      expansion [148,177] "${ALL_LLVM_TARGETS[@]/%/(-)?}"
+        (${) [148,150] "${"
+        subscript [150,169] "ALL_LLVM_TARGETS[@]"
+          variable_name [150,166] "ALL_LLVM_TARGETS"
+          ([) [166,167] "["
+          word [167,168] "@"
+          (]) [168,169] "]"
+        (/%) [169,171] "/%"
+        (/) [171,172] "/"
+        word [172,176] "(-)?"
+        (}) [176,177] "}"
+  command [178,210] "filterdiff -p1 ${paths[@]/#/-i }"
+    command_name [178,188] "filterdiff"
+      word [178,188] "filterdiff"
+    word [189,192] "-p1"
+    expansion [193,210] "${paths[@]/#/-i }"
+      (${) [193,195] "${"
+      subscript [195,203] "paths[@]"
+        variable_name [195,200] "paths"
+        ([) [200,201] "["
+        word [201,202] "@"
+        (]) [202,203] "]"
+      (/#) [203,205] "/#"
+      (/) [205,206] "/"
+      word [206,209] "-i "
+      (}) [209,210] "}"
+  command [211,241] "${cflags//-O? /$(get-flag O) }"
+    command_name [211,241] "${cflags//-O? /$(get-flag O) }"
+      expansion [211,241] "${cflags//-O? /$(get-flag O) }"
+        (${) [211,213] "${"
+        variable_name [213,219] "cflags"
+        (//) [219,221] "//"
+        regex [221,225] "-O? "
+        (/) [225,226] "/"
+        command_substitution [226,239] "$(get-flag O)"
+          ($() [226,228] "$("
+          command [228,238] "get-flag O"
+            command_name [228,236] "get-flag"
+              word [228,236] "get-flag"
+            word [237,238] "O"
+          ()) [238,239] ")"
+        word [239,240] " "
+        (}) [240,241] "}"
+  variable_assignment [242,272] "curf=\"${f%'-roff2html'*}.html\""
+    variable_name [242,246] "curf"
+    (=) [246,247] "="
+    string [247,272] "\"${f%'-roff2html'*}.html\""
+      (") [247,248] "\""
+      expansion [248,266] "${f%'-roff2html'*}"
+        (${) [248,250] "${"
+        variable_name [250,251] "f"
+        (%) [251,252] "%"
+        raw_string [252,264] "'-roff2html'"
+        regex [264,265] "*"
+        (}) [265,266] "}"
+      string_content [266,271] ".html"
+      (") [271,272] "\""
+  variable_assignment [273,310] "reff=\"${f/'-roff2html'*/'-ref'}.html\""
+    variable_name [273,277] "reff"
+    (=) [277,278] "="
+    string [278,310] "\"${f/'-roff2html'*/'-ref'}.html\""
+      (") [278,279] "\""
+      expansion [279,304] "${f/'-roff2html'*/'-ref'}"
+        (${) [279,281] "${"
+        variable_name [281,282] "f"
+        (/) [282,283] "/"
+        regex [283,296] "'-roff2html'*"
+        (/) [296,297] "/"
+        raw_string [297,303] "'-ref'"
+        (}) [303,304] "}"
+      string_content [304,309] ".html"
+      (") [309,310] "\""
+===
+===
+@known-diff expansion-default-array: corpus literals.txt::Arrays and array expansions
+a=()
+b=(1 2 3)
+
+echo ${a[@]}
+echo ${#b[@]}
+
+a[$i]=50
+a+=(foo "bar" $(baz))
+
+printf "  %-9s" "${seq0:-(default)}"
+---
+hasError: false
+program [0,112] "a=()\nb=(1 2 3)\n\necho ${a[@]}\necho ${#..."
+  variable_assignment [0,4] "a=()"
+    variable_name [0,1] "a"
+    (=) [1,2] "="
+    array [2,4] "()"
+      (() [2,3] "("
+      ()) [3,4] ")"
+  variable_assignment [5,14] "b=(1 2 3)"
+    variable_name [5,6] "b"
+    (=) [6,7] "="
+    array [7,14] "(1 2 3)"
+      (() [7,8] "("
+      number [8,9] "1"
+      number [10,11] "2"
+      number [12,13] "3"
+      ()) [13,14] ")"
+  command [16,28] "echo ${a[@]}"
+    command_name [16,20] "echo"
+      word [16,20] "echo"
+    expansion [21,28] "${a[@]}"
+      (${) [21,23] "${"
+      subscript [23,27] "a[@]"
+        variable_name [23,24] "a"
+        ([) [24,25] "["
+        word [25,26] "@"
+        (]) [26,27] "]"
+      (}) [27,28] "}"
+  command [29,42] "echo ${#b[@]}"
+    command_name [29,33] "echo"
+      word [29,33] "echo"
+    expansion [34,42] "${#b[@]}"
+      (${) [34,36] "${"
+      (#) [36,37] "#"
+      subscript [37,41] "b[@]"
+        variable_name [37,38] "b"
+        ([) [38,39] "["
+        word [39,40] "@"
+        (]) [40,41] "]"
+      (}) [41,42] "}"
+  variable_assignment [44,52] "a[$i]=50"
+    subscript [44,49] "a[$i]"
+      variable_name [44,45] "a"
+      ([) [45,46] "["
+      simple_expansion [46,48] "$i"
+        ($) [46,47] "$"
+        variable_name [47,48] "i"
+      (]) [48,49] "]"
+    (=) [49,50] "="
+    number [50,52] "50"
+  variable_assignment [53,74] "a+=(foo \"bar\" $(baz))"
+    variable_name [53,54] "a"
+    (+=) [54,56] "+="
+    array [56,74] "(foo \"bar\" $(baz))"
+      (() [56,57] "("
+      word [57,60] "foo"
+      string [61,66] "\"bar\""
+        (") [61,62] "\""
+        string_content [62,65] "bar"
+        (") [65,66] "\""
+      command_substitution [67,73] "$(baz)"
+        ($() [67,69] "$("
+        command [69,72] "baz"
+          command_name [69,72] "baz"
+            word [69,72] "baz"
+        ()) [72,73] ")"
+      ()) [73,74] ")"
+  command [76,112] "printf \"  %-9s\" \"${seq0:-(default)}\""
+    command_name [76,82] "printf"
+      word [76,82] "printf"
+    string [83,91] "\"  %-9s\""
+      (") [83,84] "\""
+      string_content [84,90] "  %-9s"
+      (") [90,91] "\""
+    string [92,112] "\"${seq0:-(default)}\""
+      (") [92,93] "\""
+      expansion [93,111] "${seq0:-(default)}"
+        (${) [93,95] "${"
+        variable_name [95,99] "seq0"
+        (:-) [99,101] ":-"
+        word [101,110] "(default)"
+        (}) [110,111] "}"
+      (") [111,112] "\""
+===
+===
+@known-diff empty-backtick: corpus literals.txt::Concatenation with double backticks
+main() {
+    local foo="asd"`
+        `"fgh"
+}
+---
+hasError: false
+program [0,46] "main() {\n    local foo=\"asd\"`\n       ..."
+  function_definition [0,46] "main() {\n    local foo=\"asd\"`\n       ..."
+    word [0,4] "main"
+    (() [4,5] "("
+    ()) [5,6] ")"
+    compound_statement [7,46] "{\n    local foo=\"asd\"`\n        `\"fgh\"\n}"
+      ({) [7,8] "{"
+      declaration_command [13,44] "local foo=\"asd\"`\n        `\"fgh\""
+        (local) [13,18] "local"
+        variable_assignment [19,44] "foo=\"asd\"`\n        `\"fgh\""
+          variable_name [19,22] "foo"
+          (=) [22,23] "="
+          concatenation [23,44] "\"asd\"`\n        `\"fgh\""
+            string [23,28] "\"asd\""
+              (") [23,24] "\""
+              string_content [24,27] "asd"
+              (") [27,28] "\""
+            command_substitution [28,39] "`\n        `"
+              (`) [28,29] "`"
+              (`) [38,39] "`"
+            string [39,44] "\"fgh\""
+              (") [39,40] "\""
+              string_content [40,43] "fgh"
+              (") [43,44] "\""
+      (}) [45,46] "}"
+===
+===
+@known-diff escaped-space-argument: corpus programs.txt::Escaped whitespace
+echo 1 \ 2 \ 3
+---
+hasError: false
+program [0,14] "echo 1 \\ 2 \\ 3"
+  command [0,14] "echo 1 \\ 2 \\ 3"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    number [5,6] "1"
+    word [7,10] "\\ 2"
+    word [11,14] "\\ 3"
+===
+===
+@known-diff expansion-equals-operator: corpus statements.txt::For statements
+for a in 1 2 $(seq 5 10); do
+  echo $a
+done
+
+for ARG; do
+  echo $ARG
+  ARG=''
+done
+
+for c in ${=1}; do
+	echo c
+done
+---
+hasError: false
+program [0,115] "for a in 1 2 $(seq 5 10); do\n  echo $..."
+  for_statement [0,43] "for a in 1 2 $(seq 5 10); do\n  echo $..."
+    (for) [0,3] "for"
+    variable_name [4,5] "a"
+    (in) [6,8] "in"
+    number [9,10] "1"
+    number [11,12] "2"
+    command_substitution [13,24] "$(seq 5 10)"
+      ($() [13,15] "$("
+      command [15,23] "seq 5 10"
+        command_name [15,18] "seq"
+          word [15,18] "seq"
+        number [19,20] "5"
+        number [21,23] "10"
+      ()) [23,24] ")"
+    (;) [24,25] ";"
+    do_group [26,43] "do\n  echo $a\ndone"
+      (do) [26,28] "do"
+      command [31,38] "echo $a"
+        command_name [31,35] "echo"
+          word [31,35] "echo"
+        simple_expansion [36,38] "$a"
+          ($) [36,37] "$"
+          variable_name [37,38] "a"
+      (done) [39,43] "done"
+  for_statement [45,82] "for ARG; do\n  echo $ARG\n  ARG=''\ndone"
+    (for) [45,48] "for"
+    variable_name [49,52] "ARG"
+    (;) [52,53] ";"
+    do_group [54,82] "do\n  echo $ARG\n  ARG=''\ndone"
+      (do) [54,56] "do"
+      command [59,68] "echo $ARG"
+        command_name [59,63] "echo"
+          word [59,63] "echo"
+        simple_expansion [64,68] "$ARG"
+          ($) [64,65] "$"
+          variable_name [65,68] "ARG"
+      variable_assignment [71,77] "ARG=''"
+        variable_name [71,74] "ARG"
+        (=) [74,75] "="
+        raw_string [75,77] "''"
+      (done) [78,82] "done"
+  for_statement [84,115] "for c in ${=1}; do\n\techo c\ndone"
+    (for) [84,87] "for"
+    variable_name [88,89] "c"
+    (in) [90,92] "in"
+    expansion [93,98] "${=1}"
+      (${) [93,95] "${"
+      (=) [95,96] "="
+      word [96,97] "1"
+      (}) [97,98] "}"
+    (;) [98,99] ";"
+    do_group [100,115] "do\n\techo c\ndone"
+      (do) [100,102] "do"
+      command [104,110] "echo c"
+        command_name [104,108] "echo"
+          word [104,108] "echo"
+        word [109,110] "c"
+      (done) [111,115] "done"
+===
+===
+@known-diff case-pattern-after-continuation: corpus statements.txt::Case statements
+case "opt" in
+  a)
+    echo a
+    ;;
+
+  b)
+    echo b
+    ;&
+
+  c)
+    echo c;;
+esac
+
+case "opt" in
+  (a)
+    echo a
+    ;;
+
+  (b)
+    echo b
+    ;&
+
+  (c)
+    echo c;;
+esac
+
+case "$Z" in
+  ab*|cd*) ef
+esac
+
+case $dest in
+  *.[1357])
+    exit $?
+    ;;
+esac
+
+case x in x) echo meow ;; esac
+
+case foo in
+  bar\ baz) : ;;
+esac
+
+case "$arg" in
+  *([0-9])([0-9])) echo "$arg"
+esac
+
+case ${lang} in
+CMakeLists.txt | \
+	cmake_modules | \
+	${PN}.pot) ;;
+*) rm -r ${lang} || die ;;
+esac
+---
+hasError: false
+program [0,478] "case \"opt\" in\n  a)\n    echo a\n    ;;\n..."
+  case_statement [0,84] "case \"opt\" in\n  a)\n    echo a\n    ;;\n..."
+    (case) [0,4] "case"
+    string [5,10] "\"opt\""
+      (") [5,6] "\""
+      string_content [6,9] "opt"
+      (") [9,10] "\""
+    (in) [11,13] "in"
+    case_item [16,36] "a)\n    echo a\n    ;;"
+      word [16,17] "a"
+      ()) [17,18] ")"
+      command [23,29] "echo a"
+        command_name [23,27] "echo"
+          word [23,27] "echo"
+        word [28,29] "a"
+      (;;) [34,36] ";;"
+    case_item [40,60] "b)\n    echo b\n    ;&"
+      word [40,41] "b"
+      ()) [41,42] ")"
+      command [47,53] "echo b"
+        command_name [47,51] "echo"
+          word [47,51] "echo"
+        word [52,53] "b"
+      (;&) [58,60] ";&"
+    case_item [64,79] "c)\n    echo c;;"
+      word [64,65] "c"
+      ()) [65,66] ")"
+      command [71,77] "echo c"
+        command_name [71,75] "echo"
+          word [71,75] "echo"
+        word [76,77] "c"
+      (;;) [77,79] ";;"
+    (esac) [80,84] "esac"
+  case_statement [86,173] "case \"opt\" in\n  (a)\n    echo a\n    ;;..."
+    (case) [86,90] "case"
+    string [91,96] "\"opt\""
+      (") [91,92] "\""
+      string_content [92,95] "opt"
+      (") [95,96] "\""
+    (in) [97,99] "in"
+    case_item [102,123] "(a)\n    echo a\n    ;;"
+      (() [102,103] "("
+      word [103,104] "a"
+      ()) [104,105] ")"
+      command [110,116] "echo a"
+        command_name [110,114] "echo"
+          word [110,114] "echo"
+        word [115,116] "a"
+      (;;) [121,123] ";;"
+    case_item [127,148] "(b)\n    echo b\n    ;&"
+      (() [127,128] "("
+      word [128,129] "b"
+      ()) [129,130] ")"
+      command [135,141] "echo b"
+        command_name [135,139] "echo"
+          word [135,139] "echo"
+        word [140,141] "b"
+      (;&) [146,148] ";&"
+    case_item [152,168] "(c)\n    echo c;;"
+      (() [152,153] "("
+      word [153,154] "c"
+      ()) [154,155] ")"
+      command [160,166] "echo c"
+        command_name [160,164] "echo"
+          word [160,164] "echo"
+        word [165,166] "c"
+      (;;) [166,168] ";;"
+    (esac) [169,173] "esac"
+  case_statement [175,206] "case \"$Z\" in\n  ab*|cd*) ef\nesac"
+    (case) [175,179] "case"
+    string [180,184] "\"$Z\""
+      (") [180,181] "\""
+      simple_expansion [181,183] "$Z"
+        ($) [181,182] "$"
+        variable_name [182,183] "Z"
+      (") [183,184] "\""
+    (in) [185,187] "in"
+    case_item [190,202] "ab*|cd*) ef\n"
+      extglob_pattern [190,193] "ab*"
+      (|) [193,194] "|"
+      extglob_pattern [194,197] "cd*"
+      ()) [197,198] ")"
+      command [199,201] "ef"
+        command_name [199,201] "ef"
+          word [199,201] "ef"
+    (esac) [202,206] "esac"
+  case_statement [208,257] "case $dest in\n  *.[1357])\n    exit $?..."
+    (case) [208,212] "case"
+    simple_expansion [213,218] "$dest"
+      ($) [213,214] "$"
+      variable_name [214,218] "dest"
+    (in) [219,221] "in"
+    case_item [224,252] "*.[1357])\n    exit $?\n    ;;"
+      concatenation [224,232] "*.[1357]"
+        word [224,226] "*."
+        word [226,227] "["
+        number [227,231] "1357"
+        word [231,232] "]"
+      ()) [232,233] ")"
+      command [238,245] "exit $?"
+        command_name [238,242] "exit"
+          word [238,242] "exit"
+        simple_expansion [243,245] "$?"
+          ($) [243,244] "$"
+          special_variable_name [244,245] "?"
+      (;;) [250,252] ";;"
+    (esac) [253,257] "esac"
+  case_statement [259,289] "case x in x) echo meow ;; esac"
+    (case) [259,263] "case"
+    word [264,265] "x"
+    (in) [266,268] "in"
+    case_item [269,284] "x) echo meow ;;"
+      word [269,270] "x"
+      ()) [270,271] ")"
+      command [272,281] "echo meow"
+        command_name [272,276] "echo"
+          word [272,276] "echo"
+        word [277,281] "meow"
+      (;;) [282,284] ";;"
+    (esac) [285,289] "esac"
+  case_statement [291,324] "case foo in\n  bar\\ baz) : ;;\nesac"
+    (case) [291,295] "case"
+    word [296,299] "foo"
+    (in) [300,302] "in"
+    case_item [305,319] "bar\\ baz) : ;;"
+      word [305,313] "bar\\ baz"
+      ()) [313,314] ")"
+      command [315,316] ":"
+        command_name [315,316] ":"
+          word [315,316] ":"
+      (;;) [317,319] ";;"
+    (esac) [320,324] "esac"
+  case_statement [326,376] "case \"$arg\" in\n  *([0-9])([0-9])) ech..."
+    (case) [326,330] "case"
+    string [331,337] "\"$arg\""
+      (") [331,332] "\""
+      simple_expansion [332,336] "$arg"
+        ($) [332,333] "$"
+        variable_name [333,336] "arg"
+      (") [336,337] "\""
+    (in) [338,340] "in"
+    case_item [343,372] "*([0-9])([0-9])) echo \"$arg\"\n"
+      extglob_pattern [343,358] "*([0-9])([0-9])"
+      ()) [358,359] ")"
+      command [360,371] "echo \"$arg\""
+        command_name [360,364] "echo"
+          word [360,364] "echo"
+        string [365,371] "\"$arg\""
+          (") [365,366] "\""
+          simple_expansion [366,370] "$arg"
+            ($) [366,367] "$"
+            variable_name [367,370] "arg"
+          (") [370,371] "\""
+    (esac) [372,376] "esac"
+  case_statement [378,478] "case ${lang} in\nCMakeLists.txt | \\\n\tc..."
+    (case) [378,382] "case"
+    expansion [383,390] "${lang}"
+      (${) [383,385] "${"
+      variable_name [385,389] "lang"
+      (}) [389,390] "}"
+    (in) [391,393] "in"
+    case_item [394,446] "CMakeLists.txt | \\\n\tcmake_modules | \\..."
+      word [394,408] "CMakeLists.txt"
+      (|) [409,410] "|"
+      extglob_pattern [414,427] "cmake_modules"
+      (|) [428,429] "|"
+      concatenation [433,442] "${PN}.pot"
+        expansion [433,438] "${PN}"
+          (${) [433,435] "${"
+          variable_name [435,437] "PN"
+          (}) [437,438] "}"
+        word [438,442] ".pot"
+      ()) [442,443] ")"
+      (;;) [444,446] ";;"
+    case_item [447,473] "*) rm -r ${lang} || die ;;"
+      extglob_pattern [447,448] "*"
+      ()) [448,449] ")"
+      list [450,470] "rm -r ${lang} || die"
+        command [450,463] "rm -r ${lang}"
+          command_name [450,452] "rm"
+            word [450,452] "rm"
+          word [453,455] "-r"
+          expansion [456,463] "${lang}"
+            (${) [456,458] "${"
+            variable_name [458,462] "lang"
+            (}) [462,463] "}"
+        (||) [464,466] "||"
+        command [467,470] "die"
+          command_name [467,470] "die"
+            word [467,470] "die"
+      (;;) [471,473] ";;"
+    (esac) [474,478] "esac"
+===
+@known-diff cfor-compound-assign-negative: corpus statements.txt::C-style for statements
+for (( c=1; c<=5; c++ ))
+do
+  echo $c
+done
+
+for (( c=1; c<=5; c++ )) {
+	echo $c
+}
+
+for (( ; ; ))
+do
+  echo 'forever'
+done
+
+for ((cx = 0; c = $cx / $pf, c < $wc - $k; )); do
+		echo "$cx"
+done
+
+for (( i = 4;;i--)) ; do echo $i; if (( $i == 0 )); then break; fi; done
+
+# added post-bash-4.2
+for (( i = j = k = 1; i % 9 || (j *= -1, $( ((i%9)) || printf " " >&2; echo 0), k++ <= 10); i += j ))
+do
+printf "$i"
+done
+
+echo
+
+( for (( i = j = k = 1; i % 9 || (j *= -1, $( ((i%9)) || printf " " >&2; echo 0), k++ <= 10); i += j ))
+do
+printf "$i"
+done )
+---
+hasError: false
+program [0,542] "for (( c=1; c<=5; c++ ))\ndo\n  echo $c..."
+  c_style_for_statement [0,42] "for (( c=1; c<=5; c++ ))\ndo\n  echo $c..."
+    (for) [0,3] "for"
+    ((() [4,6] "(("
+    variable_assignment [7,10] "c=1"
+      variable_name [7,8] "c"
+      (=) [8,9] "="
+      number [9,10] "1"
+    (;) [10,11] ";"
+    binary_expression [12,16] "c<=5"
+      word [12,13] "c"
+      (<=) [13,15] "<="
+      number [15,16] "5"
+    (;) [16,17] ";"
+    postfix_expression [18,21] "c++"
+      word [18,19] "c"
+      (++) [19,21] "++"
+    ())) [22,24] "))"
+    do_group [25,42] "do\n  echo $c\ndone"
+      (do) [25,27] "do"
+      command [30,37] "echo $c"
+        command_name [30,34] "echo"
+          word [30,34] "echo"
+        simple_expansion [35,37] "$c"
+          ($) [35,36] "$"
+          variable_name [36,37] "c"
+      (done) [38,42] "done"
+  c_style_for_statement [44,81] "for (( c=1; c<=5; c++ )) {\n\techo $c\n}"
+    (for) [44,47] "for"
+    ((() [48,50] "(("
+    variable_assignment [51,54] "c=1"
+      variable_name [51,52] "c"
+      (=) [52,53] "="
+      number [53,54] "1"
+    (;) [54,55] ";"
+    binary_expression [56,60] "c<=5"
+      word [56,57] "c"
+      (<=) [57,59] "<="
+      number [59,60] "5"
+    (;) [60,61] ";"
+    postfix_expression [62,65] "c++"
+      word [62,63] "c"
+      (++) [63,65] "++"
+    ())) [66,68] "))"
+    compound_statement [69,81] "{\n\techo $c\n}"
+      ({) [69,70] "{"
+      command [72,79] "echo $c"
+        command_name [72,76] "echo"
+          word [72,76] "echo"
+        simple_expansion [77,79] "$c"
+          ($) [77,78] "$"
+          variable_name [78,79] "c"
+      (}) [80,81] "}"
+  c_style_for_statement [83,121] "for (( ; ; ))\ndo\n  echo 'forever'\ndone"
+    (for) [83,86] "for"
+    ((() [87,89] "(("
+    (;) [90,91] ";"
+    (;) [92,93] ";"
+    ())) [94,96] "))"
+    do_group [97,121] "do\n  echo 'forever'\ndone"
+      (do) [97,99] "do"
+      command [102,116] "echo 'forever'"
+        command_name [102,106] "echo"
+          word [102,106] "echo"
+        raw_string [107,116] "'forever'"
+      (done) [117,121] "done"
+  c_style_for_statement [123,190] "for ((cx = 0; c = $cx / $pf, c < $wc ..."
+    (for) [123,126] "for"
+    ((() [127,129] "(("
+    variable_assignment [129,135] "cx = 0"
+      variable_name [129,131] "cx"
+      (=) [132,133] "="
+      number [134,135] "0"
+    (;) [135,136] ";"
+    variable_assignment [137,150] "c = $cx / $pf"
+      variable_name [137,138] "c"
+      (=) [139,140] "="
+      binary_expression [141,150] "$cx / $pf"
+        simple_expansion [141,144] "$cx"
+          ($) [141,142] "$"
+          variable_name [142,144] "cx"
+        (/) [145,146] "/"
+        simple_expansion [147,150] "$pf"
+          ($) [147,148] "$"
+          variable_name [148,150] "pf"
+    (,) [150,151] ","
+    binary_expression [152,164] "c < $wc - $k"
+      word [152,153] "c"
+      (<) [154,155] "<"
+      binary_expression [156,164] "$wc - $k"
+        simple_expansion [156,159] "$wc"
+          ($) [156,157] "$"
+          variable_name [157,159] "wc"
+        (-) [160,161] "-"
+        simple_expansion [162,164] "$k"
+          ($) [162,163] "$"
+          variable_name [163,164] "k"
+    (;) [164,165] ";"
+    ())) [166,168] "))"
+    (;) [168,169] ";"
+    do_group [170,190] "do\n\t\techo \"$cx\"\ndone"
+      (do) [170,172] "do"
+      command [175,185] "echo \"$cx\""
+        command_name [175,179] "echo"
+          word [175,179] "echo"
+        string [180,185] "\"$cx\""
+          (") [180,181] "\""
+          simple_expansion [181,184] "$cx"
+            ($) [181,182] "$"
+            variable_name [182,184] "cx"
+          (") [184,185] "\""
+      (done) [186,190] "done"
+  c_style_for_statement [192,264] "for (( i = 4;;i--)) ; do echo $i; if ..."
+    (for) [192,195] "for"
+    ((() [196,198] "(("
+    variable_assignment [199,204] "i = 4"
+      variable_name [199,200] "i"
+      (=) [201,202] "="
+      number [203,204] "4"
+    (;) [204,205] ";"
+    (;) [205,206] ";"
+    postfix_expression [206,209] "i--"
+      word [206,207] "i"
+      (--) [207,209] "--"
+    ())) [209,211] "))"
+    (;) [212,213] ";"
+    do_group [214,264] "do echo $i; if (( $i == 0 )); then br..."
+      (do) [214,216] "do"
+      command [217,224] "echo $i"
+        command_name [217,221] "echo"
+          word [217,221] "echo"
+        simple_expansion [222,224] "$i"
+          ($) [222,223] "$"
+          variable_name [223,224] "i"
+      (;) [224,225] ";"
+      if_statement [226,258] "if (( $i == 0 )); then break; fi"
+        (if) [226,228] "if"
+        command [229,242] "(( $i == 0 ))"
+          command_name [229,242] "(( $i == 0 ))"
+            arithmetic_expansion [229,242] "(( $i == 0 ))"
+              ((() [229,231] "(("
+              binary_expression [232,239] "$i == 0"
+                simple_expansion [232,234] "$i"
+                  ($) [232,233] "$"
+                  variable_name [233,234] "i"
+                (==) [235,237] "=="
+                number [238,239] "0"
+              ())) [240,242] "))"
+        (;) [242,243] ";"
+        (then) [244,248] "then"
+        command [249,254] "break"
+          command_name [249,254] "break"
+            word [249,254] "break"
+        (;) [254,255] ";"
+        (fi) [256,258] "fi"
+      (;) [258,259] ";"
+      (done) [260,264] "done"
+  comment [266,287] "# added post-bash-4.2"
+  c_style_for_statement [288,409] "for (( i = j = k = 1; i % 9 || (j *= ..."
+    (for) [288,291] "for"
+    ((() [292,294] "(("
+    variable_assignment [295,308] "i = j = k = 1"
+      variable_name [295,296] "i"
+      (=) [297,298] "="
+      variable_assignment [299,308] "j = k = 1"
+        variable_name [299,300] "j"
+        (=) [301,302] "="
+        variable_assignment [303,308] "k = 1"
+          variable_name [303,304] "k"
+          (=) [305,306] "="
+          number [307,308] "1"
+    (;) [308,309] ";"
+    binary_expression [310,378] "i % 9 || (j *= -1, $( ((i%9)) || prin..."
+      binary_expression [310,315] "i % 9"
+        word [310,311] "i"
+        (%) [312,313] "%"
+        number [314,315] "9"
+      (||) [316,318] "||"
+      parenthesized_expression [319,378] "(j *= -1, $( ((i%9)) || printf \" \" >&..."
+        (() [319,320] "("
+        binary_expression [320,327] "j *= -1"
+          word [320,321] "j"
+          (*=) [322,324] "*="
+          unary_expression [325,327] "-1"
+            (-) [325,326] "-"
+            number [326,327] "1"
+        (,) [327,328] ","
+        command_substitution [329,366] "$( ((i%9)) || printf \" \" >&2; echo 0)"
+          ($() [329,331] "$("
+          list [332,357] "((i%9)) || printf \" \" >&2"
+            command [332,339] "((i%9))"
+              command_name [332,339] "((i%9))"
+                arithmetic_expansion [332,339] "((i%9))"
+                  ((() [332,334] "(("
+                  binary_expression [334,337] "i%9"
+                    variable_name [334,335] "i"
+                    (%) [335,336] "%"
+                    number [336,337] "9"
+                  ())) [337,339] "))"
+            (||) [340,342] "||"
+            redirected_statement [343,357] "printf \" \" >&2"
+              command [343,353] "printf \" \""
+                command_name [343,349] "printf"
+                  word [343,349] "printf"
+                string [350,353] "\" \""
+                  (") [350,351] "\""
+                  string_content [351,352] " "
+                  (") [352,353] "\""
+              file_redirect [354,357] ">&2"
+                (>&) [354,356] ">&"
+                number [356,357] "2"
+          (;) [357,358] ";"
+          command [359,365] "echo 0"
+            command_name [359,363] "echo"
+              word [359,363] "echo"
+            number [364,365] "0"
+          ()) [365,366] ")"
+        (,) [366,367] ","
+        binary_expression [368,377] "k++ <= 10"
+          postfix_expression [368,371] "k++"
+            word [368,369] "k"
+            (++) [369,371] "++"
+          (<=) [372,374] "<="
+          number [375,377] "10"
+        ()) [377,378] ")"
+    (;) [378,379] ";"
+    binary_expression [380,386] "i += j"
+      word [380,381] "i"
+      (+=) [382,384] "+="
+      word [385,386] "j"
+    ())) [387,389] "))"
+    do_group [390,409] "do\nprintf \"$i\"\ndone"
+      (do) [390,392] "do"
+      command [393,404] "printf \"$i\""
+        command_name [393,399] "printf"
+          word [393,399] "printf"
+        string [400,404] "\"$i\""
+          (") [400,401] "\""
+          simple_expansion [401,403] "$i"
+            ($) [401,402] "$"
+            variable_name [402,403] "i"
+          (") [403,404] "\""
+      (done) [405,409] "done"
+  command [411,415] "echo"
+    command_name [411,415] "echo"
+      word [411,415] "echo"
+  subshell [417,542] "( for (( i = j = k = 1; i % 9 || (j *..."
+    (() [417,418] "("
+    c_style_for_statement [419,540] "for (( i = j = k = 1; i % 9 || (j *= ..."
+      (for) [419,422] "for"
+      ((() [423,425] "(("
+      variable_assignment [426,439] "i = j = k = 1"
+        variable_name [426,427] "i"
+        (=) [428,429] "="
+        variable_assignment [430,439] "j = k = 1"
+          variable_name [430,431] "j"
+          (=) [432,433] "="
+          variable_assignment [434,439] "k = 1"
+            variable_name [434,435] "k"
+            (=) [436,437] "="
+            number [438,439] "1"
+      (;) [439,440] ";"
+      binary_expression [441,509] "i % 9 || (j *= -1, $( ((i%9)) || prin..."
+        binary_expression [441,446] "i % 9"
+          word [441,442] "i"
+          (%) [443,444] "%"
+          number [445,446] "9"
+        (||) [447,449] "||"
+        parenthesized_expression [450,509] "(j *= -1, $( ((i%9)) || printf \" \" >&..."
+          (() [450,451] "("
+          binary_expression [451,458] "j *= -1"
+            word [451,452] "j"
+            (*=) [453,455] "*="
+            unary_expression [456,458] "-1"
+              (-) [456,457] "-"
+              number [457,458] "1"
+          (,) [458,459] ","
+          command_substitution [460,497] "$( ((i%9)) || printf \" \" >&2; echo 0)"
+            ($() [460,462] "$("
+            list [463,488] "((i%9)) || printf \" \" >&2"
+              command [463,470] "((i%9))"
+                command_name [463,470] "((i%9))"
+                  arithmetic_expansion [463,470] "((i%9))"
+                    ((() [463,465] "(("
+                    binary_expression [465,468] "i%9"
+                      variable_name [465,466] "i"
+                      (%) [466,467] "%"
+                      number [467,468] "9"
+                    ())) [468,470] "))"
+              (||) [471,473] "||"
+              redirected_statement [474,488] "printf \" \" >&2"
+                command [474,484] "printf \" \""
+                  command_name [474,480] "printf"
+                    word [474,480] "printf"
+                  string [481,484] "\" \""
+                    (") [481,482] "\""
+                    string_content [482,483] " "
+                    (") [483,484] "\""
+                file_redirect [485,488] ">&2"
+                  (>&) [485,487] ">&"
+                  number [487,488] "2"
+            (;) [488,489] ";"
+            command [490,496] "echo 0"
+              command_name [490,494] "echo"
+                word [490,494] "echo"
+              number [495,496] "0"
+            ()) [496,497] ")"
+          (,) [497,498] ","
+          binary_expression [499,508] "k++ <= 10"
+            postfix_expression [499,502] "k++"
+              word [499,500] "k"
+              (++) [500,502] "++"
+            (<=) [503,505] "<="
+            number [506,508] "10"
+          ()) [508,509] ")"
+      (;) [509,510] ";"
+      binary_expression [511,517] "i += j"
+        word [511,512] "i"
+        (+=) [513,515] "+="
+        word [516,517] "j"
+      ())) [518,520] "))"
+      do_group [521,540] "do\nprintf \"$i\"\ndone"
+        (do) [521,523] "do"
+        command [524,535] "printf \"$i\""
+          command_name [524,530] "printf"
+            word [524,530] "printf"
+          string [531,535] "\"$i\""
+            (") [531,532] "\""
+            simple_expansion [532,534] "$i"
+              ($) [532,533] "$"
+              variable_name [533,534] "i"
+            (") [534,535] "\""
+        (done) [536,540] "done"
+    ()) [541,542] ")"
+===
+===
+@known-diff case-short-option-pattern: corpus statements.txt::Test command paren statefulness with a case glob
+[[ ${test} == @($(IFS='|'; echo "${skip[*]}")) ]]
+
+case ${out} in
+*"not supported"*|\
+*"operation not supported"*)
+	;;
+esac
+
+case $1 in
+-o)
+	owner=$2
+	shift
+	;;
+-g) ;;
+esac
+
+[[ a == \"+(?)\" ]]
+---
+hasError: false
+program [0,193] "[[ ${test} == @($(IFS='|'; echo \"${sk..."
+  test_command [0,49] "[[ ${test} == @($(IFS='|'; echo \"${sk..."
+    ([[) [0,2] "[["
+    binary_expression [3,46] "${test} == @($(IFS='|'; echo \"${skip[..."
+      expansion [3,10] "${test}"
+        (${) [3,5] "${"
+        variable_name [5,9] "test"
+        (}) [9,10] "}"
+      (==) [11,13] "=="
+      extglob_pattern [14,16] "@("
+      command_substitution [16,45] "$(IFS='|'; echo \"${skip[*]}\")"
+        ($() [16,18] "$("
+        variable_assignment [18,25] "IFS='|'"
+          variable_name [18,21] "IFS"
+          (=) [21,22] "="
+          raw_string [22,25] "'|'"
+        (;) [25,26] ";"
+        command [27,44] "echo \"${skip[*]}\""
+          command_name [27,31] "echo"
+            word [27,31] "echo"
+          string [32,44] "\"${skip[*]}\""
+            (") [32,33] "\""
+            expansion [33,43] "${skip[*]}"
+              (${) [33,35] "${"
+              subscript [35,42] "skip[*]"
+                variable_name [35,39] "skip"
+                ([) [39,40] "["
+                word [40,41] "*"
+                (]) [41,42] "]"
+              (}) [42,43] "}"
+            (") [43,44] "\""
+        ()) [44,45] ")"
+      extglob_pattern [45,46] ")"
+    (]]) [47,49] "]]"
+  case_statement [51,123] "case ${out} in\n*\"not supported\"*|\\\n*\"..."
+    (case) [51,55] "case"
+    expansion [56,62] "${out}"
+      (${) [56,58] "${"
+      variable_name [58,61] "out"
+      (}) [61,62] "}"
+    (in) [63,65] "in"
+    case_item [66,118] "*\"not supported\"*|\\\n*\"operation not s..."
+      extglob_pattern [66,67] "*"
+      string [67,82] "\"not supported\""
+        (") [67,68] "\""
+        string_content [68,81] "not supported"
+        (") [81,82] "\""
+      extglob_pattern [82,83] "*"
+      (|) [83,84] "|"
+      concatenation [86,113] "*\"operation not supported\"*"
+        word [86,87] "*"
+        string [87,112] "\"operation not supported\""
+          (") [87,88] "\""
+          string_content [88,111] "operation not supported"
+          (") [111,112] "\""
+        word [112,113] "*"
+      ()) [113,114] ")"
+      (;;) [116,118] ";;"
+    (esac) [119,123] "esac"
+  case_statement [125,172] "case $1 in\n-o)\n\towner=$2\n\tshift\n\t;;\n-..."
+    (case) [125,129] "case"
+    simple_expansion [130,132] "$1"
+      ($) [130,131] "$"
+      variable_name [131,132] "1"
+    (in) [133,135] "in"
+    case_item [136,160] "-o)\n\towner=$2\n\tshift\n\t;;"
+      word [136,138] "-o"
+      ()) [138,139] ")"
+      variable_assignment [141,149] "owner=$2"
+        variable_name [141,146] "owner"
+        (=) [146,147] "="
+        simple_expansion [147,149] "$2"
+          ($) [147,148] "$"
+          variable_name [148,149] "2"
+      command [151,156] "shift"
+        command_name [151,156] "shift"
+          word [151,156] "shift"
+      (;;) [158,160] ";;"
+    case_item [161,167] "-g) ;;"
+      word [161,163] "-g"
+      ()) [163,164] ")"
+      (;;) [165,167] ";;"
+    (esac) [168,172] "esac"
+  test_command [174,193] "[[ a == \\\"+(?)\\\" ]]"
+    ([[) [174,176] "[["
+    binary_expression [177,190] "a == \\\"+(?)\\\""
+      word [177,178] "a"
+      (==) [179,181] "=="
+      extglob_pattern [182,190] "\\\"+(?)\\\""
+    (]]) [191,193] "]]"
+
+===
+@known-diff dollar-backtick-substitution: corpus statements.txt::Command substution with $ and backticks
+$(eval echo $`echo ${foo}`)
+---
+hasError: false
+program [0,27] "$(eval echo $`echo ${foo}`)"
+  command [0,27] "$(eval echo $`echo ${foo}`)"
+    command_name [0,27] "$(eval echo $`echo ${foo}`)"
+      command_substitution [0,27] "$(eval echo $`echo ${foo}`)"
+        ($() [0,2] "$("
+        command [2,26] "eval echo $`echo ${foo}`"
+          command_name [2,6] "eval"
+            word [2,6] "eval"
+          word [7,11] "echo"
+          concatenation [12,26] "$`echo ${foo}`"
+            ($) [12,13] "$"
+            command_substitution [13,26] "`echo ${foo}`"
+              (`) [13,14] "`"
+              command [14,25] "echo ${foo}"
+                command_name [14,18] "echo"
+                  word [14,18] "echo"
+                expansion [19,25] "${foo}"
+                  (${) [19,21] "${"
+                  variable_name [21,24] "foo"
+                  (}) [24,25] "}"
+              (`) [25,26] "`"
+        ()) [26,27] ")"
+===
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/literals.txt b/packages/tree-sitter-bash/test/fixtures/corpus/literals.txt
new file mode 100644
index 0000000000..61543ca3b4
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/literals.txt
@@ -0,0 +1,1371 @@
+================================================================================
+Literal words
+================================================================================
+
+echo a
+echo a b
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (word))
+  (command
+    (command_name
+      (word))
+    (word)
+    (word)))
+
+================================================================================
+Words with special characters
+================================================================================
+
+echo {o[k]}
+echo }}}
+echo ]]] ===
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (word)
+      (word)
+      (word)
+      (word)
+      (word)
+      (word)))
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (word)
+      (word)
+      (word)))
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (word)
+      (word)
+      (word))
+    (word)))
+
+================================================================================
+Simple variable expansions
+================================================================================
+
+echo $abc
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (simple_expansion
+      (variable_name))))
+
+================================================================================
+Special variable expansions
+================================================================================
+
+echo $# $* $@ $!
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (simple_expansion
+      (special_variable_name))
+    (simple_expansion
+      (special_variable_name))
+    (simple_expansion
+      (special_variable_name))
+    (simple_expansion
+      (special_variable_name))))
+
+================================================================================
+Variable expansions
+================================================================================
+
+echo ${}
+echo ${#}
+echo ${var1#*#}
+echo ${!abc}
+echo ${abc}
+echo ${abc:-def}
+echo ${abc:+ghi}
+echo ${abc:- }
+echo ${abc:
+}
+echo ${abc,?}
+echo ${abc^^b}
+echo ${abc@U}
+echo ${abc:- -quiet}
+
+--------------------------------------------------------------------------------
+
+(program
+  (command (command_name (word)) (expansion))
+  (command (command_name (word)) (expansion))
+  (command (command_name (word)) (expansion (variable_name) (regex)))
+  (command (command_name (word)) (expansion (variable_name)))
+  (command (command_name (word)) (expansion (variable_name)))
+  (command (command_name (word)) (expansion (variable_name) (word)))
+  (command (command_name (word)) (expansion (variable_name) (word)))
+  (command (command_name (word)) (expansion (variable_name) (word)))
+  (command (command_name (word)) (expansion (variable_name)))
+  (command (command_name (word)) (expansion (variable_name) (regex)))
+  (command (command_name (word)) (expansion (variable_name) (regex)))
+  (command (command_name (word)) (expansion (variable_name)))
+  (command (command_name (word)) (expansion (variable_name) (word))))
+
+================================================================================
+Variable expansions with operators
+================================================================================
+
+A="${B[0]# }"
+C="${D/#* -E /}"
+F="${G%% *}"
+H="${I#*;}"
+J="${K##*;}"
+L="${M%;*}"
+N="${O%%;*}"
+P="${Q%|*}"
+R="${S%()}"
+T="${U%(}"
+V="${W%)}"
+X="${Y%<}"
+Z="${A#*<B>}"
+C="${D%</E>*}"
+F="${#!}"
+G=${H,,[I]}
+J=${K^^[L]}
+L="${M/'N'*/O}"
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (subscript
+          (variable_name)
+          (number))
+            (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion)))
+  (variable_assignment
+    (variable_name)
+    (expansion
+      (variable_name)
+      (regex)))
+  (variable_assignment
+    (variable_name)
+    (expansion
+      (variable_name)
+      (regex)))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex)
+        (word)))))
+
+================================================================================
+More Variable expansions with operators
+================================================================================
+
+${parameter-default}
+${parameter:-default}
+${parameter=default}
+${parameter:=default}
+${parameter+alt_value}
+${parameter:+alt_value}
+${parameter?err_msg}
+${parameter:?err_msg}
+${var%Pattern}
+${var%%Pattern}
+${var:pos}
+${var:pos:len}
+${MATRIX:$(($RANDOM%${#MATRIX})):1}
+${PKG_CONFIG_LIBDIR:-${ESYSROOT}/usr/$(get_libdir)/pkgconfig}
+${ver_str::${#ver_str}-${#not_match}}
+${value#\{sd.cicd.}
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (regex))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (regex))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (variable_name))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (variable_name)
+        (variable_name))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (arithmetic_expansion
+          (binary_expression
+            (simple_expansion
+              (variable_name))
+            (expansion
+              (variable_name))))
+        (number))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (concatenation
+          (expansion (variable_name))
+          (word)
+          (command_substitution
+            (command (command_name (word))))
+          (word)))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (binary_expression
+          (expansion (variable_name))
+          (expansion (variable_name))))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (regex)))))
+
+================================================================================
+Variable expansions in strings
+================================================================================
+
+A="${A:-$B/c}"
+A="${b=$c/$d}"
+MY_PV="${PV/_pre/$'\x7e'pre}"
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (concatenation
+          (simple_expansion
+            (variable_name))
+          (word)))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (concatenation
+          (simple_expansion
+            (variable_name))
+          (word)
+          (simple_expansion
+            (variable_name))))))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex)
+        (concatenation (ansi_c_string) (word))))))
+
+================================================================================
+Variable expansions with regexes
+================================================================================
+
+A=${B//:;;/$'\n'}
+
+# escaped space
+C=${D/;\ *;|}
+MOFILES=${LINGUAS// /.po }.po
+MY_P="${PN/aspell/aspell"${ASPELL_VERSION}"}"
+pyc=${pyc//*\/}
+${pv/\.}
+${new_test_cp//"${old_ver_cp}"/}
+${tests_to_run//"${classes}"\/}
+${allarchives// /\\|}
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment
+    (variable_name)
+    (expansion
+      (variable_name)
+      (regex)
+      (ansi_c_string)))
+  (comment)
+  (variable_assignment
+    (variable_name)
+    (expansion (variable_name) (regex)))
+  (variable_assignment
+    (variable_name)
+    (concatenation
+      (expansion (variable_name) (regex) (word))
+      (word)))
+  (variable_assignment
+    (variable_name)
+    (string
+      (expansion
+        (variable_name)
+        (regex)
+        (concatenation
+          (word)
+          (string (expansion (variable_name)))))))
+  (variable_assignment
+    (variable_name)
+    (expansion (variable_name) (regex)))
+  (command (command_name (expansion (variable_name) (regex))))
+  (command
+    (command_name
+      (expansion (variable_name) (string (expansion (variable_name))))))
+  (command
+    (command_name
+      (expansion (variable_name) (string (expansion (variable_name))) (regex))))
+  (command (command_name (expansion (variable_name) (regex) (word)))))
+
+================================================================================
+Other variable expansion operators
+================================================================================
+
+cat ${BAR} ${ABC=def} ${GHI:?jkl}
+[ "$a" != "${a#[Bc]}" ]
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (expansion
+      (variable_name))
+    (expansion
+      (variable_name)
+      (word))
+    (expansion
+      (variable_name)
+      (word)))
+  (test_command
+    (binary_expression
+      (string
+        (simple_expansion
+          (variable_name)))
+      (string
+        (expansion
+          (variable_name)
+          (regex))))))
+
+================================================================================
+Variable Expansions: Length
+================================================================================
+
+${parameter:-1}
+
+${parameter: -1}
+
+${parameter:(-1)}
+
+${matrix:$(($random%${#matrix})):1}
+
+"${_component_to_single:${len}:2}"
+
+"${PN::-1}"
+
+${trarr:$(ver_cut 2):1}
+
+${comp[@]:start:end*2-start}
+
+--------------------------------------------------------------------------------
+
+(program
+  (command (command_name (expansion (variable_name) (word))))
+  (command (command_name (expansion (variable_name) (number))))
+  (command
+    (command_name
+      (expansion (variable_name) (parenthesized_expression (unary_expression (number))))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (arithmetic_expansion
+          (binary_expression (simple_expansion (variable_name)) (expansion (variable_name))))
+        (number))))
+  (command (command_name (string (expansion (variable_name) (expansion (variable_name)) (number)))))
+  (command (command_name (string (expansion (variable_name) (number)))))
+  (command
+    (command_name
+      (expansion (variable_name) (command_substitution (command (command_name (word)) (number))) (number))))
+  (command
+    (command_name
+      (expansion
+        (subscript (variable_name) (word))
+        (variable_name)
+        (binary_expression (binary_expression (variable_name) (number)) (variable_name))))))
+
+================================================================================
+Variable Expansions with operators
+================================================================================
+
+${parameter-default}
+${parameter- default}
+${!varprefix*}
+${!varprefix@}
+${parameter@U}
+
+--------------------------------------------------------------------------------
+
+(program
+  (command (command_name (expansion (variable_name) (word))))
+  (command (command_name (expansion (variable_name) (word))))
+  (command (command_name (expansion (variable_name))))
+  (command (command_name (expansion (variable_name))))
+  (command (command_name (expansion (variable_name)))))
+
+================================================================================
+Variable Expansions: Bizarre Cases
+================================================================================
+
+${!#}
+${!# }
+${!##}
+${!## }
+${!##/}
+# here be dragons
+echo "${kw}? ( ${cond:+${cond}? (} ${baseuri}-${ver}-${kw}.${suff} ${cond:+) })"
+
+--------------------------------------------------------------------------------
+
+(program
+  (command (command_name (expansion)))
+  (command (command_name (expansion)))
+  (command (command_name (expansion)))
+  (command (command_name (expansion)))
+  (command (command_name (expansion (special_variable_name) (regex))))
+  (comment)
+  (command
+    (command_name (word))
+    (string
+      (expansion (variable_name))
+      (string_content)
+      (expansion
+        (variable_name)
+        (concatenation (expansion (variable_name)) (word)))
+      (expansion (variable_name))
+      (string_content)
+      (expansion (variable_name))
+      (string_content)
+      (expansion (variable_name))
+      (string_content)
+      (expansion (variable_name))
+      (expansion (variable_name) (word))
+      (string_content))))
+
+================================================================================
+Variable Expansions: Weird Cases
+================================================================================
+
+${completions[*]}
+${=1}
+${2?}
+${p_key#*=}
+${abc:- }
+${B[0]# }
+${to_enables[0]##*/}
+exec "${0#-}" --rcfile "${BASH_IT_BASHRC:-${HOME?}/.bashrc}"
+recho "TDEFAULTS = ${selvecs:+-DSELECT_VECS=\"$selvecs\"}"
+local msg="${2:-command '$1' does not exist}"
+${cdir:+#}
+${dict_langs:+;}
+${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/util-linux )}
+${id}${2+ ${2}}
+${BRANDING_GCC_PKGVERSION/(/(Gentoo ${PVR}${extvers}, } # look at that parenthesis!
+some-command ${foo:+--arg <(printf '%s\n' "$foo")}
+
+--------------------------------------------------------------------------------
+
+(program
+  (command (command_name (expansion (subscript (variable_name) (word)))))
+  (command (command_name (expansion (variable_name))))
+  (command (command_name (expansion (variable_name))))
+  (command (command_name (expansion (variable_name) (regex))))
+  (command (command_name (expansion (variable_name) (word))))
+  (command (command_name (expansion (subscript (variable_name) (number)) (regex))))
+  (command (command_name (expansion (subscript (variable_name) (number)) (regex))))
+  (command
+    (command_name (word))
+    (string (expansion (special_variable_name) (regex)))
+    (word)
+    (string (expansion (variable_name) (concatenation (expansion (variable_name)) (word)))))
+  (command
+    (command_name (word))
+    (string
+      (string_content)
+      (expansion (variable_name) (concatenation (word) (simple_expansion (variable_name)) (word)))))
+  (declaration_command
+    (variable_assignment
+      (variable_name)
+      (string (expansion (variable_name) (concatenation (word) (raw_string) (word))))))
+  (command (command_name (expansion (variable_name) (word))))
+  (command (command_name (expansion (variable_name) (word))))
+  (command (command_name (expansion (subscript (variable_name) (word)) (word))))
+  (command
+    (command_name
+      (concatenation (expansion (variable_name)) (expansion (variable_name) (expansion (variable_name))))))
+  (command
+    (command_name
+      (expansion
+        (variable_name)
+        (regex)
+        (concatenation (word) (expansion (variable_name)) (expansion (variable_name)) (word)))))
+  (comment)
+      (command
+        (command_name (word))
+        (expansion
+          (variable_name)
+          (concatenation
+            (word)
+            (process_substitution
+              (command
+                (command_name (word))
+                (raw_string)
+                (string (simple_expansion (variable_name)))))))))
+
+================================================================================
+Variable Expansions: Regex
+================================================================================
+
+A=${B//:;;/$'\n'}
+C="${D/#* -E /}"
+BASH_IT_GIT_URL="${BASH_IT_GIT_URL/git@/https://}"
+10#${command_start##*.}
+echo ${LIB_DEPEND//\[static-libs(+)]}
+${ALL_LLVM_TARGETS[@]/%/(-)?}
+filterdiff -p1 ${paths[@]/#/-i }
+${cflags//-O? /$(get-flag O) }
+curf="${f%'-roff2html'*}.html"
+reff="${f/'-roff2html'*/'-ref'}.html"
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment (variable_name) (expansion (variable_name) (regex) (ansi_c_string)))
+  (variable_assignment (variable_name) (string (expansion (variable_name) (regex))))
+  (variable_assignment (variable_name) (string (expansion (variable_name) (regex) (word))))
+  (command (command_name (number (expansion (variable_name) (regex)))))
+  (command (command_name (word)) (expansion (variable_name) (regex)))
+  (command (command_name (expansion (subscript (variable_name) (word)) (word))))
+  (command (command_name (word)) (word) (expansion (subscript (variable_name) (word)) (word)))
+  (command
+    (command_name
+      (expansion (variable_name) (regex) (command_substitution (command (command_name (word)) (word))) (word))))
+  (variable_assignment
+    (variable_name)
+    (string (expansion (variable_name) (raw_string) (regex)) (string_content)))
+  (variable_assignment
+    (variable_name)
+    (string (expansion (variable_name) (regex) (raw_string)) (string_content))))
+
+================================================================================
+Words ending with '$'
+================================================================================
+
+grep ^${var}$
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (word)
+      (expansion
+        (variable_name)))))
+
+================================================================================
+Command substitutions
+================================================================================
+
+echo `echo hi`
+echo `echo hi; echo there`
+echo $(echo $(echo hi))
+echo $(< some-file)
+
+# both of these are concatenations!
+echo `echo otherword`word
+echo word`echo otherword`
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (command_substitution
+      (command
+        (command_name
+          (word))
+        (word))))
+  (command
+    (command_name
+      (word))
+    (command_substitution
+      (command
+        (command_name
+          (word))
+        (word))
+      (command
+        (command_name
+          (word))
+        (word))))
+  (command
+    (command_name
+      (word))
+    (command_substitution
+      (command
+        (command_name
+          (word))
+        (command_substitution
+          (command
+            (command_name
+              (word))
+            (word))))))
+  (command
+    (command_name
+      (word))
+    (command_substitution
+      (file_redirect
+        (word))))
+  (comment)
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (command_substitution
+        (command
+          (command_name
+            (word))
+          (word)))
+      (word)))
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (word)
+      (command_substitution
+        (command
+          (command_name
+            (word))
+          (word))))))
+
+================================================================================
+Process substitutions
+================================================================================
+
+wc -c <(echo abc && echo def)
+wc -c <(echo abc; echo def)
+echo abc > >(wc -c)
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (word)
+    (process_substitution
+      (list
+        (command
+          (command_name
+            (word))
+          (word))
+        (command
+          (command_name
+            (word))
+          (word)))))
+  (command
+    (command_name
+      (word))
+    (word)
+    (process_substitution
+      (command
+        (command_name
+          (word))
+        (word))
+      (command
+        (command_name
+          (word))
+        (word))))
+  (redirected_statement
+    (command
+      (command_name
+        (word))
+      (word))
+    (file_redirect
+      (process_substitution
+        (command
+          (command_name
+            (word))
+          (word))))))
+
+================================================================================
+Single quoted strings
+================================================================================
+
+echo 'a b' 'c d'
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (raw_string)
+    (raw_string)))
+
+================================================================================
+Double quoted strings
+================================================================================
+
+echo "a" "b"
+echo "a ${b} c" "d $e"
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (string (string_content))
+    (string (string_content)))
+  (command
+    (command_name
+      (word))
+    (string
+      (string_content)
+      (expansion
+        (variable_name))
+      (string_content))
+    (string
+      (string_content)
+      (simple_expansion
+        (variable_name)))))
+
+================================================================================
+Strings containing command substitutions
+================================================================================
+
+find "`dirname $file`" -name "$base"'*'
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (string
+      (command_substitution
+        (command
+          (command_name
+            (word))
+          (simple_expansion
+            (variable_name)))))
+    (word)
+    (concatenation
+      (string
+        (simple_expansion
+          (variable_name)))
+      (raw_string))))
+
+================================================================================
+Strings containing escape sequence
+================================================================================
+
+echo "\"The great escape\`\${var}"
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (string (string_content))))
+
+================================================================================
+Strings containing special characters
+================================================================================
+
+echo "s/$/'/"
+echo "#"
+echo "s$"
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (string (string_content) (string_content)))
+  (command
+    (command_name
+      (word))
+    (string (string_content)))
+  (command
+    (command_name
+      (word))
+    (string (string_content))))
+
+================================================================================
+Strings with ANSI-C quoting
+================================================================================
+
+echo $'Here\'s Johnny!\r\n'
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (ansi_c_string)))
+
+================================================================================
+Arrays and array expansions
+================================================================================
+
+a=()
+b=(1 2 3)
+
+echo ${a[@]}
+echo ${#b[@]}
+
+a[$i]=50
+a+=(foo "bar" $(baz))
+
+printf "  %-9s" "${seq0:-(default)}"
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment
+    (variable_name)
+    (array))
+  (variable_assignment
+    (variable_name)
+    (array
+      (number)
+      (number)
+      (number)))
+  (command
+    (command_name
+      (word))
+    (expansion
+      (subscript
+        (variable_name)
+        (word))))
+  (command
+    (command_name
+      (word))
+    (expansion
+      (subscript
+        (variable_name)
+        (word))))
+  (variable_assignment
+    (subscript
+      (variable_name)
+      (simple_expansion
+        (variable_name)))
+    (number))
+  (variable_assignment
+    (variable_name)
+    (array
+      (word)
+      (string (string_content))
+      (command_substitution
+        (command
+          (command_name
+            (word))))))
+  (command
+    (command_name
+      (word))
+    (string (string_content))
+    (string
+      (expansion
+        (variable_name)
+        (array
+          (word))))))
+
+================================================================================
+Escaped characters in strings
+================================================================================
+
+echo -ne "\033k$1\033\\" > /dev/stderr
+
+--------------------------------------------------------------------------------
+
+(program
+  (redirected_statement
+    (command
+      (command_name
+        (word))
+      (word)
+      (string
+        (string_content)
+        (simple_expansion
+          (variable_name))
+        (string_content)))
+    (file_redirect
+      (word))))
+
+================================================================================
+Words containing bare '#'
+================================================================================
+
+curl -# localhost #comment without space
+nix build nixpkgs#hello -v # comment with space
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (word)
+    (word))
+  (comment)
+  (command
+    (command_name
+      (word))
+    (word)
+    (word)
+    (word))
+  (comment))
+
+================================================================================
+Words containing # that are not comments
+================================================================================
+
+echo 'word'#not-comment # a legit comment
+echo $(uname -a)#not-comment # a legit comment
+echo `uname -a`#not-comment # a legit comment
+echo $hey#not-comment # a legit comment
+var=#not-comment # a legit comment
+echo "'$var'" # -> '#not-comment'
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (raw_string)
+      (word)))
+  (comment)
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (command_substitution
+        (command
+          (command_name
+            (word))
+          (word)))
+      (word)))
+  (comment)
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (command_substitution
+        (command
+          (command_name
+            (word))
+          (word)))
+      (word)))
+  (comment)
+  (command
+    (command_name
+      (word))
+    (concatenation
+      (simple_expansion
+        (variable_name))
+      (word)))
+  (comment)
+  (variable_assignment
+    (variable_name)
+    (word))
+  (comment)
+  (command
+    (command_name
+      (word))
+    (string
+      (string_content)
+      (simple_expansion
+        (variable_name))
+      (string_content)))
+  (comment))
+
+================================================================================
+Variable assignments immediately followed by a terminator
+================================================================================
+
+loop=; variables=& here=;;
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment
+    (variable_name))
+  (variable_assignment
+    (variable_name))
+  (variable_assignment
+    (variable_name)))
+
+================================================================================
+Multiple variable assignments
+================================================================================
+
+component_type="${1}" item_name="${2?}"
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignments
+    (variable_assignment
+      (variable_name)
+      (string
+        (expansion
+          (variable_name))))
+    (variable_assignment
+      (variable_name)
+      (string
+        (expansion
+          (variable_name))))))
+
+================================================================================
+Arithmetic expansions
+================================================================================
+
+echo $((1 + 2 - 3 * 4 / 5))
+a=$((6 % 7 ** 8 << 9 >> 10 & 11 | 12 ^ 13))
+$(((${1:-${SECONDS}} % 12) + 144))
+((foo=0))
+echo $((bar=1))
+echo $((-1, 1))
+echo $((! -a || ~ +b || ++c || --d))
+echo $((foo-- || bar++))
+(("${MULTIBUILD_VARIANTS}" > 1))
+$(("$(stat --printf '%05a' "${save_file}")" & 07177))
+soft_errors_count=$[soft_errors_count + 1]
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name (word))
+    (arithmetic_expansion
+      (binary_expression
+        (binary_expression (number) (number))
+        (binary_expression
+          (binary_expression (number) (number))
+          (number)))))
+  (variable_assignment
+    (variable_name)
+    (arithmetic_expansion
+      (binary_expression
+        (binary_expression
+          (binary_expression
+            (binary_expression
+              (binary_expression
+                (number)
+                (binary_expression (number) (number)))
+              (number))
+            (number))
+          (number))
+        (binary_expression (number) (number)))))
+  (command
+    (command_name
+      (arithmetic_expansion
+        (binary_expression
+          (parenthesized_expression
+            (binary_expression
+              (expansion
+                (variable_name)
+                (expansion (variable_name)))
+              (number)))
+          (number)))))
+  (command
+    (command_name
+      (arithmetic_expansion
+        (binary_expression (variable_name) (number)))))
+  (command
+    (command_name (word))
+    (arithmetic_expansion
+      (binary_expression (variable_name) (number))))
+  (command
+    (command_name (word))
+    (arithmetic_expansion (unary_expression (number)) (number)))
+  (command
+    (command_name (word))
+    (arithmetic_expansion
+      (binary_expression
+        (binary_expression
+          (binary_expression
+            (unary_expression (unary_expression (variable_name)))
+            (unary_expression (unary_expression (variable_name))))
+          (unary_expression (variable_name)))
+        (unary_expression (variable_name)))))
+  (command
+    (command_name (word))
+    (arithmetic_expansion
+      (binary_expression
+        (postfix_expression
+          (variable_name))
+        (postfix_expression
+          (variable_name)))))
+  (command
+    (command_name
+      (arithmetic_expansion
+        (binary_expression
+          (string
+            (expansion
+              (variable_name)))
+          (number)))))
+  (command
+    (command_name
+      (arithmetic_expansion
+        (binary_expression
+          (string
+            (command_substitution
+              (command
+                (command_name
+                  (word))
+                (word)
+                (raw_string)
+                (string
+                  (expansion
+                    (variable_name))))))
+          (number)))))
+  (variable_assignment
+    (variable_name)
+    (arithmetic_expansion
+      (binary_expression
+        (variable_name)
+        (number)))))
+
+================================================================================
+Concatenation with double backticks
+================================================================================
+
+main() {
+    local foo="asd"`
+        `"fgh"
+}
+
+---
+
+(program
+  (function_definition
+    (word)
+    (compound_statement
+      (declaration_command
+        (variable_assignment
+          (variable_name)
+          (concatenation
+            (string (string_content))
+            (string (string_content))))))))
+
+================================================================================
+Brace expressions and lookalikes
+================================================================================
+
+echo {1..2}
+echo {0..5}
+echo {0..2 # not a brace expression
+echo }{0..2}
+echo {0..n} # not a brace expression
+echo {0..n..2} # not a brace expression
+echo {0..2}{1..2}
+
+---
+
+(program
+  (command
+    (command_name (word))
+    (brace_expression (number) (number)))
+  (command
+    (command_name (word))
+    (brace_expression (number) (number)))
+  (command
+    (command_name (word))
+    (concatenation (word) (word)))
+  (comment)
+  (command
+    (command_name (word))
+    (concatenation
+      (word)
+      (brace_expression (number) (number))))
+  (command
+    (command_name (word))
+    (concatenation (word) (word) (word)))
+  (comment)
+  (command
+    (command_name (word))
+    (concatenation (word) (word) (word)))
+  (comment)
+  (command
+    (command_name (word))
+    (concatenation
+      (brace_expression (number) (number))
+      (brace_expression (number) (number)))))
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/programs.txt b/packages/tree-sitter-bash/test/fixtures/corpus/programs.txt
new file mode 100644
index 0000000000..cae772ea91
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/programs.txt
@@ -0,0 +1,108 @@
+===============================
+Comments
+===============================
+
+#!/bin/bash
+# hi
+
+---
+
+(program
+  (comment)
+  (comment))
+
+===============================
+Escaped newlines
+===============================
+
+abc \
+  d \
+  e
+
+f=g   \
+  h=i \
+  j   \
+  --k
+
+---
+
+(program
+  (command
+    (command_name
+      (word))
+    (word)
+    (word))
+  (command
+    (variable_assignment
+      (variable_name)
+      (word))
+    (variable_assignment
+      (variable_name)
+      (word))
+    (command_name
+      (word))
+    (word)))
+
+=============================
+escaped newline immediately after a char
+=============================
+
+echo a \
+    b
+
+echo a\
+    b
+
+echo a\
+    b\
+    c
+
+
+-----------------------------
+
+(program
+  (command
+    (command_name
+      (word))
+    (word)
+    (word))
+  (command
+    (command_name
+      (word))
+    (word)
+    (word))
+  (command
+    (command_name
+      (word))
+    (word)
+    (word)
+    (word)))
+
+=============================
+Escaped whitespace
+=============================
+
+echo 1 \ 2 \ 3
+
+---
+
+(program
+  (command
+    (command_name
+      (word))
+    (number)
+    (number)
+    (number)))
+
+====================================
+Files without trailing terminators
+====================================
+
+echo hi
+---
+
+(program
+  (command
+    (command_name
+      (word))
+    (word)))
diff --git a/packages/tree-sitter-bash/test/fixtures/corpus/statements.txt b/packages/tree-sitter-bash/test/fixtures/corpus/statements.txt
new file mode 100644
index 0000000000..b530fd4e24
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/corpus/statements.txt
@@ -0,0 +1,1622 @@
+================================================================================
+Pipelines
+================================================================================
+
+whoami | cat
+cat foo | grep -v bar
+cat baz | head -n 1
+
+--------------------------------------------------------------------------------
+
+(program
+  (pipeline
+    (command
+      name: (command_name
+        (word)))
+    (command
+      name: (command_name
+        (word))))
+  (pipeline
+    (command
+      name: (command_name
+        (word))
+      argument: (word))
+    (command
+      name: (command_name
+        (word))
+      argument: (word)
+      argument: (word)))
+  (pipeline
+    (command
+      name: (command_name
+        (word))
+      argument: (word))
+    (command
+      name: (command_name
+        (word))
+      argument: (word)
+      argument: (number))))
+
+
+================================================================================
+Lists
+================================================================================
+
+a | b && c && d; d e f || e g
+
+--------------------------------------------------------------------------------
+
+(program
+  (list
+    (list
+      (pipeline
+        (command
+          (command_name
+            (word)))
+        (command
+          (command_name
+            (word))))
+      (command
+        (command_name
+          (word))))
+    (command
+      (command_name
+        (word))))
+  (list
+    (command
+      (command_name
+        (word))
+      (word)
+      (word))
+    (command
+      (command_name
+        (word))
+      (word))))
+
+================================================================================
+While statements
+================================================================================
+
+while something happens; do
+  echo a
+  echo b
+done
+
+while local name="$1" val="$2"; shift 2; do
+  printf "%s (%s)\n" "$val" "$name"
+done
+
+--------------------------------------------------------------------------------
+
+(program
+  (while_statement
+    condition: (command
+      name: (command_name
+        (word))
+      argument: (word))
+    body: (do_group
+      (command
+        name: (command_name
+          (word))
+        argument: (word))
+      (command
+        name: (command_name
+          (word))
+        argument: (word))))
+  (while_statement
+    condition: (declaration_command
+      (variable_assignment
+        name: (variable_name)
+        value: (string
+          (simple_expansion
+            (variable_name))))
+      (variable_assignment
+        name: (variable_name)
+        value: (string
+          (simple_expansion
+            (variable_name)))))
+    condition: (command
+      name: (command_name
+        (word))
+      argument: (number))
+    body: (do_group
+      (command
+        name: (command_name
+          (word))
+        argument: (string (string_content))
+        argument: (string
+          (simple_expansion
+            (variable_name)))
+        argument: (string
+          (simple_expansion
+            (variable_name)))))))
+
+================================================================================
+Until statements
+================================================================================
+
+until something happens; do
+  echo a
+  echo b
+done
+
+--------------------------------------------------------------------------------
+
+(program
+  (while_statement
+    condition: (command
+      name: (command_name
+        (word))
+      argument: (word))
+    body: (do_group
+      (command
+        name: (command_name
+          (word))
+        argument: (word))
+      (command
+        name: (command_name
+          (word))
+        argument: (word)))))
+
+================================================================================
+While statements with IO redirects
+================================================================================
+
+while read line; do
+  echo $line
+done < <(cat file)
+
+--------------------------------------------------------------------------------
+
+(program
+  (redirected_statement
+    body: (while_statement
+      condition: (command
+        name: (command_name
+          (word))
+        argument: (word))
+      body: (do_group
+        (command
+          name: (command_name
+            (word))
+          argument: (simple_expansion
+            (variable_name)))))
+    redirect: (file_redirect
+      destination: (process_substitution
+        (command
+          name: (command_name
+            (word))
+          argument: (word))))))
+
+================================================================================
+For statements
+================================================================================
+
+for a in 1 2 $(seq 5 10); do
+  echo $a
+done
+
+for ARG; do
+  echo $ARG
+  ARG=''
+done
+
+for c in ${=1}; do
+	echo c
+done
+
+--------------------------------------------------------------------------------
+
+(program
+  (for_statement
+    variable: (variable_name)
+    value: (number)
+    value: (number)
+    value: (command_substitution
+      (command
+        name: (command_name
+          (word))
+        argument: (number)
+        argument: (number)))
+    body: (do_group
+      (command
+        name: (command_name
+          (word))
+        argument: (simple_expansion
+          (variable_name)))))
+  (for_statement
+    variable: (variable_name)
+    body: (do_group
+      (command
+        name: (command_name
+          (word))
+        argument: (simple_expansion
+          (variable_name)))
+      (variable_assignment
+        name: (variable_name)
+        value: (raw_string))))
+  (for_statement
+    variable: (variable_name)
+    value: (expansion
+      (variable_name))
+    body: (do_group
+      (command
+        name: (command_name
+          (word))
+        argument: (word)))))
+
+================================================================================
+Select statements
+================================================================================
+
+select choice in X Y $(ls); do
+  echo $choice
+  break
+done
+
+select ARG; do
+  echo $ARG
+  ARG=''
+done
+
+--------------------------------------------------------------------------------
+
+(program
+  (for_statement
+    (variable_name)
+    (word)
+    (word)
+    (command_substitution
+      (command
+        (command_name
+          (word))))
+    (do_group
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))
+      (command
+        (command_name
+          (word)))))
+  (for_statement
+    (variable_name)
+    (do_group
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))
+      (variable_assignment
+        (variable_name)
+        (raw_string)))))
+
+================================================================================
+C-style for statements
+================================================================================
+
+for (( c=1; c<=5; c++ ))
+do
+  echo $c
+done
+
+for (( c=1; c<=5; c++ )) {
+	echo $c
+}
+
+for (( ; ; ))
+do
+  echo 'forever'
+done
+
+for ((cx = 0; c = $cx / $pf, c < $wc - $k; )); do
+		echo "$cx"
+done
+
+for (( i = 4;;i--)) ; do echo $i; if (( $i == 0 )); then break; fi; done
+
+# added post-bash-4.2
+for (( i = j = k = 1; i % 9 || (j *= -1, $( ((i%9)) || printf " " >&2; echo 0), k++ <= 10); i += j ))
+do
+printf "$i"
+done
+
+echo
+
+( for (( i = j = k = 1; i % 9 || (j *= -1, $( ((i%9)) || printf " " >&2; echo 0), k++ <= 10); i += j ))
+do
+printf "$i"
+done )
+
+--------------------------------------------------------------------------------
+
+(program
+  (c_style_for_statement
+    (variable_assignment
+      (variable_name)
+      (number))
+    (binary_expression
+      (word)
+      (number))
+    (postfix_expression
+      (word))
+    (do_group
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))))
+  (c_style_for_statement
+    (variable_assignment
+      (variable_name)
+      (number))
+    (binary_expression
+      (word)
+      (number))
+    (postfix_expression
+      (word))
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))))
+  (c_style_for_statement
+    (do_group
+      (command
+        (command_name
+          (word))
+        (raw_string))))
+  (c_style_for_statement
+    (variable_assignment
+      (variable_name)
+      (number))
+    (variable_assignment
+      (variable_name)
+      (binary_expression
+        (simple_expansion
+          (variable_name))
+        (simple_expansion
+          (variable_name))))
+    (binary_expression
+      (word)
+      (binary_expression
+        (simple_expansion
+          (variable_name))
+        (simple_expansion
+          (variable_name))))
+    (do_group
+      (command
+        (command_name
+          (word))
+        (string
+          (simple_expansion
+            (variable_name))))))
+  (c_style_for_statement
+    (variable_assignment
+      (variable_name)
+      (number))
+    (postfix_expression
+      (word))
+    (do_group
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))
+      (if_statement
+        (command
+          (command_name
+            (arithmetic_expansion
+              (binary_expression
+                (simple_expansion
+                  (variable_name))
+                (number)))))
+        (command
+          (command_name
+            (word))))))
+  (comment)
+  (c_style_for_statement
+    (variable_assignment
+      (variable_name)
+      (variable_assignment
+        (variable_name)
+        (variable_assignment
+          (variable_name)
+          (number))))
+    (binary_expression
+      (binary_expression
+        (word)
+        (number))
+      (parenthesized_expression
+        (binary_expression
+          (word)
+          (number))
+        (command_substitution
+          (redirected_statement
+            (list
+              (command
+                (command_name
+                  (arithmetic_expansion
+                    (binary_expression
+                      (variable_name)
+                      (number)))))
+              (command
+                (command_name
+                  (word))
+                (string)))
+            (file_redirect
+              (number)))
+          (command
+            (command_name
+              (word))
+            (number)))
+        (binary_expression
+          (postfix_expression
+            (word))
+          (number))))
+    (binary_expression
+      (word)
+      (word))
+    (do_group
+      (command
+        (command_name
+          (word))
+        (string
+          (simple_expansion
+            (variable_name))))))
+  (command
+    (command_name
+      (word)))
+  (subshell
+    (c_style_for_statement
+      (variable_assignment
+        (variable_name)
+        (variable_assignment
+          (variable_name)
+          (variable_assignment
+            (variable_name)
+            (number))))
+      (binary_expression
+        (binary_expression
+          (word)
+          (number))
+        (parenthesized_expression
+          (binary_expression
+            (word)
+            (number))
+          (command_substitution
+            (redirected_statement
+              (list
+                (command
+                  (command_name
+                    (arithmetic_expansion
+                      (binary_expression
+                        (variable_name)
+                        (number)))))
+                (command
+                  (command_name
+                    (word))
+                  (string)))
+              (file_redirect
+                (number)))
+            (command
+              (command_name
+                (word))
+              (number)))
+          (binary_expression
+            (postfix_expression
+              (word))
+            (number))))
+      (binary_expression
+        (word)
+        (word))
+      (do_group
+        (command
+          (command_name
+            (word))
+          (string
+            (simple_expansion
+              (variable_name))))))))
+
+================================================================================
+If statements
+================================================================================
+
+if cat some_file | grep -v ok; then
+  echo one
+elif cat other_file | grep -v ok; then
+  echo two
+else
+  exit
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    (pipeline
+      (command
+        (command_name
+          (word))
+        (word))
+      (command
+        (command_name
+          (word))
+        (word)
+        (word)))
+    (command
+      (command_name
+        (word))
+      (word))
+    (elif_clause
+      (pipeline
+        (command
+          (command_name
+            (word))
+          (word))
+        (command
+          (command_name
+            (word))
+          (word)
+          (word)))
+      (command
+        (command_name
+          (word))
+        (word)))
+    (else_clause
+      (command
+        (command_name
+          (word))))))
+
+================================================================================
+If statements with conditional expressions
+================================================================================
+
+if [ "$(uname)" == 'Darwin' ]; then
+  echo one
+fi
+
+if [ a = -d ]; then
+  echo two
+fi
+
+[[ abc == +(a|b|c) ]] && echo 1
+[[ abc != +(a|b|c) ]] && echo 2
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    (test_command
+      (binary_expression
+        (string
+          (command_substitution
+            (command
+              (command_name
+                (word)))))
+        (raw_string)))
+    (command
+      (command_name
+        (word))
+      (word)))
+  (if_statement
+    (test_command
+      (binary_expression
+        (word)
+        (word)))
+    (command
+      (command_name
+        (word))
+      (word)))
+  (list
+    (test_command
+      (binary_expression
+        (word)
+        (extglob_pattern)))
+    (command
+      (command_name
+        (word))
+      (number)))
+  (list
+    (test_command
+      (binary_expression
+        (word)
+        (extglob_pattern)))
+    (command
+      (command_name
+        (word))
+      (number))))
+
+================================================================================
+If statements with negated command
+================================================================================
+
+if ! command -v echo; then
+  echo 'hello'
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    condition: (negated_command
+      (command
+        name: (command_name
+          (word))
+        argument: (word)
+        argument: (word)))
+    (command
+      name: (command_name
+        (word))
+      argument: (raw_string))))
+
+================================================================================
+If statements with command
+================================================================================
+
+if command -v echo; then
+  echo 'hello'
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    condition: (command
+      name: (command_name
+        (word))
+      argument: (word)
+      argument: (word))
+    (command
+      name: (command_name
+        (word))
+      argument: (raw_string))))
+
+================================================================================
+If statements with variable assignment by command substitution
+================================================================================
+
+if result=$(echo 'hello'); then
+  echo 'hello'
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    condition: (variable_assignment
+      name: (variable_name)
+      value: (command_substitution
+        (command
+          name: (command_name
+            (word))
+          argument: (raw_string))))
+    (command
+      name: (command_name
+        (word))
+      argument: (raw_string))))
+
+================================================================================
+If statements with negated variable assignment by command substitution
+================================================================================
+
+if ! result=$(echo 'hello'); then
+  echo 'hello'
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    condition: (negated_command
+      (variable_assignment
+        name: (variable_name)
+        value: (command_substitution
+          (command
+            name: (command_name
+              (word))
+            argument: (raw_string)))))
+    (command
+      name: (command_name
+        (word))
+      argument: (raw_string))))
+
+================================================================================
+If statements with variable assignment
+================================================================================
+
+if foo=1; then
+  echo 'hello'
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    condition: (variable_assignment
+      name: (variable_name)
+      value: (number))
+    (command
+      name: (command_name
+        (word))
+      argument: (raw_string))))
+
+================================================================================
+If statements with negated variable assignment
+================================================================================
+
+if ! foo=1; then
+  echo 'hello'
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    condition: (negated_command
+      (variable_assignment
+        name: (variable_name)
+        value: (number)))
+    (command
+      name: (command_name
+        (word))
+      argument: (raw_string))))
+
+================================================================================
+Case statements
+================================================================================
+
+case "opt" in
+  a)
+    echo a
+    ;;
+
+  b)
+    echo b
+    ;&
+
+  c)
+    echo c;;
+esac
+
+case "opt" in
+  (a)
+    echo a
+    ;;
+
+  (b)
+    echo b
+    ;&
+
+  (c)
+    echo c;;
+esac
+
+case "$Z" in
+  ab*|cd*) ef
+esac
+
+case $dest in
+  *.[1357])
+    exit $?
+    ;;
+esac
+
+case x in x) echo meow ;; esac
+
+case foo in
+  bar\ baz) : ;;
+esac
+
+case "$arg" in
+  *([0-9])([0-9])) echo "$arg"
+esac
+
+case ${lang} in
+CMakeLists.txt | \
+	cmake_modules | \
+	${PN}.pot) ;;
+*) rm -r ${lang} || die ;;
+esac
+
+--------------------------------------------------------------------------------
+
+(program
+  (case_statement
+    (string (string_content))
+    (case_item (word) (command (command_name (word)) (word)))
+    (case_item (word) (command (command_name (word)) (word)))
+    (case_item (word) (command (command_name (word)) (word))))
+  (case_statement
+    (string (string_content))
+    (case_item (word) (command (command_name (word)) (word)))
+    (case_item (word) (command (command_name (word)) (word)))
+    (case_item (word) (command (command_name (word)) (word))))
+  (case_statement
+    (string (simple_expansion (variable_name)))
+    (case_item (extglob_pattern) (extglob_pattern) (command (command_name (word)))))
+  (case_statement
+    (simple_expansion (variable_name))
+    (case_item
+      (concatenation (word) (word) (number) (word))
+      (command (command_name (word)) (simple_expansion (special_variable_name)))))
+  (case_statement (word) (case_item (word) (command (command_name (word)) (word))))
+  (case_statement (word) (case_item (word) (command (command_name (word)))))
+  (case_statement
+    (string (simple_expansion (variable_name)))
+    (case_item
+      (extglob_pattern)
+      (command (command_name (word)) (string (simple_expansion (variable_name))))))
+  (case_statement
+    (expansion (variable_name))
+    (case_item (word) (word) (concatenation (expansion (variable_name)) (word)))
+    (case_item
+      (extglob_pattern)
+      (list
+        (command (command_name (word)) (word) (expansion (variable_name)))
+        (command (command_name (word)))))))
+
+================================================================================
+Test commands
+================================================================================
+
+if [[ "$lsb_dist" != 'Ubuntu' || $(ver_to_int "$lsb_release") < $(ver_to_int '14.04') ]]; then
+	return 1
+fi
+
+[[ ${PV} != $(sed -n -e 's/^Version: //p' "${ED}/usr/$(get_libdir)/pkgconfig/tss2-tcti-tabrmd.pc" || die) ]]
+
+[[ ${f} != */@(default).vim ]]
+
+[[ "${MY_LOCALES}" != *en_US* || a != 2 ]]
+
+[[ $(LC_ALL=C $(tc-getCC) ${LDFLAGS} -Wl,--version 2>/dev/null) != @(LLD|GNU\ ld)* ]]
+
+[[ -f "${EROOT}/usr/share/php/.packagexml/${MY_P}.xml" && \
+	-x "${EROOT}/usr/bin/peardev" ]]
+
+[[ ${test} == @($(IFS='|'; echo "${skip[*]}")) ]]
+
+[[ ${SRC_URI} == */${a}* ]]
+
+[[ a == *_@(LIB|SYMLINK) ]]
+
+[[ ${1} =~ \.(lisp|lsp|cl)$ ]]
+
+[[ a == - ]]
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    (test_command
+      (binary_expression
+        (binary_expression (string (simple_expansion (variable_name))) (raw_string))
+        (binary_expression
+          (command_substitution
+            (command (command_name (word)) (string (simple_expansion (variable_name)))))
+          (command_substitution (command (command_name (word)) (raw_string))))))
+    (command (command_name (word)) (number)))
+  (test_command
+    (binary_expression
+      (expansion (variable_name))
+      (command_substitution
+        (list
+          (command
+            (command_name (word))
+            (word)
+            (word)
+            (raw_string)
+            (string
+              (expansion (variable_name))
+              (string_content)
+              (command_substitution (command (command_name (word))))
+              (string_content)))
+          (command (command_name (word)))))))
+  (test_command (binary_expression (expansion (variable_name)) (extglob_pattern)))
+  (test_command
+    (binary_expression
+      (binary_expression (string (expansion (variable_name))) (extglob_pattern))
+      (binary_expression (word) (number))))
+  (test_command
+    (binary_expression
+      (command_substitution
+        (redirected_statement
+          (command
+            (variable_assignment (variable_name) (word))
+            (command_name (command_substitution (command (command_name (word)))))
+            (expansion (variable_name))
+            (word))
+          (file_redirect (file_descriptor) (word))))
+      (extglob_pattern)))
+  (test_command
+    (binary_expression
+      (unary_expression
+        (test_operator)
+        (string
+          (expansion (variable_name))
+          (string_content)
+          (expansion (variable_name))
+          (string_content)))
+      (unary_expression
+        (test_operator)
+        (string (expansion (variable_name)) (string_content)))))
+  (test_command
+    (binary_expression
+      (expansion (variable_name))
+      (extglob_pattern)
+      (command_substitution
+        (variable_assignment (variable_name) (raw_string))
+        (command
+          (command_name (word))
+          (string (expansion (subscript (variable_name) (word))))))
+      (extglob_pattern)))
+  (test_command
+    (binary_expression
+      (expansion (variable_name))
+      (extglob_pattern)
+      (expansion (variable_name))
+      (extglob_pattern)))
+  (test_command (binary_expression (word) (extglob_pattern)))
+  (test_command (binary_expression (expansion (variable_name)) (regex)))
+  (test_command (binary_expression (word) (extglob_pattern))))
+
+================================================================================
+Test commands with ternary
+================================================================================
+
+if (( 1 < 2 ? 1 : 2 )); then
+	return 1
+fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    (command
+      (command_name
+        (arithmetic_expansion
+          (ternary_expression
+            (binary_expression
+              (number)
+              (number))
+            (number)
+            (number)))))
+    (command
+      (command_name
+        (word))
+      (number))))
+
+================================================================================
+Ternary expressions
+================================================================================
+
+$((n < 10 ? n : 10))
+$(($n < 10 ? $n : 10))
+$((${n} < 10 ? ${n} : 10))
+
+--------------------------------------------------------------------------------
+
+(program
+  (command
+    (command_name
+      (arithmetic_expansion
+        (ternary_expression
+          (binary_expression
+            (variable_name)
+            (number))
+          (variable_name)
+          (number)))))
+  (command
+    (command_name
+      (arithmetic_expansion
+        (ternary_expression
+          (binary_expression
+            (simple_expansion
+              (variable_name))
+            (number))
+          (simple_expansion
+            (variable_name))
+          (number)))))
+  (command
+    (command_name
+      (arithmetic_expansion
+        (ternary_expression
+          (binary_expression
+            (expansion
+              (variable_name))
+            (number))
+          (expansion
+            (variable_name))
+          (number))))))
+
+================================================================================
+Test commands with regexes
+================================================================================
+
+[[ "35d8b" =~ ^[0-9a-fA-F] ]]
+[[ $CMD =~ (^|;)update_terminal_cwd($|;) ]]
+[[ ! " ${completions[*]} " =~ " $alias_cmd " ]]
+! [[ "$a" =~ ^a|b\ *c|d$ ]]
+[[ "$1" =~ ^${var}${var}*=..* ]]
+[[ "$1" =~ ^\-${var}+ ]]
+[[ ${var1} == *${var2}* ]]
+[[ "$server" =~ [0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3} ]]
+[[ "$primary_wins" =~ ([0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}) ]]
+[[ -f ${x} && $(od -t x1 -N 4 "${x}") == *"7f 45 4c 46"* ]]
+[[ ${PV} =~ 99999999$ ]]
+[[ $1 == -- ]]
+[[ " ${REPLACING_VERSIONS} " != *\ ${PVR}\ * ]]
+[[ ${file} == @(*${GENTOO_PATCH_NAME}.tar.xz|*.asc|*.sig) ]]
+[[ $RUBY_TARGETS != *$( eselect ruby show | awk 'NR==2' | tr  -d ' '  )* ]]
+[[ " ${m[0]##*/}" =~ ^(\ ${skip_files[*]/%/.*|\\} ) ]]
+[[ ' boop xyz' =~ ' boop '(.*)$ ]]
+[[ $b = foo* ]]
+
+--------------------------------------------------------------------------------
+
+(program
+  (test_command (binary_expression (string (string_content)) (regex)))
+  (test_command (binary_expression (simple_expansion (variable_name)) (regex)))
+  (test_command
+    (binary_expression
+      (unary_expression
+        (string (expansion (subscript (variable_name) (word)))))
+      (string (simple_expansion (variable_name)))))
+  (negated_command (test_command (binary_expression (string (simple_expansion (variable_name))) (regex))))
+  (test_command (binary_expression (string (simple_expansion (variable_name))) (regex)))
+  (test_command (binary_expression (string (simple_expansion (variable_name))) (regex)))
+  (test_command
+    (binary_expression
+      (expansion (variable_name))
+      (extglob_pattern)
+      (expansion (variable_name))
+      (extglob_pattern)))
+  (test_command (binary_expression (string (simple_expansion (variable_name))) (regex)))
+  (test_command (binary_expression (string (simple_expansion (variable_name))) (regex)))
+  (test_command
+    (binary_expression
+      (unary_expression (test_operator) (expansion (variable_name)))
+      (binary_expression
+        (command_substitution
+          (command
+            (command_name (word))
+            (word)
+            (word)
+            (word)
+            (number)
+            (string (expansion (variable_name)))))
+        (extglob_pattern)
+        (string (string_content))
+        (extglob_pattern))))
+  (test_command (binary_expression (expansion (variable_name)) (regex)))
+  (test_command (binary_expression (simple_expansion (variable_name)) (extglob_pattern)))
+  (test_command
+    (binary_expression
+      (string (expansion (variable_name)))
+      (extglob_pattern)
+      (expansion (variable_name))
+      (extglob_pattern)))
+  (test_command
+    (binary_expression
+      (expansion (variable_name))
+      (extglob_pattern)
+      (expansion (variable_name))
+      (extglob_pattern)))
+  (test_command
+    (binary_expression
+      (simple_expansion (variable_name))
+      (extglob_pattern)
+      (command_substitution
+        (pipeline
+          (command (command_name (word)) (word) (word))
+          (command (command_name (word)) (raw_string))
+          (command (command_name (word)) (word) (raw_string))))
+      (extglob_pattern)))
+  (test_command
+    (binary_expression (string (expansion (subscript (variable_name) (number)) (regex))) (regex)))
+  (test_command (binary_expression (raw_string) (regex)))
+  (test_command (binary_expression (simple_expansion (variable_name)) (regex))))
+
+================================================================================
+Test command paren statefulness with a case glob
+================================================================================
+
+[[ ${test} == @($(IFS='|'; echo "${skip[*]}")) ]]
+
+case ${out} in
+*"not supported"*|\
+*"operation not supported"*)
+	;;
+esac
+
+case $1 in
+-o)
+	owner=$2
+	shift
+	;;
+-g) ;;
+esac
+
+[[ a == \"+(?)\" ]]
+
+---
+
+(program
+  (test_command
+    (binary_expression
+      (expansion (variable_name))
+      (extglob_pattern)
+      (command_substitution
+        (variable_assignment (variable_name) (raw_string))
+        (command
+        (command_name (word))
+          (string (expansion (subscript (variable_name) (word))))))
+      (extglob_pattern)))
+  (case_statement
+    (expansion (variable_name))
+    (case_item
+      (extglob_pattern)
+      (string (string_content))
+      (extglob_pattern)
+      (concatenation (word) (string (string_content)) (word))))
+  (case_statement
+    (simple_expansion (variable_name))
+    (case_item
+      (extglob_pattern)
+      (variable_assignment (variable_name) (simple_expansion (variable_name)))
+      (command (command_name (word))))
+    (case_item (word)))
+  (test_command (binary_expression (word) (extglob_pattern))))
+
+================================================================================
+Subshells
+================================================================================
+
+(
+  ./start-server --port=80
+) &
+
+time ( cd tests && sh run-tests.sh )
+
+--------------------------------------------------------------------------------
+
+(program
+  (subshell (command (command_name (word)) (word)))
+  (command
+    (command_name (word))
+    (subshell
+      (list
+        (command (command_name (word)) (word))
+        (command (command_name (word)) (word))))))
+
+================================================================================
+Function definitions
+================================================================================
+
+do_something() {
+  echo ok
+}
+
+run_subshell_command() (
+  true
+)
+
+run_test_command() [[ -e foo ]]
+
+function do_something_else() {
+  a | xargs -I{} find xml/{} -type f
+}
+
+function do_yet_another_thing {
+  echo ok
+} 2>&1
+
+do_nothing() { return 0; }
+
+foo::bar() {
+  echo what
+}
+
+foo::baz() {
+  echo how
+}
+
+assert()
+  if ! $1; then
+    return 1
+  fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (function_definition
+    (word)
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (word))))
+  (function_definition
+    (word)
+    (subshell
+      (command
+        (command_name
+          (word)))))
+  (function_definition
+    (word)
+    (test_command
+      (unary_expression
+        (test_operator)
+        (word))))
+  (function_definition
+    (word)
+    (compound_statement
+      (pipeline
+        (command
+          (command_name
+            (word)))
+        (command
+          (command_name
+            (word))
+          (concatenation
+            (word)
+            (word)
+            (word))
+          (word)
+          (concatenation
+            (word)
+            (word)
+            (word))
+          (word)
+          (word)))))
+  (function_definition
+    (word)
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (word)))
+    (file_redirect
+      (file_descriptor)
+      (number)))
+  (function_definition
+    (word)
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (number))))
+  (function_definition
+    (word)
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (word))))
+  (function_definition
+    (word)
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (word))))
+  (function_definition
+    (word)
+    (if_statement
+      (negated_command
+        (command
+          (command_name
+            (simple_expansion
+              (variable_name)))))
+      (command
+        (command_name
+          (word))
+        (number)))))
+
+================================================================================
+Variable declaration: declare & typeset
+================================================================================
+
+declare var1
+typeset -i -r var2=42 var3=10
+
+--------------------------------------------------------------------------------
+
+(program
+  (declaration_command
+    (variable_name))
+  (declaration_command
+    (word)
+    (word)
+    (variable_assignment
+      (variable_name)
+      (number))
+    (variable_assignment
+      (variable_name)
+      (number))))
+
+================================================================================
+Variable declaration: readonly
+================================================================================
+
+readonly var1
+readonly var2=42
+
+--------------------------------------------------------------------------------
+
+(program
+  (declaration_command
+    (variable_name))
+  (declaration_command
+    (variable_assignment
+      (variable_name)
+      (number))))
+
+================================================================================
+Variable declaration: local
+================================================================================
+
+local a=42 b
+local -r c
+local var=word1\ word2
+
+--------------------------------------------------------------------------------
+
+(program
+  (declaration_command
+    (variable_assignment
+      (variable_name)
+      (number))
+    (variable_name))
+  (declaration_command
+    (word)
+    (variable_name))
+  (declaration_command
+    (variable_assignment
+      (variable_name)
+      (word))))
+
+================================================================================
+Variable declaration: export
+================================================================================
+
+export PATH
+export FOOBAR PATH="$PATH:/usr/foobar/bin"
+export $FOO:$BAR
+
+--------------------------------------------------------------------------------
+
+(program
+  (declaration_command
+    (variable_name))
+  (declaration_command
+    (variable_name)
+    (variable_assignment
+      (variable_name)
+      (string
+        (simple_expansion
+          (variable_name))
+          (string_content))))
+  (declaration_command
+    (concatenation
+      (simple_expansion
+        (variable_name))
+      (word)
+      (simple_expansion
+        (variable_name)))))
+
+================================================================================
+Variable declaration: command substitution with semi-colon
+================================================================================
+
+_path=$(
+  while statement; do
+    cd ".."
+  done;
+  echo $PWD
+)
+
+--------------------------------------------------------------------------------
+
+(program
+  (variable_assignment
+    (variable_name)
+    (command_substitution
+      (while_statement
+        (command
+          (command_name
+            (word)))
+        (do_group
+          (command
+            (command_name
+              (word))
+            (string (string_content)))))
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name))))))
+
+===========================================================
+Command substution with $ and backticks
+===========================================================
+
+$(eval echo $`echo ${foo}`)
+
+---
+
+(program
+  (command
+    (command_name
+      (command_substitution
+        (command
+          (command_name
+            (word))
+          (word)
+          (command_substitution
+            (command
+              (command_name
+                (word))
+              (expansion
+                (variable_name)))))))))
+
+================================================================================
+Expressions passed to declaration commands
+================================================================================
+
+export "$(echo ${key} | tr [:lower:] [:upper:])=${p_key#*=}"
+
+--------------------------------------------------------------------------------
+
+(program
+  (declaration_command
+    (string
+      (command_substitution
+        (pipeline
+          (command
+            (command_name
+              (word))
+            (expansion
+              (variable_name)))
+          (command
+            (command_name
+              (word))
+            (concatenation
+              (word)
+              (word)
+              (word))
+            (concatenation
+              (word)
+              (word)
+              (word)))))
+      (string_content)
+      (expansion
+        (variable_name)
+        (regex)))))
+
+================================================================================
+Unset commands
+================================================================================
+
+unset A
+unset "$variable_name"
+unsetenv -f ONE TWO
+
+--------------------------------------------------------------------------------
+
+(program
+  (unset_command
+    (variable_name))
+  (unset_command
+    (string
+      (simple_expansion
+        (variable_name))))
+  (unset_command
+    (word)
+    (variable_name)
+    (variable_name)))
+
+================================================================================
+Compound statements
+================================================================================
+
+a () {
+    ls || { echo "b"; return 0; }
+    echo c
+}
+
+{ echo "a"
+  echo "b"
+} >&2
+
+--------------------------------------------------------------------------------
+
+(program
+  (function_definition
+    (word)
+    (compound_statement
+      (list
+        (command
+          (command_name
+            (word)))
+        (compound_statement
+          (command
+            (command_name
+              (word))
+            (string (string_content)))
+          (command
+            (command_name
+              (word))
+            (number))))
+      (command
+        (command_name
+          (word))
+        (word))))
+  (redirected_statement
+    (compound_statement
+      (command
+        (command_name
+          (word))
+        (string (string_content)))
+      (command
+        (command_name
+          (word))
+        (string (string_content))))
+    (file_redirect
+      (number))))
+
+================================================================================
+If condition with subshell
+================================================================================
+
+if (echo $BASHPID; true); then echo $BASHPID; fi
+
+--------------------------------------------------------------------------------
+
+(program
+  (if_statement
+    (subshell
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))
+      (command
+        (command_name
+          (word))))
+    (command
+      (command_name
+        (word))
+      (simple_expansion
+        (variable_name)))))
+
+================================================================================
+While condition with subshell
+================================================================================
+
+while (echo $BASHPID; true); do echo $BASHPID; break; done
+
+--------------------------------------------------------------------------------
+
+(program
+  (while_statement
+    (subshell
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))
+      (command
+        (command_name
+          (word))))
+    (do_group
+      (command
+        (command_name
+          (word))
+        (simple_expansion
+          (variable_name)))
+      (command
+        (command_name
+          (word))))))
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/arithmetic.txt b/packages/tree-sitter-bash/test/fixtures/differential/arithmetic.txt
new file mode 100644
index 0000000000..f2583d371a
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/arithmetic.txt
@@ -0,0 +1,236 @@
+@match: for ((i=0;i<3;i++)); do echo $i; done
+for ((i=0;i<3;i++)); do echo $i; done
+===
+@match: for (( i = 0; i < 10; i += 2 )); do x; done
+for (( i = 0; i < 10; i += 2 )); do x; done
+===
+@match: echo $((1 + 2 * 3))
+echo $((1 + 2 * 3))
+===
+@match: echo $((a > b ? a : b))
+echo $((a > b ? a : b))
+===
+@match: echo $(( (1+2) ** 3 ))
+echo $(( (1+2) ** 3 ))
+===
+@match: echo $((i++ + --j))
+echo $((i++ + --j))
+===
+@match: echo $((-x + ~y))
+echo $((-x + ~y))
+===
+@match: echo $((arr[0] + 1))
+echo $((arr[0] + 1))
+===
+@match: echo $((x += 5))
+echo $((x += 5))
+===
+@match: echo $((1, 2, 3))
+echo $((1, 2, 3))
+===
+@match: ((x = y + 1))
+((x = y + 1))
+===
+@match: echo $((n % 2 == 0))
+echo $((n % 2 == 0))
+===
+@match: echo $((2 ** 3 ** 2))
+echo $((2 ** 3 ** 2))
+===
+@match: echo $((x = y = 1))
+echo $((x = y = 1))
+===
+@match: echo $((
+echo $((
+1 +
+2
+))
+===
+@match: (( ls ))
+(( ls ))
+===
+@match: ((x=1)); echo $x
+((x=1)); echo $x
+===
+@match: for ((;;)); do x; done
+for ((;;)); do x; done
+===
+@match: for ((i=0,j=1; i<3; i++,j--)); do x; done
+for ((i=0,j=1; i<3; i++,j--)); do x; done
+===
+@match: for ((i=0;i<3;i++)) { echo $i; }
+for ((i=0;i<3;i++)) { echo $i; }
+===
+@match: echo $[1+2]
+echo $[1+2]
+===
+@match: echo $((x[1]+y[z]))
+echo $((x[1]+y[z]))
+===
+@match: ((a++))
+((a++))
+===
+@match: ((a--))
+((a--))
+===
+@match: ((a++ + b))
+((a++ + b))
+===
+@match: ((b + a++))
+((b + a++))
+===
+@known-diff arithmetic-hex: echo $((0x1F))
+echo $((0x1F))
+---
+hasError: false
+program [0,14] "echo $((0x1F))"
+  command [0,14] "echo $((0x1F))"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    arithmetic_expansion [5,14] "$((0x1F))"
+      ($(() [5,8] "$(("
+      number [8,12] "0x1F"
+      ())) [12,14] "))"
+===
+@match: echo ((a++))
+echo ((a++))
+===
+@match: ((a[i]++))
+((a[i]++))
+===
+@match: ! ((a++))
+! ((a++))
+===
+@match: ((ab-cd++))
+((ab-cd++))
+===
+@match: echo $((1+2)) more
+echo $((1+2)) more
+===
+@match: $(((${1:-${SECONDS}} % 12) + 144))
+$(((${1:-${SECONDS}} % 12) + 144))
+===
+@match: echo $((-1, 1))
+echo $((-1, 1))
+===
+@match: echo $((! -a || ~ +b || ++c || --d))
+echo $((! -a || ~ +b || ++c || --d))
+===
+@match: (("${MULTIBUILD_VARIANTS}" > 1))
+(("${MULTIBUILD_VARIANTS}" > 1))
+===
+@match: $(("$(stat --printf '%05a' "${save_file}")" & 07177))
+$(("$(stat --printf '%05a' "${save_file}")" & 07177))
+===
+@match: soft_errors_count=$[soft_errors_count + 1]
+soft_errors_count=$[soft_errors_count + 1]
+===
+@known-diff cfor-compound-assign-negative: for (( i = j = k = 1; i % 9 || (j *= -1, $( ((i%9)) || pr...
+for (( i = j = k = 1; i % 9 || (j *= -1, $( ((i%9)) || printf " " >&2; echo 0), k++ <= 10); i += j ))
+do
+printf "$i"
+done
+---
+hasError: false
+program [0,121] "for (( i = j = k = 1; i % 9 || (j *= ..."
+  c_style_for_statement [0,121] "for (( i = j = k = 1; i % 9 || (j *= ..."
+    (for) [0,3] "for"
+    ((() [4,6] "(("
+    variable_assignment [7,20] "i = j = k = 1"
+      variable_name [7,8] "i"
+      (=) [9,10] "="
+      variable_assignment [11,20] "j = k = 1"
+        variable_name [11,12] "j"
+        (=) [13,14] "="
+        variable_assignment [15,20] "k = 1"
+          variable_name [15,16] "k"
+          (=) [17,18] "="
+          number [19,20] "1"
+    (;) [20,21] ";"
+    binary_expression [22,90] "i % 9 || (j *= -1, $( ((i%9)) || prin..."
+      binary_expression [22,27] "i % 9"
+        word [22,23] "i"
+        (%) [24,25] "%"
+        number [26,27] "9"
+      (||) [28,30] "||"
+      parenthesized_expression [31,90] "(j *= -1, $( ((i%9)) || printf \" \" >&..."
+        (() [31,32] "("
+        binary_expression [32,39] "j *= -1"
+          word [32,33] "j"
+          (*=) [34,36] "*="
+          unary_expression [37,39] "-1"
+            (-) [37,38] "-"
+            number [38,39] "1"
+        (,) [39,40] ","
+        command_substitution [41,78] "$( ((i%9)) || printf \" \" >&2; echo 0)"
+          ($() [41,43] "$("
+          list [44,69] "((i%9)) || printf \" \" >&2"
+            command [44,51] "((i%9))"
+              command_name [44,51] "((i%9))"
+                arithmetic_expansion [44,51] "((i%9))"
+                  ((() [44,46] "(("
+                  binary_expression [46,49] "i%9"
+                    variable_name [46,47] "i"
+                    (%) [47,48] "%"
+                    number [48,49] "9"
+                  ())) [49,51] "))"
+            (||) [52,54] "||"
+            redirected_statement [55,69] "printf \" \" >&2"
+              command [55,65] "printf \" \""
+                command_name [55,61] "printf"
+                  word [55,61] "printf"
+                string [62,65] "\" \""
+                  (") [62,63] "\""
+                  string_content [63,64] " "
+                  (") [64,65] "\""
+              file_redirect [66,69] ">&2"
+                (>&) [66,68] ">&"
+                number [68,69] "2"
+          (;) [69,70] ";"
+          command [71,77] "echo 0"
+            command_name [71,75] "echo"
+              word [71,75] "echo"
+            number [76,77] "0"
+          ()) [77,78] ")"
+        (,) [78,79] ","
+        binary_expression [80,89] "k++ <= 10"
+          postfix_expression [80,83] "k++"
+            word [80,81] "k"
+            (++) [81,83] "++"
+          (<=) [84,86] "<="
+          number [87,89] "10"
+        ()) [89,90] ")"
+    (;) [90,91] ";"
+    binary_expression [92,98] "i += j"
+      word [92,93] "i"
+      (+=) [94,96] "+="
+      word [97,98] "j"
+    ())) [99,101] "))"
+    do_group [102,121] "do\nprintf \"$i\"\ndone"
+      (do) [102,104] "do"
+      command [105,116] "printf \"$i\""
+        command_name [105,111] "printf"
+          word [105,111] "printf"
+        string [112,116] "\"$i\""
+          (") [112,113] "\""
+          simple_expansion [113,115] "$i"
+            ($) [113,114] "$"
+            variable_name [114,115] "i"
+          (") [115,116] "\""
+      (done) [117,121] "done"
+===
+@match: ((x = -1))
+((x = -1))
+===
+@match: ((x += -1))
+((x += -1))
+===
+@match: echo $((-1))
+echo $((-1))
+===
+@match: echo $((x * -1))
+echo $((x * -1))
+===
+@match: echo $((x=-1))
+echo $((x=-1))
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/case.txt b/packages/tree-sitter-bash/test/fixtures/differential/case.txt
new file mode 100644
index 0000000000..bbb781340e
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/case.txt
@@ -0,0 +1,436 @@
+@match: case $x in
+case $x in
+  a) echo A ;;
+  b|c) echo BC ;&
+  *) echo other ;;
+esac
+===
+@match: case $x in (a) x ;; esac
+case $x in (a) x ;; esac
+===
+@match: case $x in a|b) x ;; esac
+case $x in a|b) x ;; esac
+===
+@match: case $x in foo*|bar) x ;; esac
+case $x in foo*|bar) x ;; esac
+===
+@match: case $x in [a-z]) x ;; esac
+case $x in [a-z]) x ;; esac
+===
+@match: case $x in "a") x ;; esac
+case $x in "a") x ;; esac
+===
+@match: case $x in $v|${w}) x ;; esac
+case $x in $v|${w}) x ;; esac
+===
+@known-diff recovery-partial-nodes: case $x in a) x ;;& esac
+case $x in a) x ;;& esac
+---
+hasError: true
+program [0,24] "case $x in a) x ;;& esac"
+  case_statement [0,24] "case $x in a) x ;;& esac"
+    (case) [0,4] "case"
+    simple_expansion [5,7] "$x"
+      ($) [5,6] "$"
+      variable_name [6,7] "x"
+    (in) [8,10] "in"
+    case_item [11,19] "a) x ;;&"
+      word [11,12] "a"
+      ()) [12,13] ")"
+      command [14,15] "x"
+        command_name [14,15] "x"
+          word [14,15] "x"
+      (;;&) [16,19] ";;&"
+    (esac) [20,24] "esac"
+===
+@match: case $x in a) x esac
+case $x in a) x esac
+===
+@match: case $x in esac
+case $x in esac
+===
+@match: case x in a) ;; esac
+case x in a) ;; esac
+===
+@match: case $x in
+case $x in
+a) y ;;
+esac
+===
+@match: case $x in a) echo esac ;; esac
+case $x in a) echo esac ;; esac
+===
+@match: case $x in a) x
+case $x in a) x
+esac
+===
+@match: x=$(case y in a) 1;; esac)
+x=$(case y in a) 1;; esac)
+===
+@match: x=$(case y in (b) 2;; esac)
+x=$(case y in (b) 2;; esac)
+===
+@match: diff <(case y in a) 1;; esac) other
+diff <(case y in a) 1;; esac) other
+===
+@match: case $x in [1]) x ;; esac
+case $x in [1]) x ;; esac
+===
+@known-diff recovery-partial-nodes: if a; then case x in b) 1;; esac elif c; then d; fi
+if a; then case x in b) 1;; esac elif c; then d; fi
+---
+hasError: true
+program [0,51] "if a; then case x in b) 1;; esac elif..."
+  if_statement [0,51] "if a; then case x in b) 1;; esac elif..."
+    (if) [0,2] "if"
+    command [3,4] "a"
+      command_name [3,4] "a"
+        word [3,4] "a"
+    (;) [4,5] ";"
+    (then) [6,10] "then"
+    case_statement [11,32] "case x in b) 1;; esac"
+      (case) [11,15] "case"
+      word [16,17] "x"
+      (in) [18,20] "in"
+      case_item [21,27] "b) 1;;"
+        word [21,22] "b"
+        ()) [22,23] ")"
+        command [24,25] "1"
+          command_name [24,25] "1"
+            number [24,25] "1"
+        (;;) [25,27] ";;"
+      (esac) [28,32] "esac"
+    elif_clause [33,48] "elif c; then d;"
+      (elif) [33,37] "elif"
+      command [38,39] "c"
+        command_name [38,39] "c"
+          word [38,39] "c"
+      (;) [39,40] ";"
+      (then) [41,45] "then"
+      command [46,47] "d"
+        command_name [46,47] "d"
+          word [46,47] "d"
+      (;) [47,48] ";"
+    (fi) [49,51] "fi"
+===
+@match: case $dest in
+case $dest in
+  *.[1357])
+    exit $?
+    ;;
+esac
+===
+@match: case "$arg" in
+case "$arg" in
+  *([0-9])([0-9])) echo "$arg"
+esac
+===
+@known-diff case-pattern-after-continuation: case ${lang} in
+case ${lang} in
+CMakeLists.txt | \
+	cmake_modules | \
+	${PN}.pot) ;;
+*) rm -r ${lang} || die ;;
+esac
+---
+hasError: false
+program [0,100] "case ${lang} in\nCMakeLists.txt | \\\n\tc..."
+  case_statement [0,100] "case ${lang} in\nCMakeLists.txt | \\\n\tc..."
+    (case) [0,4] "case"
+    expansion [5,12] "${lang}"
+      (${) [5,7] "${"
+      variable_name [7,11] "lang"
+      (}) [11,12] "}"
+    (in) [13,15] "in"
+    case_item [16,68] "CMakeLists.txt | \\\n\tcmake_modules | \\..."
+      word [16,30] "CMakeLists.txt"
+      (|) [31,32] "|"
+      extglob_pattern [36,49] "cmake_modules"
+      (|) [50,51] "|"
+      concatenation [55,64] "${PN}.pot"
+        expansion [55,60] "${PN}"
+          (${) [55,57] "${"
+          variable_name [57,59] "PN"
+          (}) [59,60] "}"
+        word [60,64] ".pot"
+      ()) [64,65] ")"
+      (;;) [66,68] ";;"
+    case_item [69,95] "*) rm -r ${lang} || die ;;"
+      extglob_pattern [69,70] "*"
+      ()) [70,71] ")"
+      list [72,92] "rm -r ${lang} || die"
+        command [72,85] "rm -r ${lang}"
+          command_name [72,74] "rm"
+            word [72,74] "rm"
+          word [75,77] "-r"
+          expansion [78,85] "${lang}"
+            (${) [78,80] "${"
+            variable_name [80,84] "lang"
+            (}) [84,85] "}"
+        (||) [86,88] "||"
+        command [89,92] "die"
+          command_name [89,92] "die"
+            word [89,92] "die"
+      (;;) [93,95] ";;"
+    (esac) [96,100] "esac"
+===
+@match: case ${out} in
+case ${out} in
+*"not supported"*|\
+*"operation not supported"*)
+	;;
+esac
+===
+@known-diff case-short-option-pattern: case $1 in
+case $1 in
+-o)
+	owner=$2
+	shift
+	;;
+-g) ;;
+esac
+---
+hasError: false
+program [0,47] "case $1 in\n-o)\n\towner=$2\n\tshift\n\t;;\n-..."
+  case_statement [0,47] "case $1 in\n-o)\n\towner=$2\n\tshift\n\t;;\n-..."
+    (case) [0,4] "case"
+    simple_expansion [5,7] "$1"
+      ($) [5,6] "$"
+      variable_name [6,7] "1"
+    (in) [8,10] "in"
+    case_item [11,35] "-o)\n\towner=$2\n\tshift\n\t;;"
+      word [11,13] "-o"
+      ()) [13,14] ")"
+      variable_assignment [16,24] "owner=$2"
+        variable_name [16,21] "owner"
+        (=) [21,22] "="
+        simple_expansion [22,24] "$2"
+          ($) [22,23] "$"
+          variable_name [23,24] "2"
+      command [26,31] "shift"
+        command_name [26,31] "shift"
+          word [26,31] "shift"
+      (;;) [33,35] ";;"
+    case_item [36,42] "-g) ;;"
+      word [36,38] "-g"
+      ()) [38,39] ")"
+      (;;) [40,42] ";;"
+    (esac) [43,47] "esac"
+===
+@match: case x in *.txt) ;; esac
+case x in *.txt) ;; esac
+===
+@match: case x in [abc]) ;; esac
+case x in [abc]) ;; esac
+===
+@match: case x in a|b) ;; esac
+case x in a|b) ;; esac
+===
+@match: case x in -o|--output) ;; esac
+case x in -o|--output) ;; esac
+===
+@match: case x in x) ;; esac
+case x in x) ;; esac
+===
+@match: case x in foo.txt) ;; esac
+case x in foo.txt) ;; esac
+===
+@match: case x in *foo) ;; esac
+case x in *foo) ;; esac
+===
+@match: case x in foo*) ;; esac
+case x in foo*) ;; esac
+===
+@match: case x in *) ;; esac
+case x in *) ;; esac
+===
+@match: case x in ?) ;; esac
+case x in ?) ;; esac
+===
+@match: case x in *.*) ;; esac
+case x in *.*) ;; esac
+===
+@match: case x in foo|bar) ;; esac
+case x in foo|bar) ;; esac
+===
+@match: case x in foo.txt|*.log) ;; esac
+case x in foo.txt|*.log) ;; esac
+===
+@match: case x in [a-z]*) ;; esac
+case x in [a-z]*) ;; esac
+===
+@match: case x in *.[1357]) ;; esac
+case x in *.[1357]) ;; esac
+===
+@match: case x in -o) ;; esac
+case x in -o) ;; esac
+===
+@match: case x in --output) ;; esac
+case x in --output) ;; esac
+===
+@match: case x in +(a|b)) ;; esac
+case x in +(a|b)) ;; esac
+===
+@match: case x in *(a|b)) ;; esac
+case x in *(a|b)) ;; esac
+===
+@match: case x in @(a|b)) ;; esac
+case x in @(a|b)) ;; esac
+===
+@known-diff case-rejected-group: case x in x@(y)) ;; esac
+case x in x@(y)) ;; esac
+---
+hasError: true
+program [0,24] "case x in x@(y)) ;; esac"
+  case_statement [0,24] "case x in x@(y)) ;; esac"
+    (case) [0,4] "case"
+    word [5,6] "x"
+    (in) [7,9] "in"
+    case_item [10,19] "x@(y)) ;;"
+      ERROR [10,15] "x@(y)"
+      ()) [15,16] ")"
+      (;;) [17,19] ";;"
+    (esac) [20,24] "esac"
+===
+@match: case x in *([0-9])([0-9])) ;; esac
+case x in *([0-9])([0-9])) ;; esac
+===
+@match: case x in *"s"*) ;; esac
+case x in *"s"*) ;; esac
+===
+@match: case x in *${x}*) ;; esac
+case x in *${x}*) ;; esac
+===
+@match: case x in ${x}) ;; esac
+case x in ${x}) ;; esac
+===
+@match: case x in ?(a)) ;; esac
+case x in ?(a)) ;; esac
+===
+@match: case x in !(a)) ;; esac
+case x in !(a)) ;; esac
+===
+@match: case x in a\ b) ;; esac
+case x in a\ b) ;; esac
+===
+@match: case $x in foo_bar) ;; esac
+case $x in foo_bar) ;; esac
+===
+@match: case $x in word1) ;; esac
+case $x in word1) ;; esac
+===
+@match: case $x in abc1d) ;; esac
+case $x in abc1d) ;; esac
+===
+@match: case $x in v2.0) ;; esac
+case $x in v2.0) ;; esac
+===
+@match: case $x in a1) ;; esac
+case $x in a1) ;; esac
+===
+@match: case $x in a_) ;; esac
+case $x in a_) ;; esac
+===
+@match: case $x in A1) ;; esac
+case $x in A1) ;; esac
+===
+@match: case $x in a-b) ;; esac
+case $x in a-b) ;; esac
+===
+@match: case $x in a-bc) ;; esac
+case $x in a-bc) ;; esac
+===
+@match: case $x in ab-c) ;; esac
+case $x in ab-c) ;; esac
+===
+@match: case $x in a-b-c) ;; esac
+case $x in a-b-c) ;; esac
+===
+@match: case $x in cmake_modules) ;; esac
+case $x in cmake_modules) ;; esac
+===
+@match: case $x in a-b|y) ;; esac
+case $x in a-b|y) ;; esac
+===
+@match: case $x in abc|y) ;; esac
+case $x in abc|y) ;; esac
+===
+@match: case $x in abc | y) ;; esac
+case $x in abc | y) ;; esac
+===
+@match: case $x in -x-) ;; esac
+case $x in -x-) ;; esac
+===
+@match: case $x in -x-y) ;; esac
+case $x in -x-y) ;; esac
+===
+@known-diff case-rejected-group: case $x in *.(a|b)) ;; esac
+case $x in *.(a|b)) ;; esac
+---
+hasError: true
+program [0,27] "case $x in *.(a|b)) ;; esac"
+  case_statement [0,27] "case $x in *.(a|b)) ;; esac"
+    (case) [0,4] "case"
+    simple_expansion [5,7] "$x"
+      ($) [5,6] "$"
+      variable_name [6,7] "x"
+    (in) [8,10] "in"
+    case_item [11,22] "*.(a|b)) ;;"
+      ERROR [11,18] "*.(a|b)"
+      ()) [18,19] ")"
+      (;;) [20,22] ";;"
+    (esac) [23,27] "esac"
+===
+@known-diff case-rejected-group: case $x in x.(y)) ;; esac
+case $x in x.(y)) ;; esac
+---
+hasError: true
+program [0,25] "case $x in x.(y)) ;; esac"
+  case_statement [0,25] "case $x in x.(y)) ;; esac"
+    (case) [0,4] "case"
+    simple_expansion [5,7] "$x"
+      ($) [5,6] "$"
+      variable_name [6,7] "x"
+    (in) [8,10] "in"
+    case_item [11,20] "x.(y)) ;;"
+      ERROR [11,16] "x.(y)"
+      ()) [16,17] ")"
+      (;;) [18,20] ";;"
+    (esac) [21,25] "esac"
+===
+@known-diff case-pattern-after-continuation: case $x in cont\<LF>ued) ;; esac
+case $x in cont\
+ued) ;; esac
+---
+hasError: false
+program [0,29] "case $x in cont\\\nued) ;; esac"
+  case_statement [0,29] "case $x in cont\\\nued) ;; esac"
+    (case) [0,4] "case"
+    simple_expansion [5,7] "$x"
+      ($) [5,6] "$"
+      variable_name [6,7] "x"
+    (in) [8,10] "in"
+    case_item [11,24] "cont\\\nued) ;;"
+      word [11,20] "cont\\\nued"
+      ()) [20,21] ")"
+      (;;) [22,24] ";;"
+    (esac) [25,29] "esac"
+===
+@known-diff case-pattern-after-continuation: case $x in a\<LF>1) ;; esac
+case $x in a\
+1) ;; esac
+---
+hasError: false
+program [0,24] "case $x in a\\\n1) ;; esac"
+  case_statement [0,24] "case $x in a\\\n1) ;; esac"
+    (case) [0,4] "case"
+    simple_expansion [5,7] "$x"
+      ($) [5,6] "$"
+      variable_name [6,7] "x"
+    (in) [8,10] "in"
+    case_item [11,19] "a\\\n1) ;;"
+      word [11,15] "a\\\n1"
+      ()) [15,16] ")"
+      (;;) [17,19] ";;"
+    (esac) [20,24] "esac"
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/expansions.txt b/packages/tree-sitter-bash/test/fixtures/differential/expansions.txt
new file mode 100644
index 0000000000..e7394998a7
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/expansions.txt
@@ -0,0 +1,599 @@
+@match: for f in a b c; do echo "$f"; done
+for f in a b c; do echo "$f"; done
+===
+@match: for f; do echo "$f"; done
+for f; do echo "$f"; done
+===
+@match: select opt in a b; do echo $opt; done
+select opt in a b; do echo $opt; done
+===
+@match: a[i+1]=$y
+a[i+1]=$y
+===
+@match: echo ${arr[1]} ${arr[@]}
+echo ${arr[1]} ${arr[@]}
+===
+@match: echo $'a\nb\t'
+echo $'a\nb\t'
+===
+@match: echo $"hello $USER"
+echo $"hello $USER"
+===
+@match: echo ${x##*/} ${x%%.*}
+echo ${x##*/} ${x%%.*}
+===
+@match: echo ${x/y/z} ${x//a/b}
+echo ${x/y/z} ${x//a/b}
+===
+@match: echo ${x^} ${x^^} ${x,} ${x,,}
+echo ${x^} ${x^^} ${x,} ${x,,}
+===
+@match: echo ${x@Q}
+echo ${x@Q}
+===
+@match: echo ${v:1:2} ${v:-default}
+echo ${v:1:2} ${v:-default}
+===
+@match: f() { arr=(1 2); echo ${arr[0]}; }
+f() { arr=(1 2); echo ${arr[0]}; }
+===
+@match: for i in $(seq 3); do echo $i; done
+for i in $(seq 3); do echo $i; done
+===
+@match: echo ${a[$i]} ${a[i+1]}
+echo ${a[$i]} ${a[i+1]}
+===
+@match: echo ${!prefix*} ${!name@}
+echo ${!prefix*} ${!name@}
+===
+@match: echo ${x##a b}
+echo ${x##a b}
+===
+@match: echo ${x#"pat"}
+echo ${x#"pat"}
+===
+@match: echo ${x/a b/c}
+echo ${x/a b/c}
+===
+@match: echo ${x/a/$v}
+echo ${x/a/$v}
+===
+@match: echo ${x/pat/rep/extra}
+echo ${x/pat/rep/extra}
+===
+@match: echo ${x:?msg} ${x:+alt} ${x:-d e f}
+echo ${x:?msg} ${x:+alt} ${x:-d e f}
+===
+@match: echo ${v:-1} ${v:=2x} ${v:-"d e"}
+echo ${v:-1} ${v:=2x} ${v:-"d e"}
+===
+@match: echo ${arr[@]:1:2}
+echo ${arr[@]:1:2}
+===
+@known-diff recovery-partial-nodes: echo ${s//"/x}
+echo ${s//"/x}
+---
+hasError: true
+program [0,14] "echo ${s//\"/x}"
+  command [0,14] "echo ${s//\"/x}"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    expansion [5,14] "${s//\"/x}"
+      (${) [5,7] "${"
+      variable_name [7,8] "s"
+      (//) [8,10] "//"
+      string [10,14] "\"/x}"
+        (") [10,11] "\""
+        string_content [11,14] "/x}"
+===
+@match: x=$"v"
+x=$"v"
+===
+@match: for f in $"a"; do x; done
+for f in $"a"; do x; done
+===
+@match: echo a$"b"
+echo a$"b"
+===
+@match: echo $"a""b"
+echo $"a""b"
+===
+@match: echo x=$"v"
+echo x=$"v"
+===
+@match: echo $0 ${0} $1 ${10}
+echo $0 ${0} $1 ${10}
+===
+@match: echo ${x: -5}
+echo ${x: -5}
+===
+@match: echo ${x: -5:2}
+echo ${x: -5:2}
+===
+@known-diff recovery-partial-nodes: ${x@}
+${x@}
+---
+hasError: true
+program [0,5] "${x@}"
+  command [0,5] "${x@}"
+    command_name [0,5] "${x@}"
+      expansion [0,5] "${x@}"
+        (${) [0,2] "${"
+        variable_name [2,3] "x"
+        word [3,4] "@"
+        (}) [4,5] "}"
+===
+@match: echo ${x[1]}
+echo ${x[1]}
+===
+@match: echo "hello $NAME" 'raw'
+echo "hello $NAME" 'raw'
+===
+@match: echo `hostname` $(date)
+echo `hostname` $(date)
+===
+@match: echo ${v:-d} ${x} $1 $@
+echo ${v:-d} ${x} $1 $@
+===
+@match: echo $HOME
+echo $HOME
+===
+@match: echo a"b"c'd'$e${f}g
+echo a"b"c'd'$e${f}g
+===
+@match: echo $v
+echo $v
+===
+@match: echo "${a} middle ${b}"
+echo "${a} middle ${b}"
+===
+@match: echo "$@"
+echo "$@"
+===
+@match: echo $(a $(b c))
+echo $(a $(b c))
+===
+@match: echo ${#v}
+echo ${#v}
+===
+@match: echo ${v:=word}
+echo ${v:=word}
+===
+@known-diff empty-command-substitution: echo $( )
+echo $( )
+---
+hasError: false
+program [0,9] "echo $( )"
+  command [0,9] "echo $( )"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    command_substitution [5,9] "$( )"
+      ($() [5,7] "$("
+      ()) [8,9] ")"
+===
+@known-diff empty-backtick: echo ``
+echo ``
+---
+hasError: false
+program [0,7] "echo ``"
+  command [0,7] "echo ``"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    command_substitution [5,7] "``"
+      (`) [5,6] "`"
+      (`) [6,7] "`"
+===
+@match: echo $
+echo $
+===
+@match: echo "a\$b"
+echo "a\$b"
+===
+@known-diff string-content-newlines: echo "line1
+echo "line1
+line2"
+---
+hasError: false
+program [0,18] "echo \"line1\nline2\""
+  command [0,18] "echo \"line1\nline2\""
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    string [5,18] "\"line1\nline2\""
+      (") [5,6] "\""
+      string_content [6,17] "line1\nline2"
+      (") [17,18] "\""
+===
+@match: echo ${v:-$(cmd)}
+echo ${v:-$(cmd)}
+===
+@match: echo ${a[0]}
+echo ${a[0]}
+===
+@match: echo ${}
+echo ${}
+===
+@match: echo ${#}
+echo ${#}
+===
+@match: echo ${var1#*#}
+echo ${var1#*#}
+===
+@match: echo ${!abc}
+echo ${!abc}
+===
+@match: echo ${abc:- }
+echo ${abc:- }
+===
+@match: echo ${abc:
+echo ${abc:
+}
+===
+@match: echo ${abc,?}
+echo ${abc,?}
+===
+@match: echo ${abc^^b}
+echo ${abc^^b}
+===
+@match: echo ${abc@U}
+echo ${abc@U}
+===
+@match: echo ${abc:- -quiet}
+echo ${abc:- -quiet}
+===
+@match: A="${B[0]# }"
+A="${B[0]# }"
+===
+@match: C="${D/#* -E /}"
+C="${D/#* -E /}"
+===
+@match: F="${G%% *}"
+F="${G%% *}"
+===
+@match: H="${I#*;}"
+H="${I#*;}"
+===
+@match: R="${S%()}"
+R="${S%()}"
+===
+@match: F="${#!}"
+F="${#!}"
+===
+@match: G=${H,,[I]}
+G=${H,,[I]}
+===
+@match: L="${M/'N'*/O}"
+L="${M/'N'*/O}"
+===
+@match: A=${B//:;;/$'\n'}
+A=${B//:;;/$'\n'}
+===
+@match: C=${D/;\ *;|}
+C=${D/;\ *;|}
+===
+@match: MOFILES=${LINGUAS// /.po }.po
+MOFILES=${LINGUAS// /.po }.po
+===
+@match: MY_P="${PN/aspell/aspell"${ASPELL_VERSION}"}"
+MY_P="${PN/aspell/aspell"${ASPELL_VERSION}"}"
+===
+@match: pyc=${pyc//*\/}
+pyc=${pyc//*\/}
+===
+@match: ${pv/\.}
+${pv/\.}
+===
+@match: ${new_test_cp//"${old_ver_cp}"/}
+${new_test_cp//"${old_ver_cp}"/}
+===
+@known-diff expansion-replacement-escaped-backslash: ${allarchives// /\\|}
+${allarchives// /\\|}
+---
+hasError: false
+program [0,21] "${allarchives// /\\\\|}"
+  command [0,21] "${allarchives// /\\\\|}"
+    command_name [0,21] "${allarchives// /\\\\|}"
+      expansion [0,21] "${allarchives// /\\\\|}"
+        (${) [0,2] "${"
+        variable_name [2,13] "allarchives"
+        (//) [13,15] "//"
+        regex [15,16] " "
+        (/) [16,17] "/"
+        word [17,20] "\\\\|"
+        (}) [20,21] "}"
+===
+@match: ${!#}
+${!#}
+===
+@known-diff expansion-bang-hash-special: ${!# }
+${!# }
+---
+hasError: true
+program [0,6] "${!# }"
+  command [0,6] "${!# }"
+    command_name [0,6] "${!# }"
+      expansion [0,6] "${!# }"
+        (${) [0,2] "${"
+        (!) [2,3] "!"
+        special_variable_name [3,4] "#"
+        word [4,5] " "
+        (}) [5,6] "}"
+===
+@match: ${!##}
+${!##}
+===
+@known-diff expansion-bang-hash-special: ${!## }
+${!## }
+---
+hasError: true
+program [0,7] "${!## }"
+  command [0,7] "${!## }"
+    command_name [0,7] "${!## }"
+      expansion [0,7] "${!## }"
+        (${) [0,2] "${"
+        (!) [2,3] "!"
+        (#) [3,4] "#"
+        special_variable_name [4,5] "#"
+        word [5,6] " "
+        (}) [6,7] "}"
+===
+@known-diff expansion-bang-hash-special: ${!##/}
+${!##/}
+---
+hasError: false
+program [0,7] "${!##/}"
+  command [0,7] "${!##/}"
+    command_name [0,7] "${!##/}"
+      expansion [0,7] "${!##/}"
+        (${) [0,2] "${"
+        (!) [2,3] "!"
+        (#) [3,4] "#"
+        special_variable_name [4,5] "#"
+        (/) [5,6] "/"
+        (}) [6,7] "}"
+===
+@match: echo "${kw}? ( ${cond:+${cond}? (} ${baseuri}-${ver}-${kw...
+echo "${kw}? ( ${cond:+${cond}? (} ${baseuri}-${ver}-${kw}.${suff} ${cond:+) })"
+===
+@match: ${completions[*]}
+${completions[*]}
+===
+@known-diff expansion-equals-operator: ${=1}
+${=1}
+---
+hasError: false
+program [0,5] "${=1}"
+  command [0,5] "${=1}"
+    command_name [0,5] "${=1}"
+      expansion [0,5] "${=1}"
+        (${) [0,2] "${"
+        (=) [2,3] "="
+        word [3,4] "1"
+        (}) [4,5] "}"
+===
+@match: ${2?}
+${2?}
+===
+@match: ${p_key#*=}
+${p_key#*=}
+===
+@match: ${B[0]# }
+${B[0]# }
+===
+@match: ${to_enables[0]##*/}
+${to_enables[0]##*/}
+===
+@match: exec "${0#-}" --rcfile "${BASH_IT_BASHRC:-${HOME?}/.bashrc}"
+exec "${0#-}" --rcfile "${BASH_IT_BASHRC:-${HOME?}/.bashrc}"
+===
+@match: recho "TDEFAULTS = ${selvecs:+-DSELECT_VECS=\"$selvecs\"}"
+recho "TDEFAULTS = ${selvecs:+-DSELECT_VECS=\"$selvecs\"}"
+===
+@match: local msg="${2:-command '$1' does not exist}"
+local msg="${2:-command '$1' does not exist}"
+===
+@match: ${cdir:+#}
+${cdir:+#}
+===
+@match: ${dict_langs:+;}
+${dict_langs:+;}
+===
+@match: ${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/util-linux )}
+${UTIL_LINUX_LIBC[@]/%/? ( sys-apps/util-linux )}
+===
+@match: ${id}${2+ ${2}}
+${id}${2+ ${2}}
+===
+@match: ${BRANDING_GCC_PKGVERSION/(/(Gentoo ${PVR}${extvers}, }
+${BRANDING_GCC_PKGVERSION/(/(Gentoo ${PVR}${extvers}, }
+===
+@match: BASH_IT_GIT_URL="${BASH_IT_GIT_URL/git@/https://}"
+BASH_IT_GIT_URL="${BASH_IT_GIT_URL/git@/https://}"
+===
+@known-diff number-base-with-expansion: 10#${command_start##*.}
+10#${command_start##*.}
+---
+hasError: false
+program [0,23] "10#${command_start##*.}"
+  command [0,23] "10#${command_start##*.}"
+    command_name [0,23] "10#${command_start##*.}"
+      concatenation [0,23] "10#${command_start##*.}"
+        word [0,3] "10#"
+        expansion [3,23] "${command_start##*.}"
+          (${) [3,5] "${"
+          variable_name [5,18] "command_start"
+          (##) [18,20] "##"
+          regex [20,22] "*."
+          (}) [22,23] "}"
+===
+@match: echo ${LIB_DEPEND//\[static-libs(+)]}
+echo ${LIB_DEPEND//\[static-libs(+)]}
+===
+@match: ${ALL_LLVM_TARGETS[@]/%/(-)?}
+${ALL_LLVM_TARGETS[@]/%/(-)?}
+===
+@match: filterdiff -p1 ${paths[@]/#/-i }
+filterdiff -p1 ${paths[@]/#/-i }
+===
+@match: ${cflags//-O? /$(get-flag O) }
+${cflags//-O? /$(get-flag O) }
+===
+@match: curf="${f%'-roff2html'*}.html"
+curf="${f%'-roff2html'*}.html"
+===
+@match: reff="${f/'-roff2html'*/'-ref'}.html"
+reff="${f/'-roff2html'*/'-ref'}.html"
+===
+@match: echo "s/$/'/"
+echo "s/$/'/"
+===
+@match: echo "s$"
+echo "s$"
+===
+@match: echo $'Here\'s Johnny!\r\n'
+echo $'Here\'s Johnny!\r\n'
+===
+@known-diff expansion-default-array: printf "  %-9s" "${seq0:-(default)}"
+printf "  %-9s" "${seq0:-(default)}"
+---
+hasError: false
+program [0,36] "printf \"  %-9s\" \"${seq0:-(default)}\""
+  command [0,36] "printf \"  %-9s\" \"${seq0:-(default)}\""
+    command_name [0,6] "printf"
+      word [0,6] "printf"
+    string [7,15] "\"  %-9s\""
+      (") [7,8] "\""
+      string_content [8,14] "  %-9s"
+      (") [14,15] "\""
+    string [16,36] "\"${seq0:-(default)}\""
+      (") [16,17] "\""
+      expansion [17,35] "${seq0:-(default)}"
+        (${) [17,19] "${"
+        variable_name [19,23] "seq0"
+        (:-) [23,25] ":-"
+        word [25,34] "(default)"
+        (}) [34,35] "}"
+      (") [35,36] "\""
+===
+@known-diff empty-backtick: main() {
+main() {
+    local foo="asd"`
+        `"fgh"
+}
+---
+hasError: false
+program [0,46] "main() {\n    local foo=\"asd\"`\n       ..."
+  function_definition [0,46] "main() {\n    local foo=\"asd\"`\n       ..."
+    word [0,4] "main"
+    (() [4,5] "("
+    ()) [5,6] ")"
+    compound_statement [7,46] "{\n    local foo=\"asd\"`\n        `\"fgh\"\n}"
+      ({) [7,8] "{"
+      declaration_command [13,44] "local foo=\"asd\"`\n        `\"fgh\""
+        (local) [13,18] "local"
+        variable_assignment [19,44] "foo=\"asd\"`\n        `\"fgh\""
+          variable_name [19,22] "foo"
+          (=) [22,23] "="
+          concatenation [23,44] "\"asd\"`\n        `\"fgh\""
+            string [23,28] "\"asd\""
+              (") [23,24] "\""
+              string_content [24,27] "asd"
+              (") [27,28] "\""
+            command_substitution [28,39] "`\n        `"
+              (`) [28,29] "`"
+              (`) [38,39] "`"
+            string [39,44] "\"fgh\""
+              (") [39,40] "\""
+              string_content [40,43] "fgh"
+              (") [43,44] "\""
+      (}) [45,46] "}"
+===
+@known-diff dollar-backtick-substitution: $(eval echo $`echo ${foo}`)
+$(eval echo $`echo ${foo}`)
+---
+hasError: false
+program [0,27] "$(eval echo $`echo ${foo}`)"
+  command [0,27] "$(eval echo $`echo ${foo}`)"
+    command_name [0,27] "$(eval echo $`echo ${foo}`)"
+      command_substitution [0,27] "$(eval echo $`echo ${foo}`)"
+        ($() [0,2] "$("
+        command [2,26] "eval echo $`echo ${foo}`"
+          command_name [2,6] "eval"
+            word [2,6] "eval"
+          word [7,11] "echo"
+          concatenation [12,26] "$`echo ${foo}`"
+            ($) [12,13] "$"
+            command_substitution [13,26] "`echo ${foo}`"
+              (`) [13,14] "`"
+              command [14,25] "echo ${foo}"
+                command_name [14,18] "echo"
+                  word [14,18] "echo"
+                expansion [19,25] "${foo}"
+                  (${) [19,21] "${"
+                  variable_name [21,24] "foo"
+                  (}) [24,25] "}"
+              (`) [25,26] "`"
+        ()) [26,27] ")"
+===
+@match: echo `echo otherword`word
+echo `echo otherword`word
+===
+@match: echo word`echo otherword`
+echo word`echo otherword`
+===
+@match: assert()
+assert()
+  if ! $1; then
+    return 1
+  fi
+===
+@match: echo " ${x}"
+echo " ${x}"
+===
+@match: echo "a ${x} b"
+echo "a ${x} b"
+===
+@match: echo "$x "
+echo "$x "
+===
+@match: echo "${x} "
+echo "${x} "
+===
+@match: echo ${#@}
+echo ${#@}
+===
+@match: echo ${#*}
+echo ${#*}
+===
+@match: echo ${#-}
+echo ${#-}
+===
+@match: echo ${#x[@]}
+echo ${#x[@]}
+===
+@match: echo ${x/'abc'/y}
+echo ${x/'abc'/y}
+===
+@match: echo ${x/"abc"/y}
+echo ${x/"abc"/y}
+===
+@match: echo ${x/'a'*'b'/y}
+echo ${x/'a'*'b'/y}
+===
+@match: echo ${x##'a'}
+echo ${x##'a'}
+===
+@match: echo ${x%%a b}
+echo ${x%%a b}
+===
+@match: echo "${x} $(y)"
+echo "${x} $(y)"
+===
+@match: echo " $(y)"
+echo " $(y)"
+===
+@match: echo " $x"
+echo " $x"
+===
+@match: echo "${x} $y"
+echo "${x} $y"
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/heredoc.txt b/packages/tree-sitter-bash/test/fixtures/differential/heredoc.txt
new file mode 100644
index 0000000000..1ac604b146
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/heredoc.txt
@@ -0,0 +1,597 @@
+@match: echo $((x << 2 | 1))
+echo $((x << 2 | 1))
+===
+@known-diff heredoc-content-chunks: foo() { cat <<EOF
+foo() { cat <<EOF
+body $x
+EOF
+}
+---
+hasError: false
+program [0,31] "foo() { cat <<EOF\nbody $x\nEOF\n}"
+  function_definition [0,31] "foo() { cat <<EOF\nbody $x\nEOF\n}"
+    word [0,3] "foo"
+    (() [3,4] "("
+    ()) [4,5] ")"
+    compound_statement [6,31] "{ cat <<EOF\nbody $x\nEOF\n}"
+      ({) [6,7] "{"
+      redirected_statement [8,29] "cat <<EOF\nbody $x\nEOF"
+        command [8,11] "cat"
+          command_name [8,11] "cat"
+            word [8,11] "cat"
+        heredoc_redirect [12,29] "<<EOF\nbody $x\nEOF"
+          (<<) [12,14] "<<"
+          heredoc_start [14,17] "EOF"
+          heredoc_body [18,26] "body $x\n"
+            heredoc_content [18,23] "body "
+            simple_expansion [23,25] "$x"
+              ($) [23,24] "$"
+              variable_name [24,25] "x"
+            heredoc_content [25,26] "\n"
+          heredoc_end [26,29] "EOF"
+      (}) [30,31] "}"
+===
+@known-diff recovery-partial-nodes: if cat <<EOF; then x; fi
+if cat <<EOF; then x; fi
+body
+EOF
+---
+hasError: true
+program [0,33] "if cat <<EOF; then x; fi\nbody\nEOF"
+  if_statement [0,33] "if cat <<EOF; then x; fi\nbody\nEOF"
+    (if) [0,2] "if"
+    redirected_statement [3,33] "cat <<EOF; then x; fi\nbody\nEOF"
+      command [3,6] "cat"
+        command_name [3,6] "cat"
+          word [3,6] "cat"
+      heredoc_redirect [7,33] "<<EOF; then x; fi\nbody\nEOF"
+        (<<) [7,9] "<<"
+        heredoc_start [9,12] "EOF"
+        (;) [12,13] ";"
+        command [14,20] "then x"
+          command_name [14,18] "then"
+            word [14,18] "then"
+          word [19,20] "x"
+        (;) [20,21] ";"
+        command [22,24] "fi"
+          command_name [22,24] "fi"
+            word [22,24] "fi"
+        heredoc_body [25,30] "body\n"
+          heredoc_content [25,30] "body\n"
+        heredoc_end [30,33] "EOF"
+===
+@match: cmd <<< "$str"
+cmd <<< "$str"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+hello $USER
+EOF
+---
+hasError: false
+program [0,25] "cat <<EOF\nhello $USER\nEOF"
+  redirected_statement [0,25] "cat <<EOF\nhello $USER\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,25] "<<EOF\nhello $USER\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,22] "hello $USER\n"
+        heredoc_content [10,16] "hello "
+        simple_expansion [16,21] "$USER"
+          ($) [16,17] "$"
+          variable_name [17,21] "USER"
+        heredoc_content [21,22] "\n"
+      heredoc_end [22,25] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<-EOF
+cat <<-EOF
+	indented
+	EOF
+---
+hasError: false
+program [0,25] "cat <<-EOF\n\tindented\n\tEOF"
+  redirected_statement [0,25] "cat <<-EOF\n\tindented\n\tEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,25] "<<-EOF\n\tindented\n\tEOF"
+      (<<-) [4,7] "<<-"
+      heredoc_start [7,10] "EOF"
+      heredoc_body [12,22] "indented\n\t"
+        heredoc_content [12,22] "indented\n\t"
+      heredoc_end [22,25] "EOF"
+===
+@match: cat <<'EOF'
+cat <<'EOF'
+raw $notexpanded
+EOF
+===
+@known-diff heredoc-multiple-on-line: cat <<A <<B
+cat <<A <<B
+body a
+A
+body b
+B
+---
+hasError: true
+program [0,29] "cat <<A <<B\nbody a\nA\nbody b\nB"
+  redirected_statement [0,20] "cat <<A <<B\nbody a\nA"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,20] "<<A <<B\nbody a\nA"
+      (<<) [4,6] "<<"
+      heredoc_start [6,7] "A"
+      ERROR [8,11] "<<B"
+        (<<) [8,10] "<<"
+        word [10,11] "B"
+      heredoc_body [12,19] "body a\n"
+        heredoc_content [12,19] "body a\n"
+      heredoc_end [19,20] "A"
+  command [21,27] "body b"
+    command_name [21,25] "body"
+      word [21,25] "body"
+    word [26,27] "b"
+  command [28,29] "B"
+    command_name [28,29] "B"
+      word [28,29] "B"
+===
+@match: echo $((1<<2))
+echo $((1<<2))
+===
+@known-diff recovery-partial-nodes: cat <<EOF
+cat <<EOF
+no end marker
+---
+hasError: true
+program [0,23] "cat <<EOF\nno end marker"
+  redirected_statement [0,23] "cat <<EOF\nno end marker"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,23] "<<EOF\nno end marker"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,23] "no end marker"
+        heredoc_content [10,23] "no end marker"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+hello $USER and `date` ${x}
+done
+EOF
+---
+hasError: false
+program [0,46] "cat <<EOF\nhello $USER and `date` ${x}..."
+  redirected_statement [0,46] "cat <<EOF\nhello $USER and `date` ${x}..."
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,46] "<<EOF\nhello $USER and `date` ${x}\ndon..."
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,43] "hello $USER and `date` ${x}\ndone\n"
+        heredoc_content [10,16] "hello "
+        simple_expansion [16,21] "$USER"
+          ($) [16,17] "$"
+          variable_name [17,21] "USER"
+        heredoc_content [21,26] " and "
+        command_substitution [26,32] "`date`"
+          (`) [26,27] "`"
+          command [27,31] "date"
+            command_name [27,31] "date"
+              word [27,31] "date"
+          (`) [31,32] "`"
+        heredoc_content [32,33] " "
+        expansion [33,37] "${x}"
+          (${) [33,35] "${"
+          variable_name [35,36] "x"
+          (}) [36,37] "}"
+        heredoc_content [37,43] "\ndone\n"
+      heredoc_end [43,46] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+plain only
+EOF
+---
+hasError: false
+program [0,24] "cat <<EOF\nplain only\nEOF"
+  redirected_statement [0,24] "cat <<EOF\nplain only\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,24] "<<EOF\nplain only\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,21] "plain only\n"
+        heredoc_content [10,21] "plain only\n"
+      heredoc_end [21,24] "EOF"
+===
+@match: cat <<-"EOF"
+cat <<-"EOF"
+	x
+	EOF
+===
+@known-diff heredoc-content-chunks: cat <<-EOF
+cat <<-EOF
+		a
+	b
+		EOF
+---
+hasError: false
+program [0,23] "cat <<-EOF\n\t\ta\n\tb\n\t\tEOF"
+  redirected_statement [0,23] "cat <<-EOF\n\t\ta\n\tb\n\t\tEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,23] "<<-EOF\n\t\ta\n\tb\n\t\tEOF"
+      (<<-) [4,7] "<<-"
+      heredoc_start [7,10] "EOF"
+      heredoc_body [13,20] "a\n\tb\n\t\t"
+        heredoc_content [13,20] "a\n\tb\n\t\t"
+      heredoc_end [20,23] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+$(cmd) `other` $v ${w}
+EOF
+---
+hasError: false
+program [0,36] "cat <<EOF\n$(cmd) `other` $v ${w}\nEOF"
+  redirected_statement [0,36] "cat <<EOF\n$(cmd) `other` $v ${w}\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,36] "<<EOF\n$(cmd) `other` $v ${w}\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,33] "$(cmd) `other` $v ${w}\n"
+        command_substitution [10,16] "$(cmd)"
+          ($() [10,12] "$("
+          command [12,15] "cmd"
+            command_name [12,15] "cmd"
+              word [12,15] "cmd"
+          ()) [15,16] ")"
+        heredoc_content [16,17] " "
+        command_substitution [17,24] "`other`"
+          (`) [17,18] "`"
+          command [18,23] "other"
+            command_name [18,23] "other"
+              word [18,23] "other"
+          (`) [23,24] "`"
+        heredoc_content [24,25] " "
+        simple_expansion [25,27] "$v"
+          ($) [25,26] "$"
+          variable_name [26,27] "v"
+        heredoc_content [27,28] " "
+        expansion [28,32] "${w}"
+          (${) [28,30] "${"
+          variable_name [30,31] "w"
+          (}) [31,32] "}"
+        heredoc_content [32,33] "\n"
+      heredoc_end [33,36] "EOF"
+===
+@known-diff heredoc-tail-statements: cat <<EOF arg1 arg2
+cat <<EOF arg1 arg2
+body
+EOF
+---
+hasError: false
+program [0,28] "cat <<EOF arg1 arg2\nbody\nEOF"
+  redirected_statement [0,28] "cat <<EOF arg1 arg2\nbody\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,28] "<<EOF arg1 arg2\nbody\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      word [10,14] "arg1"
+      word [15,19] "arg2"
+      heredoc_body [20,25] "body\n"
+        heredoc_content [20,25] "body\n"
+      heredoc_end [25,28] "EOF"
+===
+@known-diff heredoc-tail-statements: cat <<EOF > out
+cat <<EOF > out
+body
+EOF
+---
+hasError: false
+program [0,24] "cat <<EOF > out\nbody\nEOF"
+  redirected_statement [0,24] "cat <<EOF > out\nbody\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,24] "<<EOF > out\nbody\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      file_redirect [10,15] "> out"
+        (>) [10,11] ">"
+        word [12,15] "out"
+      heredoc_body [16,21] "body\n"
+        heredoc_content [16,21] "body\n"
+      heredoc_end [21,24] "EOF"
+===
+@known-diff heredoc-tail-statements: cat <<EOF | grep x
+cat <<EOF | grep x
+body
+EOF
+---
+hasError: false
+program [0,27] "cat <<EOF | grep x\nbody\nEOF"
+  redirected_statement [0,27] "cat <<EOF | grep x\nbody\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,27] "<<EOF | grep x\nbody\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      pipeline [10,18] "| grep x"
+        (|) [10,11] "|"
+        command [12,18] "grep x"
+          command_name [12,16] "grep"
+            word [12,16] "grep"
+          word [17,18] "x"
+      heredoc_body [19,24] "body\n"
+        heredoc_content [19,24] "body\n"
+      heredoc_end [24,27] "EOF"
+===
+@known-diff heredoc-tail-statements: cat <<EOF && echo done
+cat <<EOF && echo done
+body
+EOF
+---
+hasError: false
+program [0,31] "cat <<EOF && echo done\nbody\nEOF"
+  redirected_statement [0,31] "cat <<EOF && echo done\nbody\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,31] "<<EOF && echo done\nbody\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      (&&) [10,12] "&&"
+      command [13,22] "echo done"
+        command_name [13,17] "echo"
+          word [13,17] "echo"
+        word [18,22] "done"
+      heredoc_body [23,28] "body\n"
+        heredoc_content [23,28] "body\n"
+      heredoc_end [28,31] "EOF"
+===
+@match: cmd <<< x > out
+cmd <<< x > out
+===
+@known-diff heredoc-multiple-on-line: cat <<A <<B
+cat <<A <<B
+ba
+A
+bb
+B
+---
+hasError: true
+program [0,21] "cat <<A <<B\nba\nA\nbb\nB"
+  redirected_statement [0,16] "cat <<A <<B\nba\nA"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,16] "<<A <<B\nba\nA"
+      (<<) [4,6] "<<"
+      heredoc_start [6,7] "A"
+      ERROR [8,11] "<<B"
+        (<<) [8,10] "<<"
+        word [10,11] "B"
+      heredoc_body [12,15] "ba\n"
+        heredoc_content [12,15] "ba\n"
+      heredoc_end [15,16] "A"
+  command [17,19] "bb"
+    command_name [17,19] "bb"
+      word [17,19] "bb"
+  command [20,21] "B"
+    command_name [20,21] "B"
+      word [20,21] "B"
+===
+@known-diff heredoc-tail-statements: cat <<EOF; echo x
+cat <<EOF; echo x
+body
+EOF
+---
+hasError: false
+program [0,26] "cat <<EOF; echo x\nbody\nEOF"
+  redirected_statement [0,26] "cat <<EOF; echo x\nbody\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,26] "<<EOF; echo x\nbody\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      (;) [9,10] ";"
+      command [11,17] "echo x"
+        command_name [11,15] "echo"
+          word [11,15] "echo"
+        word [16,17] "x"
+      heredoc_body [18,23] "body\n"
+        heredoc_content [18,23] "body\n"
+      heredoc_end [23,26] "EOF"
+===
+@known-diff heredoc-tail-statements: cat <<EOF &
+cat <<EOF &
+body
+EOF
+---
+hasError: false
+program [0,20] "cat <<EOF &\nbody\nEOF"
+  redirected_statement [0,20] "cat <<EOF &\nbody\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,20] "<<EOF &\nbody\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      (&) [10,11] "&"
+      heredoc_body [12,17] "body\n"
+        heredoc_content [12,17] "body\n"
+      heredoc_end [17,20] "EOF"
+===
+@known-diff heredoc-tail-statements: wc <<EOF | cat | sort
+wc <<EOF | cat | sort
+body
+EOF
+---
+hasError: false
+program [0,30] "wc <<EOF | cat | sort\nbody\nEOF"
+  redirected_statement [0,30] "wc <<EOF | cat | sort\nbody\nEOF"
+    command [0,2] "wc"
+      command_name [0,2] "wc"
+        word [0,2] "wc"
+    heredoc_redirect [3,30] "<<EOF | cat | sort\nbody\nEOF"
+      (<<) [3,5] "<<"
+      heredoc_start [5,8] "EOF"
+      pipeline [9,21] "| cat | sort"
+        (|) [9,10] "|"
+        pipeline [11,21] "cat | sort"
+          command [11,14] "cat"
+            command_name [11,14] "cat"
+              word [11,14] "cat"
+          (|) [15,16] "|"
+          command [17,21] "sort"
+            command_name [17,21] "sort"
+              word [17,21] "sort"
+      heredoc_body [22,27] "body\n"
+        heredoc_content [22,27] "body\n"
+      heredoc_end [27,30] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+# check out this regex '^EOF$'
+EOF
+---
+hasError: false
+program [0,44] "cat <<EOF\n# check out this regex '^EO..."
+  redirected_statement [0,44] "cat <<EOF\n# check out this regex '^EO..."
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,44] "<<EOF\n# check out this regex '^EOF$'\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,41] "# check out this regex '^EOF$'\n"
+        heredoc_content [10,41] "# check out this regex '^EOF$'\n"
+      heredoc_end [41,44] "EOF"
+===
+@match: a=$((6 % 7 ** 8 << 9 >> 10 & 11 | 12 ^ 13))
+a=$((6 % 7 ** 8 << 9 >> 10 & 11 | 12 ^ 13))
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+# comment
+EOF
+---
+hasError: false
+program [0,23] "cat <<EOF\n# comment\nEOF"
+  redirected_statement [0,23] "cat <<EOF\n# comment\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,23] "<<EOF\n# comment\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,20] "# comment\n"
+        heredoc_content [10,20] "# comment\n"
+      heredoc_end [20,23] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+has 'quote'
+EOF
+---
+hasError: false
+program [0,25] "cat <<EOF\nhas 'quote'\nEOF"
+  redirected_statement [0,25] "cat <<EOF\nhas 'quote'\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,25] "<<EOF\nhas 'quote'\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,22] "has 'quote'\n"
+        heredoc_content [10,22] "has 'quote'\n"
+      heredoc_end [22,25] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+has $dollar
+EOF
+---
+hasError: false
+program [0,25] "cat <<EOF\nhas $dollar\nEOF"
+  redirected_statement [0,25] "cat <<EOF\nhas $dollar\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,25] "<<EOF\nhas $dollar\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,22] "has $dollar\n"
+        heredoc_content [10,14] "has "
+        simple_expansion [14,21] "$dollar"
+          ($) [14,15] "$"
+          variable_name [15,21] "dollar"
+        heredoc_content [21,22] "\n"
+      heredoc_end [22,25] "EOF"
+===
+@known-diff heredoc-content-chunks: cat <<EOF
+cat <<EOF
+regex '^EOF$'
+EOF
+---
+hasError: false
+program [0,27] "cat <<EOF\nregex '^EOF$'\nEOF"
+  redirected_statement [0,27] "cat <<EOF\nregex '^EOF$'\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,27] "<<EOF\nregex '^EOF$'\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,24] "regex '^EOF$'\n"
+        heredoc_content [10,24] "regex '^EOF$'\n"
+      heredoc_end [24,27] "EOF"
+===
+@known-diff heredoc-body-dollar-quote: cat <<EOF
+cat <<EOF
+$'x'
+EOF
+---
+hasError: false
+program [0,18] "cat <<EOF\n$'x'\nEOF"
+  redirected_statement [0,18] "cat <<EOF\n$'x'\nEOF"
+    command [0,3] "cat"
+      command_name [0,3] "cat"
+        word [0,3] "cat"
+    heredoc_redirect [4,18] "<<EOF\n$'x'\nEOF"
+      (<<) [4,6] "<<"
+      heredoc_start [6,9] "EOF"
+      heredoc_body [10,15] "$'x'\n"
+        heredoc_content [10,15] "$'x'\n"
+      heredoc_end [15,18] "EOF"
+===
+@known-diff heredoc-at-statement-start: <<EOF cat
+<<EOF cat
+body
+EOF
+---
+hasError: false
+program [0,18] "<<EOF cat\nbody\nEOF"
+  redirected_statement [0,18] "<<EOF cat\nbody\nEOF"
+    heredoc_redirect [0,18] "<<EOF cat\nbody\nEOF"
+      (<<) [0,2] "<<"
+      heredoc_start [2,5] "EOF"
+      word [6,9] "cat"
+      heredoc_body [10,15] "body\n"
+        heredoc_content [10,15] "body\n"
+      heredoc_end [15,18] "EOF"
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/recovery.txt b/packages/tree-sitter-bash/test/fixtures/differential/recovery.txt
new file mode 100644
index 0000000000..b8814fc7be
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/recovery.txt
@@ -0,0 +1,112 @@
+@known-diff recovery-partial-nodes: foo () ls
+foo () ls
+---
+hasError: true
+program [0,9] "foo () ls"
+  command [0,9] "foo () ls"
+    command_name [0,3] "foo"
+      word [0,3] "foo"
+    subshell [4,6] "()"
+      (() [4,5] "("
+      ()) [5,6] ")"
+    word [7,9] "ls"
+===
+@known-diff recovery-partial-nodes: if a; then b
+if a; then b
+---
+hasError: true
+program [0,12] "if a; then b"
+  if_statement [0,12] "if a; then b"
+    (if) [0,2] "if"
+    command [3,4] "a"
+      command_name [3,4] "a"
+        word [3,4] "a"
+    (;) [4,5] ";"
+    (then) [6,10] "then"
+    command [11,12] "b"
+      command_name [11,12] "b"
+        word [11,12] "b"
+===
+@known-diff recovery-partial-nodes: if a; then b;
+if a; then b;
+---
+hasError: true
+program [0,13] "if a; then b;"
+  if_statement [0,13] "if a; then b;"
+    (if) [0,2] "if"
+    command [3,4] "a"
+      command_name [3,4] "a"
+        word [3,4] "a"
+    (;) [4,5] ";"
+    (then) [6,10] "then"
+    command [11,12] "b"
+      command_name [11,12] "b"
+        word [11,12] "b"
+    (;) [12,13] ";"
+===
+@known-diff recovery-partial-nodes: if a
+if a
+---
+hasError: true
+program [0,4] "if a"
+  if_statement [0,4] "if a"
+    (if) [0,2] "if"
+    command [3,4] "a"
+      command_name [3,4] "a"
+        word [3,4] "a"
+===
+@known-diff recovery-partial-nodes: if a; then b fi
+if a; then b fi
+---
+hasError: true
+program [0,15] "if a; then b fi"
+  if_statement [0,15] "if a; then b fi"
+    (if) [0,2] "if"
+    command [3,4] "a"
+      command_name [3,4] "a"
+        word [3,4] "a"
+    (;) [4,5] ";"
+    (then) [6,10] "then"
+    command [11,15] "b fi"
+      command_name [11,12] "b"
+        word [11,12] "b"
+      word [13,15] "fi"
+===
+@known-diff recovery-partial-nodes: { ls }
+{ ls }
+---
+hasError: true
+program [0,6] "{ ls }"
+  compound_statement [0,6] "{ ls }"
+    ({) [0,1] "{"
+    command [2,6] "ls }"
+      command_name [2,4] "ls"
+        word [2,4] "ls"
+      word [5,6] "}"
+===
+@known-diff recovery-partial-nodes: while x do y done
+while x do y done
+---
+hasError: true
+program [0,17] "while x do y done"
+  while_statement [0,17] "while x do y done"
+    (while) [0,5] "while"
+    command [6,17] "x do y done"
+      command_name [6,7] "x"
+        word [6,7] "x"
+      word [8,10] "do"
+      word [11,12] "y"
+      word [13,17] "done"
+===
+@known-diff recovery-partial-nodes: echo "unterminated
+echo "unterminated
+---
+hasError: true
+program [0,18] "echo \"unterminated"
+  command [0,18] "echo \"unterminated"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    string [5,18] "\"unterminated"
+      (") [5,6] "\""
+      string_content [6,18] "unterminated"
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/redirects.txt b/packages/tree-sitter-bash/test/fixtures/differential/redirects.txt
new file mode 100644
index 0000000000..eb68c0c127
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/redirects.txt
@@ -0,0 +1,120 @@
+@match: while read -r line; do echo "$line"; done < file
+while read -r line; do echo "$line"; done < file
+===
+@match: foo() { x; } > out
+foo() { x; } > out
+===
+@match: if a; then b; fi > out
+if a; then b; fi > out
+===
+@match: cat > $"out"
+cat > $"out"
+===
+@match: diff <(cmd) >(other)
+diff <(cmd) >(other)
+===
+@match: cmd > out 2>&1
+cmd > out 2>&1
+===
+@match: cmd &>> log
+cmd &>> log
+===
+@known-diff redirect-readwrite: exec 3<>file
+exec 3<>file
+---
+hasError: false
+program [0,12] "exec 3<>file"
+  redirected_statement [0,12] "exec 3<>file"
+    command [0,4] "exec"
+      command_name [0,4] "exec"
+        word [0,4] "exec"
+    file_redirect [5,12] "3<>file"
+      file_descriptor [5,6] "3"
+      (<>) [6,8] "<>"
+      word [8,12] "file"
+===
+@match: > out
+> out
+===
+@match: cmd 2>/dev/null
+cmd 2>/dev/null
+===
+@match: cmd >&-
+cmd >&-
+===
+@match: cmd >| file
+cmd >| file
+===
+@known-diff redirect-readwrite: cmd <>file
+cmd <>file
+---
+hasError: false
+program [0,10] "cmd <>file"
+  redirected_statement [0,10] "cmd <>file"
+    command [0,3] "cmd"
+      command_name [0,3] "cmd"
+        word [0,3] "cmd"
+    file_redirect [4,10] "<>file"
+      (<>) [4,6] "<>"
+      word [6,10] "file"
+===
+@match: cmd < in
+cmd < in
+===
+@match: cmd <&0-
+cmd <&0-
+===
+@match: cmd > out
+cmd > out
+===
+@match: cmd 2>&1 >/dev/null
+cmd 2>&1 >/dev/null
+===
+@match: > a cmd x
+> a cmd x
+===
+@match: (ls) > out
+(ls) > out
+===
+@match: ! ls > out
+! ls > out
+===
+@match: cmd > out arg
+cmd > out arg
+===
+@match: 2>&1 whoami
+2>&1 whoami
+===
+@match: echo "foobar" >&2
+echo "foobar" >&2
+===
+@match: exec {VIRTWL[0]} {VIRTWL[1]} <&- >&-
+exec {VIRTWL[0]} {VIRTWL[1]} <&- >&-
+===
+@match: exec {VIRTWL[0]}<&- {VIRTWL[1]}>&-
+exec {VIRTWL[0]}<&- {VIRTWL[1]}>&-
+===
+@match: x <x a b c
+x <x a b c
+===
+@match: Z="${A#*<B>}"
+Z="${A#*<B>}"
+===
+@match: C="${D%</E>*}"
+C="${D%</E>*}"
+===
+@match: some-command ${foo:+--arg <(printf '%s\n' "$foo")}
+some-command ${foo:+--arg <(printf '%s\n' "$foo")}
+===
+@match: echo $(< some-file)
+echo $(< some-file)
+===
+@match: exec {fd}<&-
+exec {fd}<&-
+===
+@match: exec {a[0]}<&- {b[1]}>&-
+exec {a[0]}<&- {b[1]}>&-
+===
+@match: exec {a[0]}<&-
+exec {a[0]}<&-
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/statements.txt b/packages/tree-sitter-bash/test/fixtures/differential/statements.txt
new file mode 100644
index 0000000000..24c53c5ea2
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/statements.txt
@@ -0,0 +1,391 @@
+@match: if true; then echo yes; fi
+if true; then echo yes; fi
+===
+@match: if a; then b; elif c; then d; else e; fi
+if a; then b; elif c; then d; else e; fi
+===
+@match: if a
+if a
+then
+  b
+fi
+===
+@match: until x; do y; done
+until x; do y; done
+===
+@match: foo() { echo hi; }
+foo() { echo hi; }
+===
+@match: function bar { echo hi; }
+function bar { echo hi; }
+===
+@match: function baz() { echo hi; return 0; }
+function baz() { echo hi; return 0; }
+===
+@match: foo() ( echo sub )
+foo() ( echo sub )
+===
+@match: { echo a; echo b; }
+{ echo a; echo b; }
+===
+@match: coproc myjob { echo hi; }
+coproc myjob { echo hi; }
+===
+@match: echo if for while
+echo if for while
+===
+@match: arr=(a b "c d")
+arr=(a b "c d")
+===
+@match: arr[0]=x
+arr[0]=x
+===
+@match: export FOO=bar BAZ
+export FOO=bar BAZ
+===
+@match: declare -r x=1
+declare -r x=1
+===
+@match: local y
+local y
+===
+@match: readonly z=2
+readonly z=2
+===
+@match: unset a b
+unset a b
+===
+@match: echo {1..10}
+echo {1..10}
+===
+@match: echo {1..10..2}
+echo {1..10..2}
+===
+@match: echo {a..z}
+echo {a..z}
+===
+@match: echo a{b,c}d
+echo a{b,c}d
+===
+@match: time ls -la
+time ls -la
+===
+@match: ls
+ls
+echo b
+===
+@match: echo [ x ]
+echo [ x ]
+===
+@match: if [ -f f ]; then x; fi
+if [ -f f ]; then x; fi
+===
+@match: foo() if a; then b; fi
+foo() if a; then b; fi
+===
+@match: if a; then b; fi | grep x
+if a; then b; fi | grep x
+===
+@match: if ! grep -q x f; then echo missing; fi
+if ! grep -q x f; then echo missing; fi
+===
+@match: x[1]+=2
+x[1]+=2
+===
+@match: until grep -q done log; do sleep 1; done &
+until grep -q done log; do sleep 1; done &
+===
+@match: ( (ls) )
+( (ls) )
+===
+@match: {ls;}
+{ls;}
+===
+@match: fi
+fi
+===
+@match: done
+done
+===
+@match: }
+}
+===
+@match: echo a{1..3}b
+echo a{1..3}b
+===
+@match: x={1..5}
+x={1..5}
+===
+@match: while x; do done
+while x; do done
+===
+@match: echo {a..z..2}
+echo {a..z..2}
+===
+@match: echo {10..1}
+echo {10..1}
+===
+@match: echo {-5..5}
+echo {-5..5}
+===
+@match: select x in; do y; done
+select x in; do y; done
+===
+@match: function f() if x; then y; fi
+function f() if x; then y; fi
+===
+@match: f () { x; }
+f () { x; }
+===
+@match: if a && b; then c; fi
+if a && b; then c; fi
+===
+@match: foo (ls)
+foo (ls)
+===
+@match: for f in a b
+for f in a b
+do x
+done
+===
+@match: while a; b; do x; done
+while a; b; do x; done
+===
+@match: if a; then b; fi && c
+if a; then b; fi && c
+===
+@match: if if a; then b; fi; then c; fi
+if if a; then b; fi; then c; fi
+===
+@match: x=1 if a; then b; fi
+x=1 if a; then b; fi
+===
+@match: echo hi | if a; then b; fi
+echo hi | if a; then b; fi
+===
+@match: ! while x; do y; done
+! while x; do y; done
+===
+@match: ! [ -f x ]
+! [ -f x ]
+===
+@match: x+=(c d)
+x+=(c d)
+===
+@match: declare -A m
+declare -A m
+===
+@match: {1..3}
+{1..3}
+===
+@match: echo {1..3}
+echo {1..3}
+===
+@match: if echo then; then x; fi
+if echo then; then x; fi
+===
+@match: while echo do; do x; done
+while echo do; do x; done
+===
+@match: for f in a; do echo done; done
+for f in a; do echo done; done
+===
+@match: { echo }; }
+{ echo }; }
+===
+@match: [1]
+[1]
+===
+@match: echo [1] a[2]b
+echo [1] a[2]b
+===
+@match: echo a[-5]b
+echo a[-5]b
+===
+@match: if a; then b; elif c; then d
+if a; then b; elif c; then d
+else e
+fi
+===
+@match: ls -la
+ls -la
+===
+@match: FOO=bar env | grep FOO && echo ok || echo fail
+FOO=bar env | grep FOO && echo ok || echo fail
+===
+@match: ! grep -q x file
+! grep -q x file
+===
+@match: # a comment
+# a comment
+echo hi # trailing
+===
+@match: (ls)
+(ls)
+===
+@match: FOO=bar ls
+FOO=bar ls
+===
+@known-diff trailing-connector: ls &&
+ls &&
+---
+hasError: true
+program [0,5] "ls &&"
+  list [0,5] "ls &&"
+    command [0,2] "ls"
+      command_name [0,2] "ls"
+        word [0,2] "ls"
+    (&&) [3,5] "&&"
+===
+@match: a & b &
+a & b &
+===
+@match: echo 123 45.6
+echo 123 45.6
+===
+@match: VAR=value
+VAR=value
+===
+@match: ls; echo a
+ls; echo a
+echo b
+===
+@match: FOO=1 BAR=2 cmd arg
+FOO=1 BAR=2 cmd arg
+===
+@match: ! a | b
+! a | b
+===
+@match: echo ""
+echo ""
+===
+@match: VAR=
+VAR=
+===
+@match: a|b|c
+a|b|c
+===
+@match: ls ;
+ls ;
+===
+@match: echo hi;# c
+echo hi;# c
+ls
+===
+@match: A=1 B=2
+A=1 B=2
+===
+@match: a && b && c
+a && b && c
+===
+@match: echo x\
+echo x\
+y
+===
+@match: echo {a,b}
+echo {a,b}
+===
+@known-diff trailing-connector: ls |
+ls |
+---
+hasError: true
+program [0,4] "ls |"
+  pipeline [0,4] "ls |"
+    command [0,2] "ls"
+      command_name [0,2] "ls"
+        word [0,2] "ls"
+    (|) [3,4] "|"
+===
+@match: time ls
+time ls
+===
+@match: FOO="a b" cmd
+FOO="a b" cmd
+===
+@match: FOO=a"b"c
+FOO=a"b"c
+===
+@known-diff escaped-space-argument: echo 1 \ 2 \ 3
+echo 1 \ 2 \ 3
+---
+hasError: false
+program [0,14] "echo 1 \\ 2 \\ 3"
+  command [0,14] "echo 1 \\ 2 \\ 3"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    number [5,6] "1"
+    word [7,10] "\\ 2"
+    word [11,14] "\\ 3"
+===
+@known-diff expansion-equals-operator: for c in ${=1}; do
+for c in ${=1}; do
+	echo c
+done
+---
+hasError: false
+program [0,31] "for c in ${=1}; do\n\techo c\ndone"
+  for_statement [0,31] "for c in ${=1}; do\n\techo c\ndone"
+    (for) [0,3] "for"
+    variable_name [4,5] "c"
+    (in) [6,8] "in"
+    expansion [9,14] "${=1}"
+      (${) [9,11] "${"
+      (=) [11,12] "="
+      word [12,13] "1"
+      (}) [13,14] "}"
+    (;) [14,15] ";"
+    do_group [16,31] "do\n\techo c\ndone"
+      (do) [16,18] "do"
+      command [20,26] "echo c"
+        command_name [20,24] "echo"
+          word [20,24] "echo"
+        word [25,26] "c"
+      (done) [27,31] "done"
+===
+@match: foo::bar() {
+foo::bar() {
+  echo what
+}
+===
+@match: echo "a "
+echo "a "
+===
+@match: echo 你好 && ls
+echo 你好 && ls
+===
+@match: echo "🎉 done" | grep 🎉 && echo ✓
+echo "🎉 done" | grep 🎉 && echo ✓
+===
+@known-diff nonascii-identifier-assignment: 变量=值 echo $变量
+变量=值 echo $变量
+---
+hasError: false
+program [0,13] "变量=值 echo $变量"
+  command [0,13] "变量=值 echo $变量"
+    command_name [0,4] "变量=值"
+      word [0,4] "变量=值"
+    word [5,9] "echo"
+    concatenation [10,13] "$变量"
+      ($) [10,11] "$"
+      word [11,13] "变量"
+===
+@known-diff escaped-space-argument: echo a\<TAB>b
+echo a\	b
+---
+hasError: false
+program [0,9] "echo a\\\tb"
+  command [0,9] "echo a\\\tb"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    word [5,9] "a\\\tb"
+===
+@known-diff escaped-space-argument: echo \<TAB>x
+echo \	x
+---
+hasError: false
+program [0,8] "echo \\\tx"
+  command [0,8] "echo \\\tx"
+    command_name [0,4] "echo"
+      word [0,4] "echo"
+    word [5,8] "\\\tx"
+===
diff --git a/packages/tree-sitter-bash/test/fixtures/differential/test-command.txt b/packages/tree-sitter-bash/test/fixtures/differential/test-command.txt
new file mode 100644
index 0000000000..02b22149ef
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fixtures/differential/test-command.txt
@@ -0,0 +1,650 @@
+@match: [[ -f file.txt ]]
+[[ -f file.txt ]]
+===
+@match: [[ $x == "foo" ]]
+[[ $x == "foo" ]]
+===
+@match: [[ $x =~ ^ab+c$ ]]
+[[ $x =~ ^ab+c$ ]]
+===
+@match: [[ -n $s && -d /tmp ]]
+[[ -n $s && -d /tmp ]]
+===
+@match: [[ a < b || ! -e f ]]
+[[ a < b || ! -e f ]]
+===
+@match: [ -f file ]
+[ -f file ]
+===
+@match: [ "$a" = "b" ]
+[ "$a" = "b" ]
+===
+@match: ! [[ -f x ]]
+! [[ -f x ]]
+===
+@match: [ -f file
+[ -f file
+===
+@match: [ x ]
+[ x ]
+===
+@match: while [[ -n $s ]]; do s=; done
+while [[ -n $s ]]; do s=; done
+===
+@match: foo() [[ -f x ]]
+foo() [[ -f x ]]
+===
+@match: [[ -z $s ]]
+[[ -z $s ]]
+===
+@match: [[ $a -eq 3 ]]
+[[ $a -eq 3 ]]
+===
+@match: [[ $a -nt $b ]]
+[[ $a -nt $b ]]
+===
+@match: [[ "x" ]]
+[[ "x" ]]
+===
+@match: [[ ]]
+[[ ]]
+===
+@match: [[ a = b*c ]]
+[[ a = b*c ]]
+===
+@match: [[ a = bc ]]
+[[ a = bc ]]
+===
+@match: [[ $x == b*c ]]
+[[ $x == b*c ]]
+===
+@match: [[ $x == a"b" ]]
+[[ $x == a"b" ]]
+===
+@match: [[ $x =~ ^a.*b$ ]]
+[[ $x =~ ^a.*b$ ]]
+===
+@match: [[ a != b ]]
+[[ a != b ]]
+===
+@match: [[ (a == b) && c ]]
+[[ (a == b) && c ]]
+===
+@match: [[ -v var ]]
+[[ -v var ]]
+===
+@match: [[ -o opt ]]
+[[ -o opt ]]
+===
+@match: [[ $x -eq 1 && $y -ne 2 ]]
+[[ $x -eq 1 && $y -ne 2 ]]
+===
+@match: [[ $x == b ]]
+[[ $x == b ]]
+===
+@known-diff recovery-partial-nodes: [[ $x =~ ^a b$ ]]
+[[ $x =~ ^a b$ ]]
+---
+hasError: true
+program [0,17] "[[ $x =~ ^a b$ ]]"
+  test_command [0,17] "[[ $x =~ ^a b$ ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,11] "$x =~ ^a"
+      simple_expansion [3,5] "$x"
+        ($) [3,4] "$"
+        variable_name [4,5] "x"
+      (=~) [6,8] "=~"
+      regex [9,11] "^a"
+    ERROR [12,15] "b$ "
+    (]]) [15,17] "]]"
+===
+@match: [[ $x =~ a|b ]]
+[[ $x =~ a|b ]]
+===
+@match: [[ $x == b && $y == c ]]
+[[ $x == b && $y == c ]]
+===
+@match: [[ $x == 123 ]]
+[[ $x == 123 ]]
+===
+@match: [[ -t 1 ]]
+[[ -t 1 ]]
+===
+@match: [[ $x =~ $re ]]
+[[ $x =~ $re ]]
+===
+@match: [[ $x =~ "a b" ]]
+[[ $x =~ "a b" ]]
+===
+@match: [[ a=b ]]
+[[ a=b ]]
+===
+@match: [[ $a==b* ]]
+[[ $a==b* ]]
+===
+@match: [[ ( $a == x* ) ]]
+[[ ( $a == x* ) ]]
+===
+@match: [[ ( $a = x* ) ]]
+[[ ( $a = x* ) ]]
+===
+@match: [[ $x == b=c ]]
+[[ $x == b=c ]]
+===
+@match: [[ a == =b ]]
+[[ a == =b ]]
+===
+@match: [[ a = b=c ]]
+[[ a = b=c ]]
+===
+@match: [[ -f file
+[[ -f file
+===
+@known-diff recovery-partial-nodes: [[ $a
+[[ $a
+---
+hasError: true
+program [0,5] "[[ $a"
+  test_command [0,5] "[[ $a"
+    ([[) [0,2] "[["
+    simple_expansion [3,5] "$a"
+      ($) [3,4] "$"
+      variable_name [4,5] "a"
+    (]]) [5,5] ""
+===
+@match: [ 1 ]
+[ 1 ]
+===
+@match: [[ 1 ]]
+[[ 1 ]]
+===
+@match: [[ $f == foo*(.txt|.log) ]]
+[[ $f == foo*(.txt|.log) ]]
+===
+@match: [[ $f == *foo*(a) ]]
+[[ $f == *foo*(a) ]]
+===
+@match: [[ $f == [a-z]*(x) ]]
+[[ $f == [a-z]*(x) ]]
+===
+@known-diff test-extglob-rejected-group: [[ $f == x@(y|z)w ]]
+[[ $f == x@(y|z)w ]]
+---
+hasError: true
+program [0,20] "[[ $f == x@(y|z)w ]]"
+  test_command [0,20] "[[ $f == x@(y|z)w ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,11] "$f == x@"
+      simple_expansion [3,5] "$f"
+        ($) [3,4] "$"
+        variable_name [4,5] "f"
+      (==) [6,8] "=="
+      word [9,11] "x@"
+    ERROR [11,18] "(y|z)w "
+    (]]) [18,20] "]]"
+===
+@known-diff test-negative-decimal: [[ $n == -0.5 ]]
+[[ $n == -0.5 ]]
+---
+hasError: false
+program [0,16] "[[ $n == -0.5 ]]"
+  test_command [0,16] "[[ $n == -0.5 ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,13] "$n == -0.5"
+      simple_expansion [3,5] "$n"
+        ($) [3,4] "$"
+        variable_name [4,5] "n"
+      (==) [6,8] "=="
+      extglob_pattern [9,13] "-0.5"
+    (]]) [14,16] "]]"
+===
+@known-diff test-fused-operator-expansion: [[ -x$f && -d /tmp ]]
+[[ -x$f && -d /tmp ]]
+---
+hasError: false
+program [0,21] "[[ -x$f && -d /tmp ]]"
+  test_command [0,21] "[[ -x$f && -d /tmp ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,18] "-x$f && -d /tmp"
+      concatenation [3,7] "-x$f"
+        word [3,5] "-x"
+        simple_expansion [5,7] "$f"
+          ($) [5,6] "$"
+          variable_name [6,7] "f"
+      (&&) [8,10] "&&"
+      unary_expression [11,18] "-d /tmp"
+        test_operator [11,13] "-d"
+        word [14,18] "/tmp"
+    (]]) [19,21] "]]"
+===
+@known-diff test-escaped-pipe-pattern: [[ $a == foo\|bar ]]
+[[ $a == foo\|bar ]]
+---
+hasError: false
+program [0,20] "[[ $a == foo\\|bar ]]"
+  test_command [0,20] "[[ $a == foo\\|bar ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,17] "$a == foo\\|bar"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      extglob_pattern [9,17] "foo\\|bar"
+    (]]) [18,20] "]]"
+===
+@known-diff recovery-partial-nodes: [[ -f
+[[ -f
+---
+hasError: true
+program [0,5] "[[ -f"
+  test_command [0,5] "[[ -f"
+    ([[) [0,2] "[["
+    word [3,5] "-f"
+    (]]) [5,5] ""
+===
+@known-diff recovery-partial-nodes: [[ a
+[[ a
+---
+hasError: true
+program [0,4] "[[ a"
+  test_command [0,4] "[[ a"
+    ([[) [0,2] "[["
+    word [3,4] "a"
+    (]]) [4,4] ""
+===
+@known-diff recovery-partial-nodes: [[ a == b
+[[ a == b
+---
+hasError: true
+program [0,9] "[[ a == b"
+  test_command [0,9] "[[ a == b"
+    ([[) [0,2] "[["
+    binary_expression [3,9] "a == b"
+      word [3,4] "a"
+      (==) [5,7] "=="
+      word [8,9] "b"
+    (]]) [9,9] ""
+===
+@known-diff recovery-partial-nodes: [[ a &&
+[[ a &&
+---
+hasError: true
+program [0,7] "[[ a &&"
+  test_command [0,7] "[[ a &&"
+    ([[) [0,2] "[["
+    binary_expression [3,7] "a &&"
+      word [3,4] "a"
+      (&&) [5,7] "&&"
+    (]]) [7,7] ""
+===
+@match: [[ -f file 
+[[ -f file 
+===
+@match: [[ a = =b ]]
+[[ a = =b ]]
+===
+@match: [[ $a =~^x ]]
+[[ $a =~^x ]]
+===
+@match: [[ !x = y ]]
+[[ !x = y ]]
+===
+@match: [[ a =b ]]
+[[ a =b ]]
+===
+@match: [[ a!=b ]]
+[[ a!=b ]]
+===
+@match: [[ a ==b ]]
+[[ a ==b ]]
+===
+@match: [[ $a == foo ]]
+[[ $a == foo ]]
+===
+@match: [[ $a == x.y ]]
+[[ $a == x.y ]]
+===
+@match: [[ ( $a == x ) ]]
+[[ ( $a == x ) ]]
+===
+@match: [[ $a == x1 ]]
+[[ $a == x1 ]]
+===
+@match: [[ $a == foo- ]]
+[[ $a == foo- ]]
+===
+@match: [[ $a == x_ ]]
+[[ $a == x_ ]]
+===
+@match: [[ a == b*c=d ]]
+[[ a == b*c=d ]]
+===
+@match: [[ a == bc=d ]]
+[[ a == bc=d ]]
+===
+@match: [[ a == b?=e ]]
+[[ a == b?=e ]]
+===
+@match: [[ a == .x ]]
+[[ a == .x ]]
+===
+@match: [[ $a == a-b ]]
+[[ $a == a-b ]]
+===
+@match: [[ ((a)) == x ]]
+[[ ((a)) == x ]]
+===
+@match: [[ $a == -foo ]]
+[[ $a == -foo ]]
+===
+@match: [[ $a == -f ]]
+[[ $a == -f ]]
+===
+@match: [[ -foo == x ]]
+[[ -foo == x ]]
+===
+@match: [[ $a == -1 ]]
+[[ $a == -1 ]]
+===
+@match: [[ $a -eq -1 ]]
+[[ $a -eq -1 ]]
+===
+@match: [[ $a == +(!a) ]]
+[[ $a == +(!a) ]]
+===
+@match: [[ $a == ?(a|b) ]]
+[[ $a == ?(a|b) ]]
+===
+@match: [[ $a == !(x) ]]
+[[ $a == !(x) ]]
+===
+@match: [[ $a == b\*c ]]
+[[ $a == b\*c ]]
+===
+@match: [[ $ver == [0-9]* ]]
+[[ $ver == [0-9]* ]]
+===
+@match: [[ $a == [abc] ]]
+[[ $a == [abc] ]]
+===
+@match: [[ ( $a == [xyz] ) ]]
+[[ ( $a == [xyz] ) ]]
+===
+@match: [[ $a = [abc] ]]
+[[ $a = [abc] ]]
+===
+@match: [[ $a == [!a]* ]]
+[[ $a == [!a]* ]]
+===
+@match: [[ $a == foo\ bar ]]
+[[ $a == foo\ bar ]]
+===
+@known-diff test-paren-group-logical: [[ ((a) == x) && y ]]
+[[ ((a) == x) && y ]]
+---
+hasError: false
+program [0,21] "[[ ((a) == x) && y ]]"
+  test_command [0,21] "[[ ((a) == x) && y ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,18] "((a) == x) && y"
+      parenthesized_expression [3,13] "((a) == x)"
+        (() [3,4] "("
+        binary_expression [4,12] "(a) == x"
+          parenthesized_expression [4,7] "(a)"
+            (() [4,5] "("
+            word [5,6] "a"
+            ()) [6,7] ")"
+          (==) [8,10] "=="
+          word [11,12] "x"
+        ()) [12,13] ")"
+      (&&) [14,16] "&&"
+      word [17,18] "y"
+    (]]) [19,21] "]]"
+===
+@match: [ ! command -v go &>/dev/null ] && return
+[ ! command -v go &>/dev/null ] && return
+===
+@match: [[ ${f} != */@(default).vim ]]
+[[ ${f} != */@(default).vim ]]
+===
+@match: [[ $(LC_ALL=C $(tc-getCC) ${LDFLAGS} -Wl,--version 2>/dev...
+[[ $(LC_ALL=C $(tc-getCC) ${LDFLAGS} -Wl,--version 2>/dev/null) != @(LLD|GNU\ ld)* ]]
+===
+@match: [[ ${test} == @($(IFS='|'; echo "${skip[*]}")) ]]
+[[ ${test} == @($(IFS='|'; echo "${skip[*]}")) ]]
+===
+@match: [[ ${SRC_URI} == */${a}* ]]
+[[ ${SRC_URI} == */${a}* ]]
+===
+@match: [[ a == *_@(LIB|SYMLINK) ]]
+[[ a == *_@(LIB|SYMLINK) ]]
+===
+@match: [[ "35d8b" =~ ^[0-9a-fA-F] ]]
+[[ "35d8b" =~ ^[0-9a-fA-F] ]]
+===
+@match: [[ $CMD =~ (^|;)update_terminal_cwd($|;) ]]
+[[ $CMD =~ (^|;)update_terminal_cwd($|;) ]]
+===
+@match: [[ ! " ${completions[*]} " =~ " $alias_cmd " ]]
+[[ ! " ${completions[*]} " =~ " $alias_cmd " ]]
+===
+@match: ! [[ "$a" =~ ^a|b\ *c|d$ ]]
+! [[ "$a" =~ ^a|b\ *c|d$ ]]
+===
+@match: [[ "$1" =~ ^${var}${var}*=..* ]]
+[[ "$1" =~ ^${var}${var}*=..* ]]
+===
+@match: [[ "$1" =~ ^\-${var}+ ]]
+[[ "$1" =~ ^\-${var}+ ]]
+===
+@match: [[ ${var1} == *${var2}* ]]
+[[ ${var1} == *${var2}* ]]
+===
+@match: [[ "$server" =~ [0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]...
+[[ "$server" =~ [0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3} ]]
+===
+@match: [[ -f ${x} && $(od -t x1 -N 4 "${x}") == *"7f 45 4c 46"* ]]
+[[ -f ${x} && $(od -t x1 -N 4 "${x}") == *"7f 45 4c 46"* ]]
+===
+@match: [[ " ${REPLACING_VERSIONS} " != *\ ${PVR}\ * ]]
+[[ " ${REPLACING_VERSIONS} " != *\ ${PVR}\ * ]]
+===
+@match: [[ ${file} == @(*${GENTOO_PATCH_NAME}.tar.xz|*.asc|*.sig) ]]
+[[ ${file} == @(*${GENTOO_PATCH_NAME}.tar.xz|*.asc|*.sig) ]]
+===
+@match: [[ $RUBY_TARGETS != *$( eselect ruby show | awk 'NR==2' |...
+[[ $RUBY_TARGETS != *$( eselect ruby show | awk 'NR==2' | tr  -d ' '  )* ]]
+===
+@match: [[ " ${m[0]##*/}" =~ ^(\ ${skip_files[*]/%/.*|\\} ) ]]
+[[ " ${m[0]##*/}" =~ ^(\ ${skip_files[*]/%/.*|\\} ) ]]
+===
+@match: [[ ' boop xyz' =~ ' boop '(.*)$ ]]
+[[ ' boop xyz' =~ ' boop '(.*)$ ]]
+===
+@match: [[ $b = foo* ]]
+[[ $b = foo* ]]
+===
+@match: [[ a == \"+(?)\" ]]
+[[ a == \"+(?)\" ]]
+===
+@match: [[ ${PV} != $(sed -n -e 's/^Version: //p' "${ED}/usr/$(ge...
+[[ ${PV} != $(sed -n -e 's/^Version: //p' "${ED}/usr/$(get_libdir)/pkgconfig/tss2-tcti-tabrmd.pc" || die) ]]
+===
+@match: [[ "${MY_LOCALES}" != *en_US* || a != 2 ]]
+[[ "${MY_LOCALES}" != *en_US* || a != 2 ]]
+===
+@match: [[ -f "${EROOT}/usr/share/php/.packagexml/${MY_P}.xml" && \
+[[ -f "${EROOT}/usr/share/php/.packagexml/${MY_P}.xml" && \
+	-x "${EROOT}/usr/bin/peardev" ]]
+===
+@match: [[ ${1} =~ \.(lisp|lsp|cl)$ ]]
+[[ ${1} =~ \.(lisp|lsp|cl)$ ]]
+===
+@match: [[ a == - ]]
+[[ a == - ]]
+===
+@match: run_test_command() [[ -e foo ]]
+run_test_command() [[ -e foo ]]
+===
+@known-diff test-extglob-rejected-group: [[ $a == *.@(txt|log) ]]
+[[ $a == *.@(txt|log) ]]
+---
+hasError: true
+program [0,24] "[[ $a == *.@(txt|log) ]]"
+  test_command [0,24] "[[ $a == *.@(txt|log) ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,12] "$a == *.@"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      word [9,12] "*.@"
+    ERROR [12,22] "(txt|log) "
+    (]]) [22,24] "]]"
+===
+@known-diff test-extglob-rejected-group: [[ $a == x@(y) ]]
+[[ $a == x@(y) ]]
+---
+hasError: true
+program [0,17] "[[ $a == x@(y) ]]"
+  test_command [0,17] "[[ $a == x@(y) ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,11] "$a == x@"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      word [9,11] "x@"
+    ERROR [11,15] "(y) "
+    (]]) [15,17] "]]"
+===
+@match: [[ $a == ?(a) ]]
+[[ $a == ?(a) ]]
+===
+@known-diff test-pattern-two-constructs: [[ $a == *${x}*${y} ]]
+[[ $a == *${x}*${y} ]]
+---
+hasError: false
+program [0,22] "[[ $a == *${x}*${y} ]]"
+  test_command [0,22] "[[ $a == *${x}*${y} ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,19] "$a == *${x}*${y}"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      extglob_pattern [9,10] "*"
+      expansion [10,14] "${x}"
+        (${) [10,12] "${"
+        variable_name [12,13] "x"
+        (}) [13,14] "}"
+      extglob_pattern [14,19] "*${y}"
+    (]]) [20,22] "]]"
+===
+@match: [[ $a == ${x}* ]]
+[[ $a == ${x}* ]]
+===
+@known-diff test-pattern-two-constructs: [[ $a == *"s"*"t" ]]
+[[ $a == *"s"*"t" ]]
+---
+hasError: false
+program [0,20] "[[ $a == *\"s\"*\"t\" ]]"
+  test_command [0,20] "[[ $a == *\"s\"*\"t\" ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,17] "$a == *\"s\"*\"t\""
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      extglob_pattern [9,10] "*"
+      string [10,13] "\"s\""
+        (") [10,11] "\""
+        string_content [11,12] "s"
+        (") [12,13] "\""
+      extglob_pattern [13,17] "*\"t\""
+    (]]) [18,20] "]]"
+===
+@match: [ -f x ]
+[ -f x ]
+===
+@match: [ ! -f x ]
+[ ! -f x ]
+===
+@match: [ a = b ] 2>/dev/null
+[ a = b ] 2>/dev/null
+===
+@match: [[ $a == @(y) ]]
+[[ $a == @(y) ]]
+===
+@known-diff test-extglob-rejected-group: [[ $a == *@(y) ]]
+[[ $a == *@(y) ]]
+---
+hasError: true
+program [0,17] "[[ $a == *@(y) ]]"
+  test_command [0,17] "[[ $a == *@(y) ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,11] "$a == *@"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      word [9,11] "*@"
+    ERROR [11,15] "(y) "
+    (]]) [15,17] "]]"
+===
+@match: [[ $a == */@(y) ]]
+[[ $a == */@(y) ]]
+===
+@known-diff test-extglob-rejected-group: [[ $a == .@(y) ]]
+[[ $a == .@(y) ]]
+---
+hasError: true
+program [0,17] "[[ $a == .@(y) ]]"
+  test_command [0,17] "[[ $a == .@(y) ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,11] "$a == .@"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      word [9,11] ".@"
+    ERROR [11,15] "(y) "
+    (]]) [15,17] "]]"
+===
+@known-diff test-extglob-rejected-group: [[ $a == _@(y) ]]
+[[ $a == _@(y) ]]
+---
+hasError: true
+program [0,17] "[[ $a == _@(y) ]]"
+  test_command [0,17] "[[ $a == _@(y) ]]"
+    ([[) [0,2] "[["
+    binary_expression [3,11] "$a == _@"
+      simple_expansion [3,5] "$a"
+        ($) [3,4] "$"
+        variable_name [4,5] "a"
+      (==) [6,8] "=="
+      word [9,11] "_@"
+    ERROR [11,15] "(y) "
+    (]]) [15,17] "]]"
+===
+@match: [[ $a == ?*(y) ]]
+[[ $a == ?*(y) ]]
+===
+@match: [[ $a == a*(y) ]]
+[[ $a == a*(y) ]]
+===
+@match: [[ $a == *?(y) ]]
+[[ $a == *?(y) ]]
+===
+@match: [[ $a =~ 'a'b ]]
+[[ $a =~ 'a'b ]]
+===
+@match: [[ $a =~ a'b'c ]]
+[[ $a =~ a'b'c ]]
+===
+@match: [ ! command -v go ]
+[ ! command -v go ]
+===
+@match: [ command -v go &>/dev/null ]
+[ command -v go &>/dev/null ]
+===
diff --git a/packages/tree-sitter-bash/test/fuzz.test.ts b/packages/tree-sitter-bash/test/fuzz.test.ts
new file mode 100644
index 0000000000..cecbbbcc12
--- /dev/null
+++ b/packages/tree-sitter-bash/test/fuzz.test.ts
@@ -0,0 +1,165 @@
+// test/fuzz.test.ts
+//
+// Deterministic fuzzing (fixed-seed PRNG — no per-run randomness; scale the
+// volume with TS_BASH_FUZZ_SCALE=N for deeper local runs):
+//   a) token soup — random sequences of bash operators/keywords/fragments;
+//   b) byte mutations of the curated fixtures (replace/delete/insert,
+//      NUL included);
+//   c) nesting bombs — programmatic deep $(…) / (…) / ${…} / if nesting.
+//
+// Contract under test: parse never throws; within budget it returns either
+// ok:true or { ok:false, reason:'aborted' }; an ok:true tree is always
+// structurally sound (assertTreeIntegrity); nesting beyond the documented
+// depth caps degrades locally (hasError) instead of throwing.
+
+import { readFileSync, readdirSync } from 'node:fs';
+import path from 'node:path';
+
+import { describe, expect, it } from 'vitest';
+
+import { parse } from '#/parse';
+import type { ParseOptions, ParseResult } from '#/parse';
+import { assertTreeIntegrity, parseFixtureFile } from './helpers/differential';
+
+const PKG_ROOT = path.resolve(import.meta.dirname, '..');
+const SCALE = Math.max(1, Number(process.env['TS_BASH_FUZZ_SCALE'] ?? 1) || 1);
+
+/** Park–Miller minimal standard PRNG: small, portable, deterministic. */
+function prng(seed: number): () => number {
+  let state = seed % 2147483647;
+  if (state <= 0) state += 2147483646;
+  return () => {
+    state = (state * 16807) % 2147483647;
+    return (state - 1) / 2147483646;
+  };
+}
+
+function pick(rand: () => number, pool: readonly string[]): string {
+  return pool[Math.floor(rand() * pool.length)]!;
+}
+
+function checkContract(source: string, options?: ParseOptions): ParseResult {
+  let result: ParseResult | undefined;
+  expect(() => {
+    result = parse(source, options);
+  }).not.toThrow();
+  expect(result).toBeDefined();
+  if (result!.ok) {
+    assertTreeIntegrity(result!.rootNode, source);
+  } else {
+    expect(result!).toEqual({ ok: false, reason: 'aborted' });
+  }
+  return result!;
+}
+
+const TOKEN_POOL = [
+  'echo', 'ls', 'foo', 'bar', 'x', 'A=1', 'if', 'then', 'fi', 'while', 'do', 'done', 'for', 'in', 'case', 'esac',
+  '&&', '||', '|', '|&', ';', '&', ';;', ';&', '(', ')', '{', '}', '>', '>>', '<', '>&1', '2>', '&>', '<<EOF',
+  '<<<', '$x', '${v}', '${v:-d}', '$(cmd)', '`cmd`', '$((1+2))', '[[', ']]', '[', ']', '==', '=~', '-f', '-z',
+  '"str"', "'raw'", '$"t"', "$'a'", '*', '?', '[a-z]', '@(a|b)', '\\\\', '\\n', '\n', '#c', 'function', 'return',
+  '${v#p}', '${v/p/r}', '${v[@]}', '!<', '>(p)', '<(p)', '$#', '$?', '0x1F', '..', '<<-', '>&-', 'abc_def',
+] as const;
+
+function fixtureSources(): string[] {
+  const dir = path.join(PKG_ROOT, 'test/fixtures/differential');
+  const out: string[] = [];
+  for (const file of readdirSync(dir).filter((f) => f.endsWith('.txt')).toSorted()) {
+    for (const sample of parseFixtureFile(file, readFileSync(path.join(dir, file), 'utf8'))) {
+      out.push(sample.source);
+    }
+  }
+  return out;
+}
+
+describe('fuzz: token soup', () => {
+  it('never throws and always yields a sound tree or a clean abort', () => {
+    const rand = prng(0x5eed0001);
+    const count = 250 * SCALE;
+    let parsed = 0;
+    for (let n = 0; n < count; n++) {
+      const length = 3 + Math.floor(rand() * 22);
+      const parts: string[] = [];
+      for (let k = 0; k < length; k++) parts.push(pick(rand, TOKEN_POOL));
+      if (checkContract(parts.join(' ')).ok) parsed++;
+    }
+    expect(parsed).toBeGreaterThan(0);
+    console.log(`token soup: ${count} inputs, ${parsed} parsed / ${count - parsed} aborted`);
+  });
+});
+
+describe('fuzz: byte mutations of fixtures', () => {
+  it('never throws and always yields a sound tree or a clean abort', () => {
+    const rand = prng(0x5eed0002);
+    const bases = fixtureSources();
+    const count = 300 * SCALE;
+    let parsed = 0;
+    for (let n = 0; n < count; n++) {
+      const base = pick(rand, bases);
+      if (base.length === 0) continue;
+      const pos = Math.floor(rand() * base.length);
+      const mode = rand();
+      let mutated: string;
+      if (mode < 0.4) {
+        // replace one code unit (may be NUL)
+        const code = Math.floor(rand() * 256);
+        mutated = base.slice(0, pos) + String.fromCodePoint(code) + base.slice(pos + 1);
+      } else if (mode < 0.7) {
+        mutated = base.slice(0, pos) + base.slice(pos + 1);
+      } else {
+        const code = Math.floor(rand() * 256);
+        mutated = base.slice(0, pos) + String.fromCodePoint(code) + base.slice(pos);
+      }
+      if (checkContract(mutated).ok) parsed++;
+    }
+    expect(parsed).toBeGreaterThan(0);
+    console.log(`byte mutations: ${count} inputs, ${parsed} parsed / ${count - parsed} aborted`);
+  });
+});
+
+describe('fuzz: nesting bombs degrade locally per the documented depth caps', () => {
+  const substitution = (depth: number): string => `echo ${'$('.repeat(depth)}x${')'.repeat(depth)}`;
+  const subshell = (depth: number): string => `${'('.repeat(depth)}x${')'.repeat(depth)}`;
+  const expansion = (depth: number): string => `echo ${'${a:-'.repeat(depth)}z${'}'.repeat(depth)}`;
+  const ifs = (depth: number): string => `${'if x; then '.repeat(depth)}y${'; fi'.repeat(depth)}`;
+
+  it('within the caps the trees are clean', () => {
+    // literalDepth ticks twice per ${…} level (parseLiteral + parseExpansion),
+    // so 200 levels stay under MAX_PARSE_DEPTH = 500.
+    for (const source of [substitution(100), subshell(400), expansion(200), ifs(400)]) {
+      const result = checkContract(source, { timeoutMs: 60_000 });
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(false);
+    }
+  });
+
+  it('beyond the caps the parse still succeeds and flags hasError (MAX_SUBSTITUTION_DEPTH = 150)', () => {
+    const result = checkContract(substitution(200));
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('beyond the caps the parse still succeeds and flags hasError (MAX_PARSE_DEPTH = 500)', () => {
+    for (const source of [subshell(600), expansion(600), ifs(600)]) {
+      const result = checkContract(source, { timeoutMs: 60_000 });
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(true);
+    }
+  });
+
+  it('extreme nesting never overflows the stack', () => {
+    for (const source of [substitution(5000), subshell(5000), expansion(5000), ifs(5000)]) {
+      const result = checkContract(source, { timeoutMs: 60_000 });
+      // Either a locally degraded tree (hasError) or a clean budget abort.
+      if (result.ok) expect(result.hasError).toBe(true);
+      else expect(result.reason).toBe('aborted');
+    }
+  });
+
+  it('the node budget aborts huge flat programs, and a raised budget parses them', () => {
+    const source = 'echo a; '.repeat(20_000);
+    expect(parse(source)).toEqual({ ok: false, reason: 'aborted' });
+    const result = parse(source, { timeoutMs: 60_000, maxNodes: 10_000_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) assertTreeIntegrity(result.rootNode, source);
+  });
+});
diff --git a/packages/tree-sitter-bash/test/helpers/differential.ts b/packages/tree-sitter-bash/test/helpers/differential.ts
new file mode 100644
index 0000000000..e66e5ebe82
--- /dev/null
+++ b/packages/tree-sitter-bash/test/helpers/differential.ts
@@ -0,0 +1,271 @@
+// test/helpers/differential.ts
+//
+// Differential test harness: parses the same source with both this package's
+// parser and the real tree-sitter-bash (web-tree-sitter + the official wasm
+// build shipped in the tree-sitter-bash npm package), normalizes both trees
+// into a comparable dump, and produces structured diffs.
+//
+// Normalization: a pre-order dump of EVERY node (named and anonymous — the
+// byte-identical trees M1/M2 verified include the anonymous token layer),
+// one line per node: `<indent>type [start,end] "text preview"`, anonymous
+// types wrapped in parens, plus a leading `hasError:` line. Both sides use
+// UTF-16 code unit offsets directly: web-tree-sitter (wasm) reports UTF-16
+// offsets for string input, same as this parser (verified: for
+// "echo 你好 && ls" the reference reports the root as [0,13] and `&&` as
+// [8,10] — UTF-16 code units, not UTF-8 bytes).
+
+import path from 'node:path';
+
+import { Language, Parser as RefParser } from 'web-tree-sitter';
+import type { Node as RefNode } from 'web-tree-sitter';
+
+import type { SyntaxNode } from '#/node';
+import { parse } from '#/parse';
+
+const PACKAGE_ROOT = path.resolve(import.meta.dirname, '../..');
+const WASM_PATH = path.join(PACKAGE_ROOT, 'node_modules/tree-sitter-bash/tree-sitter-bash.wasm');
+
+let refParserPromise: Promise<RefParser> | null = null;
+
+/**
+ * Lazily load web-tree-sitter + the bash wasm exactly once per test process.
+ * Load failures throw a descriptive error (never silently skip): the
+ * differential suite is meaningless without the reference.
+ */
+export function loadReferenceParser(): Promise<RefParser> {
+  refParserPromise ??= (async () => {
+    try {
+      await RefParser.init();
+      const language = await Language.load(WASM_PATH);
+      const parser = new RefParser();
+      parser.setLanguage(language);
+      return parser;
+    } catch (error) {
+      throw new Error(
+        `failed to load the tree-sitter-bash wasm reference (expected at ${WASM_PATH}). ` +
+          'Run `pnpm install` and make sure the tree-sitter-bash devDependency is present.',
+        { cause: error },
+      );
+    }
+  })();
+  return refParserPromise;
+}
+
+interface DumpLine {
+  depth: number;
+  label: string;
+  start: number;
+  end: number;
+  text: string;
+}
+
+function render(lines: DumpLine[], hasError: boolean): string {
+  const out = [`hasError: ${hasError}`];
+  for (const line of lines) {
+    const preview =
+      line.text.length <= 40 ? JSON.stringify(line.text) : JSON.stringify(`${line.text.slice(0, 37)}...`);
+    out.push(`${'  '.repeat(line.depth)}${line.label} [${line.start},${line.end}] ${preview}`);
+  }
+  return out.join('\n');
+}
+
+function label(type: string, isNamed: boolean): string {
+  return isNamed ? type : `(${type})`;
+}
+
+/** Normalized dump of the reference tree (UTF-16 offsets as reported). */
+export async function referenceDump(source: string): Promise<string> {
+  const parser = await loadReferenceParser();
+  const tree = parser.parse(source);
+  if (tree === null) throw new Error(`reference parser returned null for ${JSON.stringify(source)}`);
+  const lines: DumpLine[] = [];
+  const stack: Array<{ node: RefNode; depth: number }> = [{ node: tree.rootNode, depth: 0 }];
+  while (stack.length > 0) {
+    const { node, depth } = stack.pop()!;
+    lines.push({
+      depth,
+      label: label(node.type, node.isNamed),
+      start: node.startIndex,
+      end: node.endIndex,
+      text: source.slice(node.startIndex, node.endIndex),
+    });
+    for (let i = node.childCount - 1; i >= 0; i--) stack.push({ node: node.child(i)!, depth: depth + 1 });
+  }
+  return render(lines, tree.rootNode.hasError);
+}
+
+/**
+ * Normalized dump of our tree. Parses with a generous budget so differential
+ * results are never polluted by the default 50 ms / 50 000-node budget;
+ * fixtures are expected to parse, so an abort here is an error.
+ */
+export function ourDump(source: string): string {
+  const result = parse(source, { timeoutMs: 60_000, maxNodes: 10_000_000 });
+  if (!result.ok) throw new Error(`our parser aborted on a differential fixture: ${JSON.stringify(source)}`);
+  const lines: DumpLine[] = [];
+  const stack: Array<{ node: SyntaxNode; depth: number }> = [{ node: result.rootNode, depth: 0 }];
+  while (stack.length > 0) {
+    const { node, depth } = stack.pop()!;
+    lines.push({ depth, label: label(node.type, node.isNamed), start: node.startIndex, end: node.endIndex, text: node.text });
+    for (let i = node.children.length - 1; i >= 0; i--) stack.push({ node: node.children[i]!, depth: depth + 1 });
+  }
+  return render(lines, result.hasError);
+}
+
+/** First-difference report between two dumps; empty string when identical. */
+export function diffDumps(ours: string, reference: string): string {
+  if (ours === reference) return '';
+  const a = ours.split('\n');
+  const b = reference.split('\n');
+  let i = 0;
+  while (i < a.length && i < b.length && a[i] === b[i]) i++;
+  const slice = (lines: string[], from: number): string =>
+    lines
+      .slice(from, from + 8)
+      .map((line, j) => `${from + j === i ? '>' : ' '} ${line}`)
+      .join('\n');
+  const at = Math.max(0, i - 2);
+  return (
+    `trees differ at dump line ${i + 1}:\n` +
+    `  ours:\n${slice(a, at)}\n` +
+    `  reference:\n${slice(b, at)}`
+  );
+}
+
+export interface Comparison {
+  equal: boolean;
+  ours: string;
+  reference: string;
+  diff: string;
+}
+
+/** Parse `source` with both parsers and compare the normalized dumps. */
+export async function compareSource(source: string): Promise<Comparison> {
+  const [ours, reference] = [ourDump(source), await referenceDump(source)];
+  return { equal: ours === reference, ours, reference, diff: diffDumps(ours, reference) };
+}
+
+/**
+ * Tree integrity self-check (used by fuzz): every node range must lie inside
+ * the source, `text` must equal the corresponding slice, and children must be
+ * contained in their parent, in source order, without overlaps. Throws with a
+ * descriptive message on the first violation.
+ */
+export function assertTreeIntegrity(root: SyntaxNode, source: string): void {
+  const stack: SyntaxNode[] = [root];
+  while (stack.length > 0) {
+    const node = stack.pop()!;
+    if (node.startIndex < 0 || node.endIndex < node.startIndex || node.endIndex > source.length) {
+      throw new Error(`node ${node.type} range [${node.startIndex}, ${node.endIndex}) escapes source length ${source.length}`);
+    }
+    if (node.text !== source.slice(node.startIndex, node.endIndex)) {
+      throw new Error(`node ${node.type} text does not match source.slice(${node.startIndex}, ${node.endIndex})`);
+    }
+    let previousEnd = node.startIndex;
+    for (const child of node.children) {
+      if (child.startIndex < node.startIndex || child.endIndex > node.endIndex) {
+        throw new Error(`child ${child.type} escapes parent ${node.type}`);
+      }
+      if (child.startIndex < previousEnd) {
+        throw new Error(`child ${child.type} overlaps its previous sibling in ${node.type}`);
+      }
+      previousEnd = child.endIndex;
+      stack.push(child);
+    }
+  }
+}
+
+// ------------------------------------------------------------- fixture I/O
+
+/** A curated differential fixture sample (see test/fixtures/differential/). */
+export interface FixtureSample {
+  /** Directive: 'match' (must equal the reference) or 'known-diff'. */
+  kind: 'match' | 'known-diff';
+  /** Known-difference registry id (known-diff samples only). */
+  id?: string;
+  /** Free-form description from the directive line. */
+  description: string;
+  source: string;
+  /** Stored expected dump of OUR parser (known-diff samples only). */
+  expectedOurs?: string;
+  /** 1-based line of the directive, for error messages. */
+  line: number;
+}
+
+const DIRECTIVE_RE = /^@(match|known-diff)(?:\s+(\S+))?:\s*(.*)$/;
+const SEPARATOR_RE = /^===\s*$/m;
+const DUMP_SPLIT_RE = /^---\s*$/m;
+
+/** One official tree-sitter-bash corpus test case (input only — the
+ *  expected S-expression is ignored; the live wasm reference is the
+ *  comparison target). */
+export interface CorpusCase {
+  name: string;
+  input: string;
+}
+
+/** Parse an official corpus file (`==== name ====` / input / `----` /
+ *  expected). The separator is any run of 3+ dashes. */
+export function parseCorpusFile(content: string): CorpusCase[] {
+  const out: CorpusCase[] = [];
+  const lines = content.split('\n');
+  let i = 0;
+  while (i < lines.length) {
+    if (!/^=+\s*$/.test(lines[i]!)) {
+      i++;
+      continue;
+    }
+    i++;
+    const nameLines: string[] = [];
+    while (i < lines.length && !/^=+\s*$/.test(lines[i]!)) nameLines.push(lines[i++]!);
+    i++;
+    const inputLines: string[] = [];
+    while (i < lines.length && !/^-{3,}\s*$/.test(lines[i]!)) inputLines.push(lines[i++]!);
+    i++;
+    while (i < lines.length && !/^=+\s*$/.test(lines[i]!)) i++;
+    out.push({ name: nameLines.join(' ').trim(), input: inputLines.join('\n').replace(/^\n+/, '').replace(/\n+$/, '') });
+  }
+  return out;
+}
+
+
+/**
+ * Parse a fixture file. Format: blocks separated by lines of `===`; each
+ * block starts with `@match: <desc>` or `@known-diff <registry-id>: <desc>`,
+ * followed by the sample source. known-diff blocks additionally carry the
+ * expected dump of our parser after a `---` line (the disclosed deviation
+ * shape — if our output drifts from it, or starts matching the reference,
+ * the test fails).
+ */
+export function parseFixtureFile(filePath: string, content: string): FixtureSample[] {
+  const samples: FixtureSample[] = [];
+  const blocks = content.split(SEPARATOR_RE);
+  let line = 1;
+  for (const rawBlock of blocks) {
+    const block = rawBlock.replace(/^\n/, '').replace(/\n$/, '');
+    const blockLine = line;
+    line += rawBlock.split('\n').length;
+    if (block.trim().length === 0) continue;
+    const nl = block.indexOf('\n');
+    const directiveLine = nl === -1 ? block : block.slice(0, nl);
+    const body = nl === -1 ? '' : block.slice(nl + 1);
+    const directive = DIRECTIVE_RE.exec(directiveLine.trim());
+    if (directive === null) {
+      throw new Error(`${filePath}:${blockLine}: expected a @match: / @known-diff <id>: directive, got ${JSON.stringify(directiveLine)}`);
+    }
+    const [, kind, id, description = ''] = directive;
+    if (kind === 'match') {
+      samples.push({ kind, description, source: body, line: blockLine });
+    } else {
+      if (id === undefined) throw new Error(`${filePath}:${blockLine}: @known-diff requires a registry id`);
+      const parts = body.split(DUMP_SPLIT_RE);
+      if (parts.length !== 2) {
+        throw new Error(`${filePath}:${blockLine}: @known-diff block must contain a --- line followed by the expected dump of our parser`);
+      }
+      const source = parts[0]!.replace(/\n$/, '');
+      const expectedOurs = parts[1]!.replace(/^\n/, '').replace(/\n$/, '');
+      samples.push({ kind: 'known-diff', id, description, source, expectedOurs, line: blockLine });
+    }
+  }
+  return samples;
+}
diff --git a/packages/tree-sitter-bash/test/helpers/known-differences.ts b/packages/tree-sitter-bash/test/helpers/known-differences.ts
new file mode 100644
index 0000000000..b3b526e6f6
--- /dev/null
+++ b/packages/tree-sitter-bash/test/helpers/known-differences.ts
@@ -0,0 +1,179 @@
+// test/helpers/known-differences.ts
+//
+// The single source of truth for the known, deliberately documented ways
+// this parser's tree deviates from tree-sitter-bash 0.25.0. Every entry is:
+//   - referenced by @known-diff samples in test/fixtures/differential/ and
+//     test/fixtures/corpus/known-diffs.txt (the stored dump in each sample
+//     pins the exact deviation shape — if our output drifts, or starts
+//     matching the reference, the differential tests fail);
+//   - anchored in README.md's "Known differences" section: `readmeAnchor`
+//     must appear there verbatim, so the list and the README cannot
+//     silently drift apart (enforced by test/differential.test.ts).
+
+export interface KnownDifference {
+  /** Identifier referenced from fixture @known-diff directives. */
+  readonly id: string;
+  /** Verbatim substring of README.md's Known differences section. */
+  readonly readmeAnchor: string;
+  /** Short human-readable description. */
+  readonly summary: string;
+}
+
+export const KNOWN_DIFFERENCES: readonly KnownDifference[] = [
+  {
+    id: 'redirect-readwrite',
+    readmeAnchor: '`<>` (read-write redirect) is parsed as a normal `file_redirect`',
+    summary: '<> parses as file_redirect; the reference fails to parse it.',
+  },
+  {
+    id: 'heredoc-multiple-on-line',
+    readmeAnchor: 'Several heredocs on one line',
+    summary: 'cat <<A <<B: the second << degrades to an ERROR node.',
+  },
+  {
+    id: 'heredoc-tail-statements',
+    readmeAnchor: 'Statements after a heredoc on the same line',
+    summary: 'cat <<EOF; echo x — the tail is absorbed into heredoc_redirect (the reference errors).',
+  },
+  {
+    id: 'heredoc-content-chunks',
+    readmeAnchor: 'every plain text chunk — including the leading one — becomes a `heredoc_content` node',
+    summary: 'Unquoted heredoc_body: leading content chunk and backtick substitutions are nodes here.',
+  },
+  {
+    id: 'heredoc-at-statement-start',
+    readmeAnchor: 'A heredoc redirect at statement start',
+    summary: '<<EOF cat parses normally; the reference errors.',
+  },
+  {
+    id: 'heredoc-body-dollar-quote',
+    readmeAnchor: "A `$'` sequence inside an unquoted heredoc body",
+    summary: "$' inside a heredoc body is plain text here; the reference errors when it starts a body line.",
+  },
+  {
+    id: 'recovery-partial-nodes',
+    readmeAnchor: 'Unterminated or invalid constructs keep their partial nodes',
+    summary: 'Invalid input keeps partial nodes with hasError; the reference degrades to ERROR nodes differently.',
+  },
+  {
+    id: 'trailing-connector',
+    readmeAnchor: 'A trailing connector (`ls &&`, `ls |`)',
+    summary: 'Trailing && / | yields a single-child list/pipeline with hasError.',
+  },
+  {
+    id: 'empty-backtick',
+    readmeAnchor: 'An empty backtick substitution',
+    summary: '`` (and whitespace-only pairs) parse as an empty command_substitution; the reference has a single `` token.',
+  },
+  {
+    id: 'empty-command-substitution',
+    readmeAnchor: 'An empty command substitution (`$( )`)',
+    summary: '$( ) is a clean empty command_substitution; the reference inserts a zero-width command.',
+  },
+  {
+    id: 'arithmetic-hex',
+    readmeAnchor: 'a hex literal is a `number`',
+    summary: '$((0x1F)) → number "0x1F"; the reference produces variable_name (a quirk).',
+  },
+  {
+    id: 'test-paren-group-logical',
+    readmeAnchor: '`[[ ((a) == x) && y ]]`',
+    summary: 'Parenthesized test group followed by && parses cleanly; the reference mis-reads it.',
+  },
+  {
+    id: 'test-extglob-rejected-group',
+    readmeAnchor: 'An extglob group the reference rejects',
+    summary: 'Groups after a literal or dot (x@(y|z)w, *.@(a)) are errors in the reference; recovery shapes differ.',
+  },
+  {
+    id: 'test-negative-decimal',
+    readmeAnchor: 'A negative decimal operand (`[[ $n == -0.5 ]]`)',
+    summary: 'This parser produces extglob_pattern "-0.5"; the reference a unary_expression.',
+  },
+  {
+    id: 'test-fused-operator-expansion',
+    readmeAnchor: 'A test operator fused with an expansion (`[[ -x$f && … ]]`)',
+    summary: 'Flat concatenation of -x and $f here; the reference wraps a unary_expression.',
+  },
+  {
+    id: 'test-escaped-pipe-pattern',
+    readmeAnchor: 'An escaped `|` inside a pattern (`[[ $a == foo\\|bar ]]`)',
+    summary: 'One extglob_pattern "foo\\|bar" here; the reference splits the expression.',
+  },
+  {
+    id: 'test-pattern-two-constructs',
+    readmeAnchor: 'A comparison right side with TWO substitutions or quotes',
+    summary: '*${x}*${y} / *"s"*"t" — the reference recovers with a nested binary_expression; shapes differ.',
+  },
+  {
+    id: 'string-content-newlines',
+    readmeAnchor: '`string_content` is not split at newlines',
+    summary: 'Multiline strings keep one string_content; the reference splits it.',
+  },
+  {
+    id: 'expansion-bang-hash-special',
+    readmeAnchor: '`${!# }` and `${!## }`',
+    summary: 'Pathological ${!#…} forms: the reference recovers with zero-width tokens; shapes differ.',
+  },
+  {
+    id: 'number-base-with-expansion',
+    readmeAnchor: 'A base prefix fused with an expansion (`10#${x}`)',
+    summary: 'The reference wraps the whole thing in a number node (a quirk); this parser a concatenation.',
+  },
+  {
+    id: 'expansion-default-array',
+    readmeAnchor: '`${v:-(default)}`',
+    summary: 'A parenthesized default value is a word here; the reference parses an array node (a quirk).',
+  },
+  {
+    id: 'escaped-space-argument',
+    readmeAnchor: 'An escaped space or tab between arguments',
+    summary: 'The reference drops escaped spaces/tabs as extras (its tree has gaps, words split); this parser keeps them in the word.',
+  },
+  {
+    id: 'expansion-equals-operator',
+    readmeAnchor: '`${=1}`',
+    summary: 'The zsh-style ${=…} form: the reference produces variable_name; this parser a word.',
+  },
+  {
+    id: 'dollar-backtick-substitution',
+    readmeAnchor: 'A `$` directly fused with a backtick substitution',
+    summary: '$`cmd` is one command_substitution with a $` token in the reference; here $ + command_substitution.',
+  },
+  {
+    id: 'cfor-compound-assign-negative',
+    readmeAnchor: 'A negative literal after a compound assignment in a c-style for header',
+    summary: 'j *= -1: the reference folds the sign into a number "-1"; this parser a unary_expression.',
+  },
+  {
+    id: 'case-short-option-pattern',
+    readmeAnchor: 'A single-dash short-option case pattern',
+    summary: '-o) is a word here (its usual reference form); the reference flips to extglob_pattern in some scanner states.',
+  },
+  {
+    id: 'case-pattern-after-continuation',
+    readmeAnchor: 'A case pattern directly after a line continuation',
+    summary: 'A pattern after \\<newline> is a word in the reference (scanner-state quirk); a continuation INSIDE a pattern is an error there. This parser classifies by content and keeps continuations.',
+  },
+  {
+    id: 'nonascii-identifier-assignment',
+    readmeAnchor: 'A non-ASCII “identifier” in assignment position',
+    summary: '变量=值 is a hasError variable_assignment in the reference; this parser keeps it a plain command word (bash itself rejects such names).',
+  },
+  {
+    id: 'case-rejected-group',
+    readmeAnchor: 'An extglob group the reference rejects in a case pattern',
+    summary: 'x@(y)): the reference reparses the group as a new case item; this parser degrades the whole pattern to ERROR.',
+  },
+  {
+    id: 'expansion-replacement-escaped-backslash',
+    readmeAnchor: 'A double backslash in a replacement value (`${x// /\\\\|}`)',
+    summary: 'The reference drops the first backslash (a quirk); this parser keeps both.',
+  },
+];
+
+const KNOWN_DIFFERENCE_IDS: ReadonlySet<string> = new Set(KNOWN_DIFFERENCES.map((entry) => entry.id));
+
+export function isKnownDifferenceId(id: string): boolean {
+  return KNOWN_DIFFERENCE_IDS.has(id);
+}
diff --git a/packages/tree-sitter-bash/test/node.test.ts b/packages/tree-sitter-bash/test/node.test.ts
new file mode 100644
index 0000000000..16d5219808
--- /dev/null
+++ b/packages/tree-sitter-bash/test/node.test.ts
@@ -0,0 +1,132 @@
+import { describe, expect, it } from 'vitest';
+
+import { createNode, descendantsOfType } from '#/node';
+
+const SOURCE = 'echo foo; echo bar';
+
+/** program
+ *      ├─ command "echo foo"
+ *      │    ├─ command_name "echo"
+ *      │    │    └─ word "echo"
+ *      │    └─ word "foo"
+ *      ├─ ";" (anonymous)
+ *      └─ command "echo bar"
+ *           └─ word "bar"
+ */
+function buildTree() {
+  const program = createNode({ type: 'program', source: SOURCE, startIndex: 0, endIndex: SOURCE.length });
+  const cmd1 = program.addChild(createNode({ type: 'command', source: SOURCE, startIndex: 0, endIndex: 8 }));
+  const name1 = cmd1.addChild(createNode({ type: 'command_name', source: SOURCE, startIndex: 0, endIndex: 4 }));
+  name1.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 4 }));
+  cmd1.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 5, endIndex: 8 }));
+  program.addChild(createNode({ type: ';', source: SOURCE, startIndex: 8, endIndex: 9, isNamed: false }));
+  const cmd2 = program.addChild(createNode({ type: 'command', source: SOURCE, startIndex: 10, endIndex: 18 }));
+  cmd2.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 15, endIndex: 18 }));
+  return program;
+}
+
+describe('createNode', () => {
+  it('pre-stores text as the UTF-16 slice of the source', () => {
+    const node = createNode({ type: 'word', source: SOURCE, startIndex: 5, endIndex: 8 });
+    expect(node.text).toBe('foo');
+    expect(node.startIndex).toBe(5);
+    expect(node.endIndex).toBe(8);
+    expect(node.isNamed).toBe(true);
+    expect(node.parent).toBeNull();
+  });
+
+  it('rejects out-of-range offsets', () => {
+    expect(() => createNode({ type: 'word', source: SOURCE, startIndex: -1, endIndex: 2 })).toThrow(RangeError);
+    expect(() => createNode({ type: 'word', source: SOURCE, startIndex: 3, endIndex: 2 })).toThrow(RangeError);
+    expect(() => createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 100 })).toThrow(RangeError);
+  });
+
+  it('rejects attaching a child that already has a parent', () => {
+    const a = createNode({ type: 'program', source: SOURCE, startIndex: 0, endIndex: SOURCE.length });
+    const b = createNode({ type: 'program', source: SOURCE, startIndex: 0, endIndex: SOURCE.length });
+    const child = createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 4 });
+    a.addChild(child);
+    expect(() => b.addChild(child)).toThrow(/already has a parent/);
+  });
+
+  it('rejects children outside the parent range', () => {
+    const parent = createNode({ type: 'command', source: SOURCE, startIndex: 0, endIndex: 8 });
+    expect(() =>
+      parent.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 9 })),
+    ).toThrow(RangeError);
+  });
+
+  it('rejects overlapping or out-of-order siblings', () => {
+    const parent = createNode({ type: 'command', source: SOURCE, startIndex: 0, endIndex: 8 });
+    parent.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 2, endIndex: 5 }));
+    expect(() => parent.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 4, endIndex: 7 }))).toThrow(
+      RangeError,
+    );
+    expect(() => parent.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 2 }))).toThrow(
+      RangeError,
+    );
+    // Adjacent (start == previous end) is fine.
+    parent.addChild(createNode({ type: 'word', source: SOURCE, startIndex: 5, endIndex: 8 }));
+    expect(parent.children).toHaveLength(2);
+  });
+});
+
+describe('children vs namedChildren', () => {
+  it('keeps anonymous nodes out of namedChildren but in children', () => {
+    const program = buildTree();
+    expect(program.children.map((c) => c.type)).toEqual(['command', ';', 'command']);
+    expect(program.namedChildren.map((c) => c.type)).toEqual(['command', 'command']);
+    expect(program.children[1]?.isNamed).toBe(false);
+  });
+
+  it('wires parent pointers', () => {
+    const program = buildTree();
+    const [cmd1] = program.namedChildren;
+    expect(cmd1?.parent).toBe(program);
+    expect(cmd1?.namedChildren[0]?.parent).toBe(cmd1);
+  });
+});
+
+describe('descendantsOfType', () => {
+  it('returns matching named descendants in pre-order', () => {
+    const program = buildTree();
+    const words = descendantsOfType(program, 'word');
+    expect(words.map((w) => w.text)).toEqual(['echo', 'foo', 'bar']);
+  });
+
+  it('matches multiple types and skips anonymous nodes', () => {
+    const program = buildTree();
+    const nodes = descendantsOfType(program, 'command', ';');
+    expect(nodes.map((n) => n.type)).toEqual(['command', 'command']);
+  });
+
+  it('returns every named descendant when no type is given', () => {
+    const program = buildTree();
+    expect(descendantsOfType(program).map((n) => n.type)).toEqual([
+      'command',
+      'command_name',
+      'word',
+      'word',
+      'command',
+      'word',
+    ]);
+  });
+
+  it('does not include the root itself', () => {
+    const leaf = createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 4 });
+    expect(descendantsOfType(leaf, 'word')).toEqual([]);
+  });
+
+  it('handles pathologically deep trees without overflowing the stack', () => {
+    // 200k-deep chain; a recursive walk would overflow the call stack.
+    const depth = 200_000;
+    const nodes = [];
+    for (let i = 0; i < depth; i++) {
+      nodes.push(createNode({ type: 'word', source: SOURCE, startIndex: 0, endIndex: 1 }));
+    }
+    for (let i = depth - 2; i >= 0; i--) {
+      nodes[i]!.addChild(nodes[i + 1]!);
+    }
+    expect(descendantsOfType(nodes[0]!, 'word')).toHaveLength(depth - 1);
+  });
+});
diff --git a/packages/tree-sitter-bash/test/parse.test.ts b/packages/tree-sitter-bash/test/parse.test.ts
new file mode 100644
index 0000000000..bb322741f4
--- /dev/null
+++ b/packages/tree-sitter-bash/test/parse.test.ts
@@ -0,0 +1,519 @@
+import { describe, expect, it } from 'vitest';
+
+import { Aborted, ParseBudget } from '#/budget';
+import type { SyntaxNode } from '#/node';
+import { descendantsOfType } from '#/node';
+import { parse } from '#/parse';
+
+/** Compact S-expression of the tree: named leaves are (type "text"),
+ *  anonymous leaves are just their quoted type, inner nodes nest. */
+function sexp(node: SyntaxNode): string {
+  if (node.children.length === 0) {
+    return node.isNamed ? `(${node.type} ${JSON.stringify(node.text)})` : JSON.stringify(node.type);
+  }
+  const label = node.isNamed ? node.type : JSON.stringify(node.type);
+  return `(${label} ${node.children.map(sexp).join(' ')})`;
+}
+
+function parseOk(source: string): { rootNode: SyntaxNode; hasError: boolean } {
+  const result = parse(source);
+  expect(result.ok).toBe(true);
+  if (!result.ok) throw new Error('unreachable');
+  return result;
+}
+
+/** Parse and assert the exact tree shape plus the hasError flag. */
+function expectTree(source: string, expected: string, hasError = false): SyntaxNode {
+  const { rootNode, hasError: actual } = parseOk(source);
+  expect(actual).toBe(hasError);
+  expect(sexp(rootNode)).toBe(expected);
+  return rootNode;
+}
+
+describe('ParseBudget', () => {
+  it('counts nodes and stays under the cap', () => {
+    const budget = new ParseBudget({ timeoutMs: 60_000, maxNodes: 3 });
+    budget.tick();
+    budget.tick();
+    expect(budget.nodesUsed).toBe(2);
+  });
+
+  it('throws Aborted when the node cap is exceeded', () => {
+    const budget = new ParseBudget({ timeoutMs: 60_000, maxNodes: 1 });
+    budget.tick();
+    expect(() => {
+      budget.tick();
+    }).toThrow(Aborted);
+  });
+
+  it('throws Aborted once the deadline is reached', () => {
+    const budget = new ParseBudget({ timeoutMs: 0, maxNodes: 1_000 });
+    expect(() => {
+      budget.tick();
+    }).toThrow(Aborted);
+  });
+
+  it('applies documented defaults', () => {
+    const budget = new ParseBudget();
+    // Default node cap is 50_000: the 50_000th node is still allowed.
+    for (let i = 0; i < 50_000; i++) budget.tick();
+    expect(() => {
+      budget.tick();
+    }).toThrow(Aborted);
+  });
+});
+
+describe('parse entry contract', () => {
+  it('parses an empty source into an empty program', () => {
+    const { rootNode, hasError } = parseOk('');
+    expect(rootNode.type).toBe('program');
+    expect(rootNode.text).toBe('');
+    expect(rootNode.children).toHaveLength(0);
+    expect(hasError).toBe(false);
+  });
+
+  it('returns { ok: false, reason: "aborted" } when the node budget is exceeded', () => {
+    expect(parse('echo hello', { timeoutMs: 60_000, maxNodes: 1 })).toEqual({ ok: false, reason: 'aborted' });
+  });
+
+  it('returns { ok: false, reason: "aborted" } when the deadline has passed', () => {
+    expect(parse('echo hello', { timeoutMs: 0 })).toEqual({ ok: false, reason: 'aborted' });
+  });
+
+  it('aborts instead of hanging on tens of thousands of statements', () => {
+    expect(parse('ls; '.repeat(20_000))).toEqual({ ok: false, reason: 'aborted' });
+  });
+});
+
+describe('simple commands', () => {
+  it('parses a command with arguments', () => {
+    expectTree('ls -la', `(program (command (command_name (word "ls")) (word "-la")))`);
+  });
+
+  it('parses double-quoted strings with expansions and raw strings', () => {
+    expectTree(
+      `echo "hello $NAME" 'raw'`,
+      `(program (command (command_name (word "echo")) (string "\\"" (string_content "hello ") (simple_expansion "$" (variable_name "NAME")) "\\"") (raw_string "'raw'")))`,
+    );
+  });
+
+  it('parses a pipeline inside && / || lists, left associative', () => {
+    expectTree(
+      'FOO=bar env | grep FOO && echo ok || echo fail',
+      `(program (list (list (pipeline (command (variable_assignment (variable_name "FOO") "=" (word "bar")) (command_name (word "env"))) "|" (command (command_name (word "grep")) (word "FOO"))) "&&" (command (command_name (word "echo")) (word "ok"))) "||" (command (command_name (word "echo")) (word "fail"))))`,
+    );
+  });
+
+  it('parses ! as negated_command', () => {
+    expectTree(
+      '! grep -q x file',
+      `(program (negated_command "!" (command (command_name (word "grep")) (word "-q") (word "x") (word "file"))))`,
+    );
+  });
+
+  it('attaches comments as named children', () => {
+    expectTree(
+      '# a comment\necho hi # trailing',
+      `(program (comment "# a comment") (command (command_name (word "echo")) (word "hi")) (comment "# trailing"))`,
+    );
+  });
+
+  it('distinguishes number from word', () => {
+    expectTree(
+      'echo 123 45.6',
+      `(program (command (command_name (word "echo")) (number "123") (word "45.6")))`,
+    );
+  });
+
+  it('parses concatenations of adjacent pieces', () => {
+    expectTree(
+      `echo a"b"c'd'$e\${f}g`,
+      `(program (command (command_name (word "echo")) (concatenation (word "a") (string "\\"" (string_content "b") "\\"") (word "c") (raw_string "'d'") (simple_expansion "$" (variable_name "e")) (expansion "\${" (variable_name "f") "}") (word "g"))))`,
+    );
+  });
+
+  it('splits { } into single-character word pieces', () => {
+    expectTree(
+      'echo {a,b}',
+      `(program (command (command_name (word "echo")) (concatenation (word "{") (word "a,b") (word "}"))))`,
+    );
+  });
+
+  it('treats a lone $ as an anonymous argument', () => {
+    expectTree('echo $', `(program (command (command_name (word "echo")) "$"))`);
+  });
+});
+
+describe('expansions and substitutions', () => {
+  it('parses backtick and $() command substitutions', () => {
+    expectTree(
+      'echo `hostname` $(date)',
+      `(program (command (command_name (word "echo")) (command_substitution "\`" (command (command_name (word "hostname"))) "\`") (command_substitution "$(" (command (command_name (word "date"))) ")")))`,
+    );
+  });
+
+  it('parses nested command substitutions recursively', () => {
+    expectTree(
+      'echo $(a $(b c))',
+      `(program (command (command_name (word "echo")) (command_substitution "$(" (command (command_name (word "a")) (command_substitution "$(" (command (command_name (word "b")) (word "c")) ")")) ")")))`,
+    );
+  });
+
+  it('parses ${...} expansions with operators and defaults', () => {
+    expectTree(
+      'echo ${v:-d} ${x} $1 $@',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "v") ":-" (word "d") "}") (expansion "\${" (variable_name "x") "}") (simple_expansion "$" (variable_name "1")) (simple_expansion "$" (special_variable_name "@"))))`,
+    );
+  });
+
+  it('parses ${#v}, ${v:=word} and ${a[0]}', () => {
+    expectTree('echo ${#v}', `(program (command (command_name (word "echo")) (expansion "\${" "#" (variable_name "v") "}")))`);
+    expectTree(
+      'echo ${v:=word}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "v") ":=" (word "word") "}")))`,
+    );
+    expectTree(
+      'echo ${a[0]}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (subscript (variable_name "a") "[" (number "0") "]") "}")))`,
+    );
+  });
+
+  it('parses expansions nested inside expansions', () => {
+    expectTree(
+      'echo ${v:-$(cmd)}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "v") ":-" (command_substitution "$(" (command (command_name (word "cmd"))) ")") "}")))`,
+    );
+  });
+
+  it('parses process substitutions', () => {
+    expectTree(
+      'diff <(cmd) >(other)',
+      `(program (command (command_name (word "diff")) (process_substitution "<(" (command (command_name (word "cmd"))) ")") (process_substitution ">(" (command (command_name (word "other"))) ")")))`,
+    );
+  });
+
+  it('parses $((...)) as a real arithmetic expression', () => {
+    expectTree(
+      'echo $((1<<2))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (number "1") "<<" (number "2")) "))")))`,
+    );
+  });
+});
+
+describe('redirects', () => {
+  it('parses output redirect and fd duplication', () => {
+    expectTree(
+      'cmd > out 2>&1',
+      `(program (redirected_statement (command (command_name (word "cmd"))) (file_redirect ">" (word "out")) (file_redirect (file_descriptor "2") ">&" (number "1"))))`,
+    );
+  });
+
+  it('parses &>> appending both streams', () => {
+    expectTree(
+      'cmd &>> log',
+      `(program (redirected_statement (command (command_name (word "cmd"))) (file_redirect "&>>" (word "log"))))`,
+    );
+  });
+
+  it('parses <> read-write redirect with a file descriptor', () => {
+    expectTree(
+      'exec 3<>file',
+      `(program (redirected_statement (command (command_name (word "exec"))) (file_redirect (file_descriptor "3") "<>" (word "file"))))`,
+    );
+  });
+
+  it('parses >|, <&-, < and fd redirects', () => {
+    expectTree(
+      'cmd >| file',
+      `(program (redirected_statement (command (command_name (word "cmd"))) (file_redirect ">|" (word "file"))))`,
+    );
+    expectTree('cmd >&-', `(program (redirected_statement (command (command_name (word "cmd"))) (file_redirect ">&-")))`);
+    expectTree(
+      'cmd < in',
+      `(program (redirected_statement (command (command_name (word "cmd"))) (file_redirect "<" (word "in"))))`,
+    );
+    expectTree(
+      'cmd 2>/dev/null',
+      `(program (redirected_statement (command (command_name (word "cmd"))) (file_redirect (file_descriptor "2") ">" (word "/dev/null"))))`,
+    );
+  });
+
+  it('keeps a herestring inside the command node', () => {
+    expectTree(
+      'cmd <<< "$str"',
+      `(program (command (command_name (word "cmd")) (herestring_redirect "<<<" (string "\\"" (simple_expansion "$" (variable_name "str")) "\\""))))`,
+    );
+  });
+
+  it('wraps a command carrying both a herestring and a file redirect', () => {
+    expectTree(
+      'cmd <<< x > out',
+      `(program (redirected_statement (command (command_name (word "cmd")) (herestring_redirect "<<<" (word "x"))) (file_redirect ">" (word "out"))))`,
+    );
+  });
+
+  it('parses a redirect-only statement', () => {
+    expectTree('> out', `(program (redirected_statement (file_redirect ">" (word "out"))))`);
+  });
+
+  it('keeps a prefix redirect inside the command', () => {
+    expectTree(
+      '> a cmd x',
+      `(program (command (file_redirect ">" (word "a")) (command_name (word "cmd")) (word "x")))`,
+    );
+  });
+
+  it('wraps subshells and negated commands with their redirects', () => {
+    expectTree(
+      '(ls) > out',
+      `(program (redirected_statement (subshell "(" (command (command_name (word "ls"))) ")") (file_redirect ">" (word "out"))))`,
+    );
+    expectTree(
+      '! ls > out',
+      `(program (redirected_statement (negated_command "!" (command (command_name (word "ls")))) (file_redirect ">" (word "out"))))`,
+    );
+  });
+});
+
+describe('heredocs', () => {
+  it('parses a heredoc with expansions in the body', () => {
+    expectTree(
+      'cat <<EOF\nhello $USER\nEOF',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "EOF") (heredoc_body (heredoc_content "hello ") (simple_expansion "$" (variable_name "USER")) (heredoc_content "\\n")) (heredoc_end "EOF"))))`,
+    );
+  });
+
+  it('strips first-line tabs for <<- and keeps the marker indented', () => {
+    expectTree(
+      'cat <<-EOF\n\tindented\n\tEOF',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<-" (heredoc_start "EOF") (heredoc_body (heredoc_content "indented\\n\\t")) (heredoc_end "EOF"))))`,
+    );
+  });
+
+  it('does not expand the body of a quoted delimiter', () => {
+    expectTree(
+      `cat <<'EOF'\nraw $notexpanded\nEOF`,
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "'EOF'") (heredoc_body "raw $notexpanded\\n") (heredoc_end "EOF"))))`,
+    );
+  });
+
+  it('absorbs trailing words and redirects into the heredoc_redirect', () => {
+    expectTree(
+      'cat <<EOF arg > out\nbody\nEOF',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "EOF") (word "arg") (file_redirect ">" (word "out")) (heredoc_body (heredoc_content "body\\n")) (heredoc_end "EOF"))))`,
+    );
+  });
+
+  it('absorbs a pipeline tail into the heredoc_redirect', () => {
+    expectTree(
+      'cat <<EOF | grep x\nbody\nEOF',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "EOF") (pipeline "|" (command (command_name (word "grep")) (word "x"))) (heredoc_body (heredoc_content "body\\n")) (heredoc_end "EOF"))))`,
+    );
+  });
+
+  it('absorbs `;` follow-up statements into the heredoc_redirect', () => {
+    expectTree(
+      'cat <<EOF; echo x\nbody\nEOF',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "EOF") ";" (command (command_name (word "echo")) (word "x")) (heredoc_body (heredoc_content "body\\n")) (heredoc_end "EOF"))))`,
+    );
+  });
+
+  it('degrades a second heredoc on the same line to ERROR (matches tree-sitter-bash)', () => {
+    expectTree(
+      'cat <<A <<B\nba\nA\nbb\nB',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "A") (ERROR "<<" (word "B")) (heredoc_body (heredoc_content "ba\\n")) (heredoc_end "A"))) (command (command_name (word "bb"))) (command (command_name (word "B"))))`,
+      true,
+    );
+  });
+
+  it('recovers from an unterminated heredoc', () => {
+    expectTree(
+      'cat <<EOF\nno end marker',
+      `(program (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "EOF") (heredoc_body (heredoc_content "no end marker")))))`,
+      true,
+    );
+  });
+});
+
+describe('statement lists', () => {
+  it('mixes semicolons and newlines as terminators', () => {
+    expectTree(
+      'ls; echo a\necho b',
+      `(program (command (command_name (word "ls"))) ";" (command (command_name (word "echo")) (word "a")) (command (command_name (word "echo")) (word "b")))`,
+    );
+  });
+
+  it('parses & background terminators', () => {
+    expectTree('a & b &', `(program (command (command_name (word "a"))) "&" (command (command_name (word "b"))) "&")`);
+  });
+
+  it('parses |& pipelines', () => {
+    expectTree(
+      'cmd1 |& cmd2',
+      `(program (pipeline (command (command_name (word "cmd1"))) "|&" (command (command_name (word "cmd2")))))`,
+    );
+  });
+
+  it('parses a subshell', () => {
+    expectTree('(ls)', `(program (subshell "(" (command (command_name (word "ls"))) ")"))`);
+  });
+
+  it('parses standalone assignments', () => {
+    expectTree('FOO=bar ls', `(program (command (variable_assignment (variable_name "FOO") "=" (word "bar")) (command_name (word "ls"))))`);
+    expectTree('VAR=value', `(program (variable_assignment (variable_name "VAR") "=" (word "value")))`);
+    expectTree(
+      'A=1 B=2',
+      `(program (variable_assignments (variable_assignment (variable_name "A") "=" (number "1")) (variable_assignment (variable_name "B") "=" (number "2"))))`,
+    );
+  });
+
+  it('recovers a trailing && with a partial list and hasError', () => {
+    expectTree('ls &&', `(program (list (command (command_name (word "ls"))) "&&"))`, true);
+  });
+
+  it('recovers a trailing | with a partial pipeline and hasError', () => {
+    expectTree('ls |', `(program (pipeline (command (command_name (word "ls"))) "|"))`, true);
+  });
+
+  it('continues a list across a newline after &&', () => {
+    expectTree(
+      'a &&\nb',
+      `(program (list (command (command_name (word "a"))) "&&" (command (command_name (word "b")))))`,
+    );
+  });
+
+  it('splits compound commands into separate command nodes', () => {
+    const root = expectTree(
+      'git status && rm -rf /',
+      `(program (list (command (command_name (word "git")) (word "status")) "&&" (command (command_name (word "rm")) (word "-rf") (word "/"))))`,
+    );
+    const commands = descendantsOfType(root, 'command');
+    expect(commands).toHaveLength(2);
+    const argv = commands.map((cmd) =>
+      cmd.namedChildren.map((part) => (part.type === 'command_name' ? part.namedChildren[0]!.text : part.text)),
+    );
+    expect(argv).toEqual([
+      ['git', 'status'],
+      ['rm', '-rf', '/'],
+    ]);
+  });
+});
+
+describe('offsets', () => {
+  it('uses UTF-16 code unit offsets', () => {
+    // 🎉 is two UTF-16 code units.
+    const { rootNode } = parseOk('x🎉y; ls');
+    expect(rootNode.endIndex).toBe(8);
+    const commands = descendantsOfType(rootNode, 'command');
+    expect(commands).toHaveLength(2);
+    expect(commands[0]!.startIndex).toBe(0);
+    expect(commands[0]!.endIndex).toBe(4);
+    expect(commands[1]!.startIndex).toBe(6);
+    expect(commands[1]!.endIndex).toBe(8);
+    expect(commands[1]!.text).toBe('ls');
+  });
+
+  it('node.text always equals the source slice', () => {
+    const source = 'FOO=bar env | grep FOO && echo "ok $X"';
+    const { rootNode } = parseOk(source);
+    const walk = (node: SyntaxNode): void => {
+      expect(node.text).toBe(source.slice(node.startIndex, node.endIndex));
+      node.children.forEach(walk);
+    };
+    walk(rootNode);
+  });
+});
+
+describe('adversarial input', () => {
+  it('recovers unterminated quotes, substitutions and expansions', () => {
+    for (const source of ['echo "abc', "echo 'abc", 'echo $(foo', 'echo ${foo', 'echo `foo']) {
+      const result = parse(source);
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(true);
+    }
+  });
+
+  it('recovers stray operators without throwing', () => {
+    for (const source of ['echo )', '&& echo', '| echo', ')']) {
+      const result = parse(source);
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(true);
+    }
+  });
+
+  it('parses a 500KB double-quoted string under the default budget', () => {
+    // Regression: per-character budget.tick() used to burn the 50k node cap
+    // on long strings even though only a handful of nodes are created.
+    const body = 'a'.repeat(500 * 1024);
+    const result = parse(`echo "${body}"`);
+    expect(result.ok).toBe(true);
+    if (!result.ok) return;
+    expect(result.hasError).toBe(false);
+    const content = descendantsOfType(result.rootNode, 'string_content');
+    expect(content).toHaveLength(1);
+    expect(content[0]!.text).toBe(body);
+  });
+
+  it('parses a 500KB heredoc body under the default budget', () => {
+    const body = 'b'.repeat(500 * 1024);
+    const result = parse(`cat <<EOF\n${body}\nEOF`);
+    expect(result.ok).toBe(true);
+    if (!result.ok) return;
+    expect(result.hasError).toBe(false);
+    const content = descendantsOfType(result.rootNode, 'heredoc_content');
+    expect(content).toHaveLength(1);
+    expect(content[0]!.text).toBe(`${body}\n`);
+  });
+
+  it('handles 1000-level nested ${} expansions within the depth cap', () => {
+    const source = 'echo ' + '${a:-'.repeat(1000) + 'x' + '}'.repeat(1000);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('handles 1000-level nested command substitutions within the depth cap', () => {
+    const source = '$('.repeat(1000) + 'x' + ')'.repeat(1000);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('handles 1000-level nested subshells within the depth cap', () => {
+    const source = '('.repeat(1000) + 'x' + ')'.repeat(1000);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('parses a multi-megabyte single word without throwing', () => {
+    const source = 'a'.repeat(2 * 1024 * 1024);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (!result.ok) return;
+    expect(result.hasError).toBe(false);
+    const word = result.rootNode.namedChildren[0]!.namedChildren[0]!.namedChildren[0]!;
+    expect(word.type).toBe('word');
+    expect(word.text).toBe(source);
+  });
+
+  it('does not throw on random binary garbage', () => {
+    let source = '';
+    for (let i = 0; i < 20_000; i++) {
+      source += String.fromCodePoint(Math.floor(Math.random() * 0xd800));
+    }
+    // The call itself must never throw; either it parses (possibly with
+    // errors) or the budget cuts it off.
+    const result = parse(source);
+    if (result.ok) {
+      expect(result.rootNode.type).toBe('program');
+    } else {
+      expect(result.reason).toBe('aborted');
+    }
+  });
+
+  it('does not throw on NUL bytes', () => {
+    const result = parse('echo \0\0 test \0');
+    expect(result.ok).toBe(true);
+  });
+});
diff --git a/packages/tree-sitter-bash/test/parser-compound.test.ts b/packages/tree-sitter-bash/test/parser-compound.test.ts
new file mode 100644
index 0000000000..a411fb4236
--- /dev/null
+++ b/packages/tree-sitter-bash/test/parser-compound.test.ts
@@ -0,0 +1,1232 @@
+import { describe, expect, it } from 'vitest';
+
+import type { SyntaxNode } from '#/node';
+import { descendantsOfType } from '#/node';
+import { parse } from '#/parse';
+
+/** Compact S-expression of the tree: named leaves are (type "text"),
+ *  anonymous leaves are just their quoted type, inner nodes nest. */
+function sexp(node: SyntaxNode): string {
+  if (node.children.length === 0) {
+    return node.isNamed ? `(${node.type} ${JSON.stringify(node.text)})` : JSON.stringify(node.type);
+  }
+  const label = node.isNamed ? node.type : JSON.stringify(node.type);
+  return `(${label} ${node.children.map(sexp).join(' ')})`;
+}
+
+function parseOk(source: string): { rootNode: SyntaxNode; hasError: boolean } {
+  const result = parse(source);
+  expect(result.ok).toBe(true);
+  if (!result.ok) throw new Error('unreachable');
+  return result;
+}
+
+/** Parse and assert the exact tree shape plus the hasError flag. */
+function expectTree(source: string, expected: string, hasError = false): SyntaxNode {
+  const { rootNode, hasError: actual } = parseOk(source);
+  expect(actual).toBe(hasError);
+  expect(sexp(rootNode)).toBe(expected);
+  return rootNode;
+}
+
+describe('if / while / until', () => {
+  it('parses a one-line if statement', () => {
+    expectTree(
+      'if true; then echo yes; fi',
+      `(program (if_statement "if" (command (command_name (word "true"))) ";" "then" (command (command_name (word "echo")) (word "yes")) ";" "fi"))`,
+    );
+  });
+
+  it('parses elif and else clauses', () => {
+    expectTree(
+      'if a; then b; elif c; then d; else e; fi',
+      `(program (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" (elif_clause "elif" (command (command_name (word "c"))) ";" "then" (command (command_name (word "d"))) ";") (else_clause "else" (command (command_name (word "e"))) ";") "fi"))`,
+    );
+  });
+
+  it('parses a multiline if without emitting newline nodes', () => {
+    expectTree(
+      'if a\nthen\n  b\nfi',
+      `(program (if_statement "if" (command (command_name (word "a"))) "then" (command (command_name (word "b"))) "fi"))`,
+    );
+  });
+
+  it('parses a while loop with a trailing redirect', () => {
+    expectTree(
+      'while read -r line; do echo "$line"; done < file',
+      `(program (redirected_statement (while_statement "while" (command (command_name (word "read")) (word "-r") (word "line")) ";" (do_group "do" (command (command_name (word "echo")) (string "\\"" (simple_expansion "$" (variable_name "line")) "\\"")) ";" "done")) (file_redirect "<" (word "file"))))`,
+    );
+  });
+
+  it('parses until as a while_statement', () => {
+    expectTree(
+      'until x; do y; done',
+      `(program (while_statement "until" (command (command_name (word "x"))) ";" (do_group "do" (command (command_name (word "y"))) ";" "done")))`,
+    );
+  });
+
+  it('parses an empty do_group', () => {
+    expectTree(
+      'while x; do done',
+      `(program (while_statement "while" (command (command_name (word "x"))) ";" (do_group "do" "done")))`,
+    );
+  });
+
+  it('only treats keywords as keywords in statement position', () => {
+    expectTree(
+      'echo if for while',
+      `(program (command (command_name (word "echo")) (word "if") (word "for") (word "while")))`,
+    );
+    expectTree('fi', `(program (command (command_name (word "fi"))))`);
+    expectTree('done', `(program (command (command_name (word "done"))))`);
+    expectTree('}', `(program (command (command_name (word "}"))))`);
+    expectTree('time ls -la', `(program (command (command_name (word "time")) (word "ls") (word "-la")))`);
+    // Reserved words are plain arguments inside conditions and bodies.
+    expectTree(
+      'if echo then; then x; fi',
+      `(program (if_statement "if" (command (command_name (word "echo")) (word "then")) ";" "then" (command (command_name (word "x"))) ";" "fi"))`,
+    );
+    // A prefix assignment disables the keyword reading.
+    expectTree(
+      'x=1 if a; then b; fi',
+      `(program (command (variable_assignment (variable_name "x") "=" (number "1")) (command_name (word "if")) (word "a")) ";" (command (command_name (word "then")) (word "b")) ";" (command (command_name (word "fi"))))`,
+    );
+  });
+});
+
+describe('for / select', () => {
+  it('parses a for-in loop', () => {
+    expectTree(
+      'for f in a b c; do echo "$f"; done',
+      `(program (for_statement "for" (variable_name "f") "in" (word "a") (word "b") (word "c") ";" (do_group "do" (command (command_name (word "echo")) (string "\\"" (simple_expansion "$" (variable_name "f")) "\\"")) ";" "done")))`,
+    );
+  });
+
+  it('parses a for loop without in', () => {
+    expectTree(
+      'for f; do x; done',
+      `(program (for_statement "for" (variable_name "f") ";" (do_group "do" (command (command_name (word "x"))) ";" "done")))`,
+    );
+  });
+
+  it('parses select as a for_statement', () => {
+    expectTree(
+      'select opt in a b; do echo $opt; done',
+      `(program (for_statement "select" (variable_name "opt") "in" (word "a") (word "b") ";" (do_group "do" (command (command_name (word "echo")) (simple_expansion "$" (variable_name "opt"))) ";" "done")))`,
+    );
+  });
+
+  it('parses newline-separated for headers', () => {
+    expectTree(
+      'for f in a b\ndo x\ndone',
+      `(program (for_statement "for" (variable_name "f") "in" (word "a") (word "b") (do_group "do" (command (command_name (word "x"))) "done")))`,
+    );
+  });
+
+  it('flags a bare `in` with no values as an error', () => {
+    expectTree(
+      'select x in; do y; done',
+      `(program (for_statement "select" (variable_name "x") (ERROR "in") ";" (do_group "do" (command (command_name (word "y"))) ";" "done")))`,
+      true,
+    );
+  });
+
+  it('parses a c-style for loop', () => {
+    expectTree(
+      'for ((i=0;i<3;i++)); do echo $i; done',
+      `(program (c_style_for_statement "for" "((" (variable_assignment (variable_name "i") "=" (number "0")) ";" (binary_expression (word "i") "<" (number "3")) ";" (postfix_expression (word "i") "++") "))" ";" (do_group "do" (command (command_name (word "echo")) (simple_expansion "$" (variable_name "i"))) ";" "done")))`,
+    );
+  });
+
+  it('parses a spaced c-style for with an update assignment operator', () => {
+    expectTree(
+      'for (( i = 0; i < 10; i += 2 )); do x; done',
+      `(program (c_style_for_statement "for" "((" (variable_assignment (variable_name "i") "=" (number "0")) ";" (binary_expression (word "i") "<" (number "10")) ";" (binary_expression (word "i") "+=" (number "2")) "))" ";" (do_group "do" (command (command_name (word "x"))) ";" "done")))`,
+    );
+  });
+
+  it('parses an empty c-style for header', () => {
+    expectTree(
+      'for ((;;)); do x; done',
+      `(program (c_style_for_statement "for" "((" ";" ";" "))" ";" (do_group "do" (command (command_name (word "x"))) ";" "done")))`,
+    );
+  });
+
+  it('parses comma-separated c-style for parts', () => {
+    expectTree(
+      'for ((i=0,j=1; i<3; i++,j--)); do x; done',
+      `(program (c_style_for_statement "for" "((" (variable_assignment (variable_name "i") "=" (number "0")) "," (variable_assignment (variable_name "j") "=" (number "1")) ";" (binary_expression (word "i") "<" (number "3")) ";" (postfix_expression (word "i") "++") "," (postfix_expression (word "j") "--") "))" ";" (do_group "do" (command (command_name (word "x"))) ";" "done")))`,
+    );
+  });
+
+  it('parses a compound_statement as the c-style for body', () => {
+    expectTree(
+      'for ((i=0;i<3;i++)) { echo $i; }',
+      `(program (c_style_for_statement "for" "((" (variable_assignment (variable_name "i") "=" (number "0")) ";" (binary_expression (word "i") "<" (number "3")) ";" (postfix_expression (word "i") "++") "))" (compound_statement "{" (command (command_name (word "echo")) (simple_expansion "$" (variable_name "i"))) ";" "}")))`,
+    );
+  });
+});
+
+describe('case', () => {
+  it('parses a multi-item case with all termination forms', () => {
+    expectTree(
+      'case $x in\n  a) echo A ;;\n  b|c) echo BC ;&\n  *) echo other ;;\nesac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" (case_item (word "a") ")" (command (command_name (word "echo")) (word "A")) ";;") (case_item (extglob_pattern "b") "|" (word "c") ")" (command (command_name (word "echo")) (word "BC")) ";&") (case_item (extglob_pattern "*") ")" (command (command_name (word "echo")) (word "other")) ";;") "esac"))`,
+    );
+  });
+
+  it('parses optional parens and empty item bodies', () => {
+    expectTree(
+      'case $x in (a) x ;; esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" (case_item "(" (word "a") ")" (command (command_name (word "x"))) ";;") "esac"))`,
+    );
+    expectTree(
+      'case x in a) ;; esac',
+      `(program (case_statement "case" (word "x") "in" (case_item (word "a") ")" ";;") "esac"))`,
+    );
+    expectTree(
+      'case $x in esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" "esac"))`,
+    );
+  });
+
+  it('parses pattern forms: globs, quotes and expansions', () => {
+    expectTree(
+      'case $x in [a-z]) x ;; esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" (case_item (extglob_pattern "[a-z]") ")" (command (command_name (word "x"))) ";;") "esac"))`,
+    );
+    expectTree(
+      'case $x in "a") x ;; esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" (case_item (string "\\"" (string_content "a") "\\"") ")" (command (command_name (word "x"))) ";;") "esac"))`,
+    );
+    expectTree(
+      'case $x in $v|${w}) x ;; esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" (case_item (simple_expansion "$" (variable_name "v")) "|" (expansion "\${" (variable_name "w") "}") ")" (command (command_name (word "x"))) ";;") "esac"))`,
+    );
+  });
+
+  it('ends the last case_item at esac even without ;;', () => {
+    expectTree(
+      'case $x in a) x esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "x")) "in" (case_item (word "a") ")" (command (command_name (word "x")))) "esac"))`,
+    );
+  });
+
+  it('flags a fallthrough terminator on the last case_item', () => {
+    const { hasError } = parseOk('case $x in a) x ;;& esac');
+    expect(hasError).toBe(true);
+  });
+});
+
+describe('functions and compound statements', () => {
+  it('parses the name() form', () => {
+    expectTree(
+      'foo() { echo hi; }',
+      `(program (function_definition (word "foo") "(" ")" (compound_statement "{" (command (command_name (word "echo")) (word "hi")) ";" "}")))`,
+    );
+  });
+
+  it('parses the function keyword forms', () => {
+    expectTree(
+      'function bar { echo hi; }',
+      `(program (function_definition "function" (word "bar") (compound_statement "{" (command (command_name (word "echo")) (word "hi")) ";" "}")))`,
+    );
+    expectTree(
+      'function baz() { echo hi; return 0; }',
+      `(program (function_definition "function" (word "baz") "(" ")" (compound_statement "{" (command (command_name (word "echo")) (word "hi")) ";" (command (command_name (word "return")) (number "0")) ";" "}")))`,
+    );
+    expectTree(
+      'f () { x; }',
+      `(program (function_definition (word "f") "(" ")" (compound_statement "{" (command (command_name (word "x"))) ";" "}")))`,
+    );
+  });
+
+  it('parses subshell, test and if bodies', () => {
+    expectTree(
+      'foo() ( echo sub )',
+      `(program (function_definition (word "foo") "(" ")" (subshell "(" (command (command_name (word "echo")) (word "sub")) ")")))`,
+    );
+    expectTree(
+      'foo() [[ -f x ]]',
+      `(program (function_definition (word "foo") "(" ")" (test_command "[[" (unary_expression (test_operator "-f") (word "x")) "]]")))`,
+    );
+    expectTree(
+      'foo() if a; then b; fi',
+      `(program (function_definition (word "foo") "(" ")" (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" "fi")))`,
+    );
+  });
+
+  it('keeps a trailing redirect inside the function_definition', () => {
+    expectTree(
+      'foo() { x; } > out',
+      `(program (function_definition (word "foo") "(" ")" (compound_statement "{" (command (command_name (word "x"))) ";" "}") (file_redirect ">" (word "out"))))`,
+    );
+  });
+
+  it('parses a standalone compound statement', () => {
+    expectTree(
+      '{ echo a; echo b; }',
+      `(program (compound_statement "{" (command (command_name (word "echo")) (word "a")) ";" (command (command_name (word "echo")) (word "b")) ";" "}"))`,
+    );
+    expectTree('{ls;}', `(program (compound_statement "{" (command (command_name (word "ls"))) ";" "}"))`);
+  });
+
+  it('parses coproc as an ordinary command (like the reference)', () => {
+    expectTree(
+      'coproc myjob { echo hi; }',
+      `(program (command (command_name (word "coproc")) (word "myjob") (word "{") (word "echo") (word "hi")) ";" (command (command_name (word "}"))))`,
+    );
+  });
+});
+
+describe('test commands', () => {
+  it('parses unary test operators', () => {
+    expectTree(
+      '[[ -f file.txt ]]',
+      `(program (test_command "[[" (unary_expression (test_operator "-f") (word "file.txt")) "]]"))`,
+    );
+    expectTree(
+      '[[ -z $s ]]',
+      `(program (test_command "[[" (unary_expression (test_operator "-z") (simple_expansion "$" (variable_name "s"))) "]]"))`,
+    );
+  });
+
+  it('parses string comparisons', () => {
+    expectTree(
+      '[[ $x == "foo" ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (string "\\"" (string_content "foo") "\\"")) "]]"))`,
+    );
+    expectTree(
+      '[ "$a" = "b" ]',
+      `(program (test_command "[" (binary_expression (string "\\"" (simple_expansion "$" (variable_name "a")) "\\"") "=" (string "\\"" (string_content "b") "\\"")) "]"))`,
+    );
+  });
+
+  it('parses =~ with a regex right side', () => {
+    expectTree(
+      '[[ $x =~ ^ab+c$ ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "=~" (regex "^ab+c$")) "]]"))`,
+    );
+    expectTree(
+      '[[ $x =~ a|b ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "=~" (regex "a|b")) "]]"))`,
+    );
+    expectTree(
+      '[[ $x =~ $re ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "=~" (simple_expansion "$" (variable_name "re"))) "]]"))`,
+    );
+  });
+
+  it('parses pattern right sides: regex after =, extglob_pattern after ==/!=', () => {
+    expectTree(
+      '[[ a = b*c ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (regex "b*c")) "]]"))`,
+    );
+    expectTree(
+      '[[ a = bc ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (word "bc")) "]]"))`,
+    );
+    expectTree(
+      '[[ $x == b*c ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (extglob_pattern "b*c")) "]]"))`,
+    );
+    expectTree(
+      '[[ $x == 123 ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (number "123")) "]]"))`,
+    );
+  });
+
+  it('parses && / || / ! combinations', () => {
+    expectTree(
+      '[[ -n $s && -d /tmp ]]',
+      `(program (test_command "[[" (binary_expression (unary_expression (test_operator "-n") (simple_expansion "$" (variable_name "s"))) "&&" (unary_expression (test_operator "-d") (word "/tmp"))) "]]"))`,
+    );
+    expectTree(
+      '[[ a < b || ! -e f ]]',
+      `(program (test_command "[[" (binary_expression (binary_expression (word "a") "<" (word "b")) "||" (unary_expression "!" (unary_expression (test_operator "-e") (word "f")))) "]]"))`,
+    );
+  });
+
+  it('parses binary test operators and parentheses', () => {
+    expectTree(
+      '[[ $a -eq 3 ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) (test_operator "-eq") (number "3")) "]]"))`,
+    );
+    expectTree(
+      '[[ $x -eq 1 && $y -ne 2 ]]',
+      `(program (test_command "[[" (binary_expression (binary_expression (simple_expansion "$" (variable_name "x")) (test_operator "-eq") (number "1")) "&&" (binary_expression (simple_expansion "$" (variable_name "y")) (test_operator "-ne") (number "2"))) "]]"))`,
+    );
+    // Inside parentheses the == right side is a word, not extglob_pattern.
+    expectTree(
+      '[[ (a == b) && c ]]',
+      `(program (test_command "[[" (binary_expression (parenthesized_expression "(" (binary_expression (word "a") "==" (word "b")) ")") "&&" (word "c")) "]]"))`,
+    );
+  });
+
+  it('parses the single-bracket form and adjacent =', () => {
+    expectTree('[ -f file ]', `(program (test_command "[" (unary_expression (test_operator "-f") (word "file")) "]"))`);
+    expectTree('[ x ]', `(program (test_command "[" (word "x") "]"))`);
+    expectTree('[[ a=b ]]', `(program (test_command "[[" (word "a=b") "]]"))`);
+  });
+
+  it('parses negated test commands', () => {
+    expectTree(
+      '! [[ -f x ]]',
+      `(program (negated_command "!" (test_command "[[" (unary_expression (test_operator "-f") (word "x")) "]]")))`,
+    );
+    expectTree(
+      '! [ -f x ]',
+      `(program (negated_command "!" (test_command "[" (unary_expression (test_operator "-f") (word "x")) "]")))`,
+    );
+  });
+
+  it('keeps [ in argument position as plain words', () => {
+    expectTree('echo [ x ]', `(program (command (command_name (word "echo")) (word "[") (word "x") (word "]")))`);
+  });
+
+  it('recovers an unterminated test command with a zero-width closer', () => {
+    expectTree('[ -f file', `(program (test_command "[" (unary_expression (test_operator "-f") (word "file")) "]"))`, true);
+  });
+});
+
+describe('arithmetic', () => {
+  it('parses operator precedence levels', () => {
+    expectTree(
+      'echo $((1 + 2 * 3))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (number "1") "+" (binary_expression (number "2") "*" (number "3"))) "))")))`,
+    );
+    expectTree(
+      'echo $((x << 2 | 1))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (binary_expression (variable_name "x") "<<" (number "2")) "|" (number "1")) "))")))`,
+    );
+    expectTree(
+      'echo $((n % 2 == 0))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (binary_expression (variable_name "n") "%" (number "2")) "==" (number "0")) "))")))`,
+    );
+  });
+
+  it('parses ternary, postfix, prefix and unary expressions', () => {
+    expectTree(
+      'echo $((a > b ? a : b))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (ternary_expression (binary_expression (variable_name "a") ">" (variable_name "b")) "?" (variable_name "a") ":" (variable_name "b")) "))")))`,
+    );
+    expectTree(
+      'echo $((i++ + --j))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (postfix_expression (variable_name "i") "++") "+" (unary_expression "--" (variable_name "j"))) "))")))`,
+    );
+    // Prefix operators grab the whole tighter-precedence expression, like
+    // the reference: -(x + ~y).
+    expectTree(
+      'echo $((-x + ~y))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (unary_expression "-" (binary_expression (variable_name "x") "+" (unary_expression "~" (variable_name "y")))) "))")))`,
+    );
+  });
+
+  it('parses parentheses and exponent (left-associative in arithmetic)', () => {
+    expectTree(
+      'echo $(( (1+2) ** 3 ))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (parenthesized_expression "(" (binary_expression (number "1") "+" (number "2")) ")") "**" (number "3")) "))")))`,
+    );
+    expectTree(
+      'echo $((2 ** 3 ** 2))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (binary_expression (number "2") "**" (number "3")) "**" (number "2")) "))")))`,
+    );
+  });
+
+  it('parses assignment operators, subscripts and comma lists', () => {
+    expectTree(
+      'echo $((x += 5))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (variable_name "x") "+=" (number "5")) "))")))`,
+    );
+    expectTree(
+      'echo $((x = y = 1))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (binary_expression (variable_name "x") "=" (variable_name "y")) "=" (number "1")) "))")))`,
+    );
+    expectTree(
+      'echo $((arr[0] + 1))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (subscript (variable_name "arr") "[" (number "0") "]") "+" (number "1")) "))")))`,
+    );
+    expectTree(
+      'echo $((1, 2, 3))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (number "1") "," (number "2") "," (number "3") "))")))`,
+    );
+  });
+
+  it('parses the ((…)) command and legacy $[…] form', () => {
+    expectTree(
+      '((x = y + 1))',
+      `(program (command (command_name (arithmetic_expansion "((" (binary_expression (variable_name "x") "=" (binary_expression (variable_name "y") "+" (number "1"))) "))"))))`,
+    );
+    expectTree(
+      '(( ls ))',
+      `(program (command (command_name (arithmetic_expansion "((" (variable_name "ls") "))"))))`,
+    );
+    expectTree(
+      'echo $[1+2]',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$[" (binary_expression (number "1") "+" (number "2")) "]")))`,
+    );
+  });
+
+  it('parses arithmetic across newlines and nested arithmetic', () => {
+    expectTree(
+      'echo $((\n1 +\n2\n))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (binary_expression (number "1") "+" (number "2")) "))")))`,
+    );
+    expectTree(
+      'for ((i = $((1 + 1)); i < 3; i++)); do x; done',
+      `(program (c_style_for_statement "for" "((" (variable_assignment (variable_name "i") "=" (arithmetic_expansion "$((" (binary_expression (number "1") "+" (number "1")) "))")) ";" (binary_expression (word "i") "<" (number "3")) ";" (postfix_expression (word "i") "++") "))" ";" (do_group "do" (command (command_name (word "x"))) ";" "done")))`,
+    );
+  });
+});
+
+describe('arrays and subscripts', () => {
+  it('parses array assignments', () => {
+    expectTree(
+      'arr=(a b "c d")',
+      `(program (variable_assignment (variable_name "arr") "=" (array "(" (word "a") (word "b") (string "\\"" (string_content "c d") "\\"") ")")))`,
+    );
+    expectTree(
+      'x+=(c d)',
+      `(program (variable_assignment (variable_name "x") "+=" (array "(" (word "c") (word "d") ")")))`,
+    );
+  });
+
+  it('parses subscript assignments', () => {
+    expectTree(
+      'arr[0]=x',
+      `(program (variable_assignment (subscript (variable_name "arr") "[" (number "0") "]") "=" (word "x")))`,
+    );
+    expectTree(
+      'a[i+1]=$y',
+      `(program (variable_assignment (subscript (variable_name "a") "[" (word "i+1") "]") "=" (simple_expansion "$" (variable_name "y"))))`,
+    );
+    expectTree(
+      'x[1]+=2',
+      `(program (variable_assignment (subscript (variable_name "x") "[" (number "1") "]") "+=" (number "2")))`,
+    );
+  });
+
+  it('parses subscript expansions including @ and $ indexes', () => {
+    expectTree(
+      'echo ${arr[@]}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (subscript (variable_name "arr") "[" (word "@") "]") "}")))`,
+    );
+    expectTree(
+      'echo ${a[$i]}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (subscript (variable_name "a") "[" (simple_expansion "$" (variable_name "i")) "]") "}")))`,
+    );
+  });
+});
+
+describe('declaration and unset commands', () => {
+  it('parses export/declare/local/readonly', () => {
+    expectTree(
+      'export FOO=bar BAZ',
+      `(program (declaration_command "export" (variable_assignment (variable_name "FOO") "=" (word "bar")) (variable_name "BAZ")))`,
+    );
+    expectTree(
+      'declare -r x=1',
+      `(program (declaration_command "declare" (word "-r") (variable_assignment (variable_name "x") "=" (number "1"))))`,
+    );
+    expectTree('local y', `(program (declaration_command "local" (variable_name "y")))`);
+    expectTree(
+      'readonly z=2',
+      `(program (declaration_command "readonly" (variable_assignment (variable_name "z") "=" (number "2"))))`,
+    );
+  });
+
+  it('parses unset', () => {
+    expectTree('unset a b', `(program (unset_command "unset" (variable_name "a") (variable_name "b")))`);
+  });
+});
+
+describe('strings and brace expressions', () => {
+  it('parses ANSI-C strings as a single node', () => {
+    expectTree(
+      `echo $'a\\nb\\t'`,
+      `(program (command (command_name (word "echo")) (ansi_c_string "$'a\\\\nb\\\\t'")))`,
+    );
+  });
+
+  it('parses $"…" as an anonymous $ plus a string (no translated_string)', () => {
+    expectTree(
+      'echo $"hello $USER"',
+      `(program (command (command_name (word "echo")) "$" (string "\\"" (string_content "hello ") (simple_expansion "$" (variable_name "USER")) "\\"")))`,
+    );
+  });
+
+  it('parses $"…" as translated_string outside argument position', () => {
+    expectTree(
+      'x=$"v"',
+      `(program (variable_assignment (variable_name "x") "=" (translated_string "$" (string "\\"" (string_content "v") "\\""))))`,
+    );
+    expectTree(
+      'cat > $"out"',
+      `(program (redirected_statement (command (command_name (word "cat"))) (file_redirect ">" (translated_string "$" (string "\\"" (string_content "out") "\\"")))))`,
+    );
+    expectTree(
+      'for f in $"a"; do x; done',
+      `(program (for_statement "for" (variable_name "f") "in" (translated_string "$" (string "\\"" (string_content "a") "\\"")) ";" (do_group "do" (command (command_name (word "x"))) ";" "done")))`,
+    );
+    // Mid-concatenation the bare $ and the string stay separate pieces.
+    expectTree(
+      'echo a$"b"',
+      `(program (command (command_name (word "echo")) (concatenation (word "a") "$" (string "\\"" (string_content "b") "\\""))))`,
+    );
+  });
+
+  it('parses {N..M} brace expressions', () => {
+    expectTree(
+      'echo {1..10}',
+      `(program (command (command_name (word "echo")) (brace_expression "{" (number "1") ".." (number "10") "}")))`,
+    );
+    expectTree(
+      'echo a{1..3}b',
+      `(program (command (command_name (word "echo")) (concatenation (word "a") (brace_expression "{" (number "1") ".." (number "3") "}") (word "b"))))`,
+    );
+    expectTree(
+      'x={1..5}',
+      `(program (variable_assignment (variable_name "x") "=" (brace_expression "{" (number "1") ".." (number "5") "}")))`,
+    );
+    expectTree(
+      '{1..3}',
+      `(program (command (command_name (brace_expression "{" (number "1") ".." (number "3") "}"))))`,
+    );
+  });
+
+  it('parses other brace forms as plain concatenations (like the reference)', () => {
+    expectTree(
+      'echo {1..10..2}',
+      `(program (command (command_name (word "echo")) (concatenation (word "{") (word "1..10..2") (word "}"))))`,
+    );
+    expectTree(
+      'echo {a..z}',
+      `(program (command (command_name (word "echo")) (concatenation (word "{") (word "a..z") (word "}"))))`,
+    );
+    expectTree(
+      'echo {-5..5}',
+      `(program (command (command_name (word "echo")) (concatenation (word "{") (word "-5..5") (word "}"))))`,
+    );
+    expectTree(
+      'echo a{b,c}d',
+      `(program (command (command_name (word "echo")) (concatenation (word "a") (word "{") (word "b,c") (word "}") (word "d"))))`,
+    );
+  });
+});
+
+describe('expansion operators', () => {
+  it('parses removal patterns as regex nodes', () => {
+    expectTree(
+      'echo ${x##*/} ${x%%.*}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "##" (regex "*/") "}") (expansion "\${" (variable_name "x") "%%" (regex ".*") "}")))`,
+    );
+    expectTree(
+      'echo ${x##a b}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "##" (regex "a b") "}")))`,
+    );
+    expectTree(
+      'echo ${x#"pat"}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "#" (string "\\"" (string_content "pat") "\\"") "}")))`,
+    );
+  });
+
+  it('parses replacements with regex pattern and literal replacement', () => {
+    expectTree(
+      'echo ${x/y/z} ${x//a/b}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "/" (regex "y") "/" (word "z") "}") (expansion "\${" (variable_name "x") "//" (regex "a") "/" (word "b") "}")))`,
+    );
+    expectTree(
+      'echo ${x/a b/c}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "/" (regex "a b") "/" (word "c") "}")))`,
+    );
+    expectTree(
+      'echo ${x/a/$v}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "/" (regex "a") "/" (simple_expansion "$" (variable_name "v")) "}")))`,
+    );
+    expectTree(
+      'echo ${x/pat/rep/extra}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "/" (regex "pat") "/" (word "rep/extra") "}")))`,
+    );
+  });
+
+  it('parses case modification and transformation operators', () => {
+    expectTree(
+      'echo ${x^} ${x,,}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "^" "}") (expansion "\${" (variable_name "x") ",," "}")))`,
+    );
+    expectTree(
+      'echo ${x@Q}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") "@" "Q" "}")))`,
+    );
+  });
+
+  it('parses max-length expansions with arithmetic values', () => {
+    expectTree(
+      'echo ${v:1:2}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "v") ":" (number "1") ":" (number "2") "}")))`,
+    );
+    expectTree(
+      'echo ${arr[@]:1:2}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (subscript (variable_name "arr") "[" (word "@") "]") ":" (number "1") ":" (number "2") "}")))`,
+    );
+  });
+
+  it('parses default values as words, splitting bare values with spaces', () => {
+    expectTree(
+      'echo ${v:-1} ${v:=2x}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "v") ":-" (word "1") "}") (expansion "\${" (variable_name "v") ":=" (word "2x") "}")))`,
+    );
+    expectTree(
+      'echo ${x:-d e f}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") ":-" (concatenation (word "d") (word " e f")) "}")))`,
+    );
+    expectTree(
+      'echo ${v:-"d e"}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "v") ":-" (string "\\"" (string_content "d e") "\\"") "}")))`,
+    );
+  });
+
+  it('parses indirect expansions with trailing * and @', () => {
+    expectTree(
+      'echo ${!prefix*} ${!name@}',
+      `(program (command (command_name (word "echo")) (expansion "\${" "!" (variable_name "prefix") "*" "}") (expansion "\${" "!" (variable_name "name") "@" "}")))`,
+    );
+  });
+});
+
+describe('combinations', () => {
+  it('nests if inside for inside case', () => {
+    expectTree(
+      'case $1 in start) for s in a b; do if [[ -n $s ]]; then echo "$s"; fi; done ;; *) echo usage ;; esac',
+      `(program (case_statement "case" (simple_expansion "$" (variable_name "1")) "in" (case_item (word "start") ")" (for_statement "for" (variable_name "s") "in" (word "a") (word "b") ";" (do_group "do" (if_statement "if" (test_command "[[" (unary_expression (test_operator "-n") (simple_expansion "$" (variable_name "s"))) "]]") ";" "then" (command (command_name (word "echo")) (string "\\"" (simple_expansion "$" (variable_name "s")) "\\"")) ";" "fi") ";" "done")) ";;") (case_item (extglob_pattern "*") ")" (command (command_name (word "echo")) (word "usage")) ";;") "esac"))`,
+    );
+  });
+
+  it('parses a heredoc inside a function body', () => {
+    expectTree(
+      'foo() { cat <<EOF\nbody $x\nEOF\n}',
+      `(program (function_definition (word "foo") "(" ")" (compound_statement "{" (redirected_statement (command (command_name (word "cat"))) (heredoc_redirect "<<" (heredoc_start "EOF") (heredoc_body (heredoc_content "body ") (simple_expansion "$" (variable_name "x")) (heredoc_content "\\n")) (heredoc_end "EOF"))) "}")))`,
+    );
+  });
+
+  it('parses compound commands in pipelines and lists', () => {
+    expectTree(
+      'if a; then b; fi | grep x',
+      `(program (pipeline (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" "fi") "|" (command (command_name (word "grep")) (word "x"))))`,
+    );
+    expectTree(
+      'if a; then b; fi && c',
+      `(program (list (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" "fi") "&&" (command (command_name (word "c")))))`,
+    );
+    expectTree(
+      'echo hi | if a; then b; fi',
+      `(program (pipeline (command (command_name (word "echo")) (word "hi")) "|" (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" "fi")))`,
+    );
+  });
+
+  it('parses redirects on compound statements', () => {
+    expectTree(
+      'if a; then b; fi > out',
+      `(program (redirected_statement (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" "fi") (file_redirect ">" (word "out"))))`,
+    );
+  });
+
+  it('parses test commands as loop conditions', () => {
+    expectTree(
+      'while [[ -n $s ]]; do s=; done',
+      `(program (while_statement "while" (test_command "[[" (unary_expression (test_operator "-n") (simple_expansion "$" (variable_name "s"))) "]]") ";" (do_group "do" (variable_assignment (variable_name "s") "=") ";" "done")))`,
+    );
+    expectTree(
+      'if [ -f f ]; then x; fi',
+      `(program (if_statement "if" (test_command "[" (unary_expression (test_operator "-f") (word "f")) "]") ";" "then" (command (command_name (word "x"))) ";" "fi"))`,
+    );
+  });
+
+  it('parses negation, lists and nested ifs in conditions', () => {
+    expectTree(
+      'if ! grep -q x f; then echo missing; fi',
+      `(program (if_statement "if" (negated_command "!" (command (command_name (word "grep")) (word "-q") (word "x") (word "f"))) ";" "then" (command (command_name (word "echo")) (word "missing")) ";" "fi"))`,
+    );
+    expectTree(
+      'if a && b; then c; fi',
+      `(program (if_statement "if" (list (command (command_name (word "a"))) "&&" (command (command_name (word "b")))) ";" "then" (command (command_name (word "c"))) ";" "fi"))`,
+    );
+    expectTree(
+      'if if a; then b; fi; then c; fi',
+      `(program (if_statement "if" (if_statement "if" (command (command_name (word "a"))) ";" "then" (command (command_name (word "b"))) ";" "fi") ";" "then" (command (command_name (word "c"))) ";" "fi"))`,
+    );
+  });
+
+  it('parses arrays and subscript expansions inside functions', () => {
+    expectTree(
+      'f() { arr=(1 2); echo ${arr[0]}; }',
+      `(program (function_definition (word "f") "(" ")" (compound_statement "{" (variable_assignment (variable_name "arr") "=" (array "(" (number "1") (number "2") ")")) ";" (command (command_name (word "echo")) (expansion "\${" (subscript (variable_name "arr") "[" (number "0") "]") "}")) ";" "}")))`,
+    );
+  });
+
+  it('parses a command subshell argument and command substitutions in for values', () => {
+    expectTree(
+      'foo (ls)',
+      `(program (command (command_name (word "foo")) (subshell "(" (command (command_name (word "ls"))) ")")))`,
+    );
+    expectTree(
+      'for i in $(seq 3); do echo $i; done',
+      `(program (for_statement "for" (variable_name "i") "in" (command_substitution "$(" (command (command_name (word "seq")) (number "3")) ")") ";" (do_group "do" (command (command_name (word "echo")) (simple_expansion "$" (variable_name "i"))) ";" "done")))`,
+    );
+  });
+});
+
+describe('recovery and adversarial input', () => {
+  it('recovers unterminated compound commands without throwing', () => {
+    for (const source of [
+      'if a; then b',
+      'if a',
+      'while x; do y',
+      'for f in a b; do x',
+      'case $x in a) x',
+      'case $x',
+      'foo() { x;',
+      '{ x;',
+      'function f',
+      'for ((i=0;i<3',
+      'echo $((1 + ',
+      '[[ -f x',
+      'arr=(1 2',
+    ]) {
+      const result = parse(source);
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(true);
+    }
+  });
+
+  it('recovers stray case terminators at top level', () => {
+    for (const source of ['a ;& b', 'a ;;& b', ';;&']) {
+      const result = parse(source);
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(true);
+    }
+  });
+
+  it('flags an empty subshell', () => {
+    const result = parse('()');
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('handles deeply nested compound commands within the depth cap', () => {
+    const source = 'if a; then '.repeat(600) + 'b' + ' fi'.repeat(600);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('handles deeply nested arithmetic parentheses within the depth cap', () => {
+    const source = 'echo $(( ' + '('.repeat(600) + '1' + ')'.repeat(600) + ' ))';
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('handles deeply nested case statements within the depth cap', () => {
+    const source = 'case x in a) '.repeat(600) + 'y' + ' ;; esac'.repeat(600);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (result.ok) expect(result.hasError).toBe(true);
+  });
+
+  it('parses a pathological test command under budget', () => {
+    const result = parse('[[ ' + 'a && '.repeat(2000) + 'a ]]');
+    // Either parses (with bounded work) or the budget cuts it off; never throws.
+    if (result.ok) {
+      expect(result.rootNode.type).toBe('program');
+    } else {
+      expect(result.reason).toBe('aborted');
+    }
+  });
+});
+
+describe('M2 review regressions', () => {
+  it('keeps a full case_statement inside $() despite pattern parens', () => {
+    expectTree(
+      'x=$(case y in a) 1;; esac)',
+      `(program (variable_assignment (variable_name "x") "=" (command_substitution "$(" (case_statement "case" (word "y") "in" (case_item (word "a") ")" (command (command_name (number "1"))) ";;") "esac") ")")))`,
+    );
+    // Optional-paren item form stays balanced too.
+    expectTree(
+      'x=$(case y in (b) 2;; esac)',
+      `(program (variable_assignment (variable_name "x") "=" (command_substitution "$(" (case_statement "case" (word "y") "in" (case_item "(" (word "b") ")" (command (command_name (number "2"))) ";;") "esac") ")")))`,
+    );
+    expectTree(
+      'diff <(case y in a) 1;; esac) other',
+      `(program (command (command_name (word "diff")) (process_substitution "<(" (case_statement "case" (word "y") "in" (case_item (word "a") ")" (command (command_name (number "1"))) ";;") "esac") ")") (word "other")))`,
+    );
+  });
+
+  it('does not mistake an argument named case for a case_statement in $()', () => {
+    expectTree(
+      '$(echo case; ls)',
+      `(program (command (command_name (command_substitution "$(" (command (command_name (word "echo")) (word "case")) ";" (command (command_name (word "ls"))) ")"))))`,
+    );
+  });
+
+  it('converts test right-hand sides by operator, glob and paren depth', () => {
+    expectTree(
+      '[[ ( $a == x* ) ]]',
+      `(program (test_command "[[" (parenthesized_expression "(" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "x*")) ")") "]]"))`,
+    );
+    expectTree(
+      '[[ ( $a = x* ) ]]',
+      `(program (test_command "[[" (parenthesized_expression "(" (binary_expression (simple_expansion "$" (variable_name "a")) "=" (word "x*")) ")") "]]"))`,
+    );
+    expectTree(
+      '[[ $x == b=c ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (word "b=c")) "]]"))`,
+    );
+    expectTree(
+      '[[ a = b=c ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (regex "b=c")) "]]"))`,
+    );
+    expectTree(
+      '[[ a = =b ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (regex "=b")) "]]"))`,
+    );
+  });
+
+  it('parses statement-position ((word++…)) as a test_command', () => {
+    expectTree('((a++))', `(program (test_command "((" (word "a++") "))"))`);
+    expectTree('((a--))', `(program (test_command "((" (word "a--") "))"))`);
+    expectTree(
+      '((a++ + b))',
+      `(program (test_command "((" (binary_expression (word "a++") "+" (word "b")) "))"))`,
+    );
+    // Other ((…)) forms and non-statement positions stay arithmetic.
+    expectTree(
+      '((b + a++))',
+      `(program (command (command_name (arithmetic_expansion "((" (binary_expression (variable_name "b") "+" (postfix_expression (variable_name "a") "++")) "))"))))`,
+    );
+    expectTree(
+      'echo ((a++))',
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "((" (postfix_expression (variable_name "a") "++") "))")))`,
+    );
+    expectTree(
+      '! ((a++))',
+      `(program (negated_command "!" (test_command "((" (word "a++") "))")))`,
+    );
+  });
+
+  it('keeps unconsumed expression input in ERROR nodes', () => {
+    const arithmetic = parseOk('echo $((x[1][2]))');
+    expect(arithmetic.hasError).toBe(true);
+    expect(sexp(arithmetic.rootNode)).toBe(
+      `(program (command (command_name (word "echo")) (arithmetic_expansion "$((" (subscript (variable_name "x") "[" (number "1") "]") (ERROR "[2]") "))")))`,
+    );
+    const test = parseOk('[[ $a =~ a b ]]');
+    expect(test.hasError).toBe(true);
+    const errors = descendantsOfType(test.rootNode, 'ERROR');
+    expect(errors).toHaveLength(1);
+    expect(errors[0]!.text).toContain('b');
+  });
+
+  it('parses negative offsets in max-length expansions as numbers', () => {
+    expectTree(
+      'echo ${x: -5}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") ":" (number "-5") "}")))`,
+    );
+    expectTree(
+      'echo ${x: -5:2}',
+      `(program (command (command_name (word "echo")) (expansion "\${" (variable_name "x") ":" (number "-5") ":" (number "2") "}")))`,
+    );
+  });
+
+  it('parses $0 and ${0} as special_variable_name', () => {
+    expectTree(
+      'echo $0 ${0} $1 ${10}',
+      `(program (command (command_name (word "echo")) (simple_expansion "$" (special_variable_name "0")) (expansion "\${" (special_variable_name "0") "}") (simple_expansion "$" (variable_name "1")) (expansion "\${" (variable_name "10") "}")))`,
+    );
+  });
+
+  it('degrades locally on deep pattern→string→expansion nesting', () => {
+    let inner = '${a#""}';
+    for (let k = 0; k < 650; k++) inner = `\${a#"${inner}"}`;
+    const result = parse(`echo ${inner}`, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (!result.ok) return;
+    expect(result.hasError).toBe(true);
+    // Local degradation: the tree keeps the outer expansions instead of
+    // collapsing into a single ERROR child under program.
+    expect(result.rootNode.children[0]!.type).not.toBe('ERROR');
+    expect(descendantsOfType(result.rootNode, 'expansion').length).toBeGreaterThan(100);
+  });
+
+  it('treats = and ! as word characters at operand position in tests', () => {
+    expectTree('[[ a!=b ]]', `(program (test_command "[[" (word "a!=b") "]]"))`);
+    expectTree('[[ a=b ]]', `(program (test_command "[[" (word "a=b") "]]"))`);
+    expectTree(
+      '[[ !x = y ]]',
+      `(program (test_command "[[" (binary_expression (word "!x") "=" (word "y")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a==b* ]]',
+      `(program (test_command "[[" (concatenation (simple_expansion "$" (variable_name "a")) (word "==b*")) "]]"))`,
+    );
+    // But they are operators at operator position, even when attached.
+    expectTree(
+      '[[ a ==b ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "==" (extglob_pattern "b")) "]]"))`,
+    );
+    expectTree(
+      '[[ a =b ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (word "b")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a =~^x ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "=~" (regex "^x")) "]]"))`,
+    );
+  });
+
+  it('parses [[ ]] as a concatenation plus a zero-width closer (reference quirk)', () => {
+    expectTree(
+      '[[ ]]',
+      `(program (test_command "[[" (concatenation (word "]") (word "]")) "]]"))`,
+      true,
+    );
+  });
+
+  it('parses bare numbers inside square brackets as number pieces', () => {
+    expectTree(
+      'echo [1] a[2]b',
+      `(program (command (command_name (word "echo")) (concatenation (word "[") (number "1") (word "]")) (concatenation (word "a") (word "[") (number "2") (word "]") (word "b"))))`,
+    );
+    expectTree(
+      'echo a[-5]b',
+      `(program (command (command_name (word "echo")) (concatenation (word "a") (word "[") (number "-5") (word "]") (word "b"))))`,
+    );
+  });
+});
+
+describe('M2 review round 2 regressions', () => {
+  it('applies the scanner.c extglob_pattern rule to ==/!= right sides', () => {
+    // Single letter followed by whitespace → glob; followed by `)` → word.
+    expectTree(
+      '[[ $a == b ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "b")) "]]"))`,
+    );
+    expectTree(
+      '[[ (a == b) && c ]]',
+      `(program (test_command "[[" (binary_expression (parenthesized_expression "(" (binary_expression (word "a") "==" (word "b")) ")") "&&" (word "c")) "]]"))`,
+    );
+    expectTree(
+      '[[ ( $a == x ) ]]',
+      `(program (test_command "[[" (parenthesized_expression "(" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "x")) ")") "]]"))`,
+    );
+    // Multi-character all-letter words stay words, dots don't count.
+    expectTree(
+      '[[ $a == foo ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (word "foo")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a == x.y ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (word "x.y")) "]]"))`,
+    );
+    // Any non-letter non-dot character makes it a glob.
+    for (const word of ['x1', 'foo-', 'x_', 'b*c=d', 'bc=d', 'b?=e', '.x', 'a-b']) {
+      const result = parseOk(`[[ $a == ${word} ]]`);
+      const globs = descendantsOfType(result.rootNode, 'extglob_pattern');
+      expect(globs.map((g) => g.text)).toEqual([word]);
+    }
+    // `=` vetoes only as the second character; a leading `=` vetoes too.
+    expectTree(
+      '[[ $x == b=c ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (word "b=c")) "]]"))`,
+    );
+    expectTree(
+      '[[ a == =b ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "==" (word "=b")) "]]"))`,
+    );
+    // Round-1 fixed behavior must not regress.
+    expectTree(
+      '[[ $x == b*c ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (extglob_pattern "b*c")) "]]"))`,
+    );
+    expectTree(
+      '[[ $x == 123 ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "x")) "==" (number "123")) "]]"))`,
+    );
+    expectTree(
+      '[[ a ==b ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "==" (extglob_pattern "b")) "]]"))`,
+    );
+  });
+
+  it('keeps the = right-side regex rule independent of the glob rule', () => {
+    expectTree(
+      '[[ a = b*c ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (regex "b*c")) "]]"))`,
+    );
+    expectTree(
+      '[[ a = bc ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (word "bc")) "]]"))`,
+    );
+    expectTree(
+      '[[ a = b=c ]]',
+      `(program (test_command "[[" (binary_expression (word "a") "=" (regex "b=c")) "]]"))`,
+    );
+    expectTree(
+      '[[ ( $a = x* ) ]]',
+      `(program (test_command "[[" (parenthesized_expression "(" (binary_expression (simple_expansion "$" (variable_name "a")) "=" (word "x*")) ")") "]]"))`,
+    );
+  });
+
+  it('survives 2500 levels of lexer-side pattern nesting', () => {
+    let inner = '${a#""}';
+    for (let k = 0; k < 2500; k++) inner = `\${a#"${inner}"}`;
+    const result = parse(`echo ${inner}`, { timeoutMs: 30_000 });
+    // No RangeError, no abort: the scan depth cap degrades locally.
+    expect(result.ok).toBe(true);
+    if (!result.ok) return;
+    expect(result.hasError).toBe(true);
+    expect(result.rootNode.children[0]!.type).not.toBe('ERROR');
+  });
+
+  it('parses ((…)) inside a test as an arithmetic_expansion', () => {
+    expectTree(
+      '[[ ((a)) == x ]]',
+      `(program (test_command "[[" (binary_expression (arithmetic_expansion "((" (variable_name "a") "))") "==" (extglob_pattern "x")) "]]"))`,
+    );
+    // Without the closing )), parens still nest as parenthesized_expressions.
+    expectTree(
+      '[[ ((a) == x) && y ]]',
+      `(program (test_command "[[" (binary_expression (parenthesized_expression "(" (binary_expression (parenthesized_expression "(" (word "a") ")") "==" (word "x")) ")") "&&" (word "y")) "]]"))`,
+    );
+  });
+
+  it('parses ((ab-cd++)) as a test_command', () => {
+    expectTree('((ab-cd++))', `(program (test_command "((" (word "ab-cd++") "))"))`);
+  });
+
+  it('flags a missing separator before a compound keyword', () => {
+    for (const source of [
+      'if a; then case x in b) 1;; esac elif c; then d; fi',
+      'if a; then b fi',
+      '{ ls }',
+      'while x do y done',
+      'for f in a b; do x done',
+    ]) {
+      const result = parse(source);
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(true);
+    }
+    // Valid inputs stay clean: case items need no terminator before esac,
+    // subshells none before ), compounds with proper terminators.
+    for (const source of ['if a; then b; fi', '{ ls; }', '(ls)', 'case $x in a) x esac', 'while x; do y; done']) {
+      const result = parse(source);
+      expect(result.ok).toBe(true);
+      if (result.ok) expect(result.hasError).toBe(false);
+    }
+  });
+});
+
+describe('M2 review round 3 regressions', () => {
+  it('converts […]-containing glob right sides as a whole', () => {
+    expectTree(
+      '[[ $ver == [0-9]* ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "ver")) "==" (extglob_pattern "[0-9]*")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a == [abc] ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "[abc]")) "]]"))`,
+    );
+    expectTree(
+      '[[ ( $a == [xyz] ) ]]',
+      `(program (test_command "[[" (parenthesized_expression "(" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "[xyz]")) ")") "]]"))`,
+    );
+    expectTree(
+      '[[ $a == [!a]* ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "[!a]*")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a = [abc] ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "=" (regex "[abc]")) "]]"))`,
+    );
+    // Bracket fragments outside test commands are untouched.
+    expectTree(
+      'echo a[2]b',
+      `(program (command (command_name (word "echo")) (concatenation (word "a") (word "[") (number "2") (word "]") (word "b"))))`,
+    );
+  });
+
+  it('keeps -word as a word when no operand follows it', () => {
+    expectTree(
+      '[[ $a == -foo ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (word "-foo")) "]]"))`,
+    );
+    expectTree(
+      '[[ -foo == x ]]',
+      `(program (test_command "[[" (binary_expression (word "-foo") "==" (extglob_pattern "x")) "]]"))`,
+    );
+    expectTree('[[ -f ]]', `(program (test_command "[[" (word "-f") "]]"))`);
+    // A real operand still makes it a test_operator.
+    expectTree(
+      '[[ -f file.txt ]]',
+      `(program (test_command "[[" (unary_expression (test_operator "-f") (word "file.txt")) "]]"))`,
+    );
+  });
+
+  it('parses extglob group patterns after ==/!=', () => {
+    expectTree(
+      '[[ $a == +(!a) ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "+(!a)")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a == ?(a|b) ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "?(a|b)")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a == !(x) ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "!(x)")) "]]"))`,
+    );
+  });
+
+  it('parses a negative number operand in tests as unary minus', () => {
+    expectTree(
+      '[[ $a == -1 ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (unary_expression "-" (number "1"))) "]]"))`,
+    );
+    expectTree(
+      '[[ $a -eq -1 ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) (test_operator "-eq") (unary_expression "-" (number "1"))) "]]"))`,
+    );
+  });
+
+  it('handles escaped characters in glob right sides', () => {
+    expectTree(
+      '[[ $a == foo\\ bar ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (word "foo\\\\ bar")) "]]"))`,
+    );
+    expectTree(
+      '[[ $a == b\\*c ]]',
+      `(program (test_command "[[" (binary_expression (simple_expansion "$" (variable_name "a")) "==" (extglob_pattern "b\\\\*c")) "]]"))`,
+    );
+  });
+
+  it('extends elif/else clause ranges over the trailing newline', () => {
+    const { rootNode, hasError } = parseOk('if a; then b; elif c; then d\nelse e\nfi');
+    expect(hasError).toBe(false);
+    const elifClause = descendantsOfType(rootNode, 'elif_clause')[0]!;
+    const elseClause = descendantsOfType(rootNode, 'else_clause')[0]!;
+    expect(elifClause.text).toBe('elif c; then d\n');
+    expect(elseClause.text).toBe('else e\n');
+  });
+
+  it('degrades 400-level command substitution nesting locally', () => {
+    const source = 'echo ' + '$('.repeat(400) + 'x' + ')'.repeat(400);
+    const result = parse(source, { timeoutMs: 10_000 });
+    expect(result.ok).toBe(true);
+    if (!result.ok) return;
+    expect(result.hasError).toBe(true);
+    // Local degradation (MAX_SUBSTITUTION_DEPTH), not a last-resort
+    // single-ERROR root.
+    expect(result.rootNode.children[0]!.type).not.toBe('ERROR');
+    expect(descendantsOfType(result.rootNode, 'command_substitution').length).toBeGreaterThan(100);
+  });
+});
diff --git a/packages/tree-sitter-bash/test/performance.test.ts b/packages/tree-sitter-bash/test/performance.test.ts
new file mode 100644
index 0000000000..13b094a346
--- /dev/null
+++ b/packages/tree-sitter-bash/test/performance.test.ts
@@ -0,0 +1,95 @@
+// test/performance.test.ts
+//
+// Performance smoke tests — order-of-magnitude guards only (no absolute
+// timing assertions beyond the documented abort path), meant to catch
+// accidental quadratic complexity in future changes.
+
+import { describe, expect, it } from 'vitest';
+
+import { parse } from '#/parse';
+
+/** ~100 KB of realistic script-shaped text (functions, loops, redirects,
+ *  expansions, heredocs — not a pathological single construct). */
+function realisticScript(): string {
+  const block = `deploy() {
+  local version="$1"
+  if [[ -z "\${version}" ]]; then
+    echo "usage: deploy <version>" >&2
+    return 1
+  fi
+  for pkg in core web worker; do
+    tar -czf "dist/\${pkg}-\${version}.tar.gz" "build/\${pkg}" || return 2
+  done
+  grep -q "release" CHANGELOG.md && echo "releasing \${version}" >> deploy.log
+  ssh deploy@example.com "mkdir -p /srv/app/releases/\${version}" < /dev/null
+  rsync -a --delete dist/ deploy@example.com:/srv/app/releases/\${version}/ | tee -a deploy.log
+}
+
+while read -r name status; do
+  case "$status" in
+    ok) echo "\${name}: fine" ;;
+    *) echo "\${name}: \${status}" >&2 ;;
+  esac
+done < services.txt
+
+cat <<'SCRIPT_EOF'
+deploy finished at $(date)
+SCRIPT_EOF
+
+`;
+  return block.repeat(Math.ceil(100_000 / block.length));
+}
+
+describe('performance smoke', () => {
+  it('parses a 100KB realistic script fast (default budget suffices)', () => {
+    const source = realisticScript();
+    expect(source.length).toBeGreaterThan(100_000);
+    const start = performance.now();
+    const result = parse(source);
+    const elapsed = performance.now() - start;
+    if (!result.ok) {
+      // An abort is acceptable only when it is prompt (< 100 ms).
+      expect(result).toEqual({ ok: false, reason: 'aborted' });
+      expect(elapsed).toBeLessThan(100);
+      return;
+    }
+    // Completed within the default 50 ms budget; re-parse unbounded and
+    // assert the order of magnitude (< 1 s for 100KB) as a quadratic guard.
+    const restart = performance.now();
+    const full = parse(source, { timeoutMs: Number.POSITIVE_INFINITY });
+    const fullElapsed = performance.now() - restart;
+    expect(full.ok).toBe(true);
+    expect(fullElapsed).toBeLessThan(1000);
+    console.log(`100KB realistic script: default-budget parse ${elapsed.toFixed(1)}ms, unbounded ${fullElapsed.toFixed(1)}ms`);
+  });
+
+  it('the abort path is prompt (< 100 ms) on a node-budget bomb', () => {
+    const source = 'echo a; '.repeat(50_000);
+    const start = performance.now();
+    const result = parse(source);
+    const elapsed = performance.now() - start;
+    expect(result).toEqual({ ok: false, reason: 'aborted' });
+    expect(elapsed).toBeLessThan(100);
+    console.log(`abort after ${elapsed.toFixed(1)}ms on a 400KB node-budget bomb`);
+  });
+
+  it('typical one-line commands parse in well under a millisecond', () => {
+    // Warm up.
+    parse('git status && rm -rf /');
+    const iterations = 200;
+    const start = performance.now();
+    for (let i = 0; i < iterations; i++) {
+      parse('git status && rm -rf /');
+    }
+    const perParse = (performance.now() - start) / iterations;
+    expect(perParse).toBeLessThan(5);
+    console.log(`typical command: ${(perParse * 1000).toFixed(0)}µs per parse`);
+  });
+
+  it('a 500KB heredoc body parses within the default budget (few nodes)', () => {
+    const source = `cat <<EOF\n${'line of text\n'.repeat(40_000)}EOF`;
+    expect(source.length).toBeGreaterThan(500_000);
+    const result = parse(source);
+    expect(result.ok).toBe(true);
+  });
+});
diff --git a/packages/tree-sitter-bash/tsconfig.json b/packages/tree-sitter-bash/tsconfig.json
new file mode 100644
index 0000000000..46f71c10fe
--- /dev/null
+++ b/packages/tree-sitter-bash/tsconfig.json
@@ -0,0 +1,5 @@
+{
+  "extends": "../../tsconfig.json",
+  "compilerOptions": {},
+  "include": ["src", "test"]
+}
diff --git a/packages/tree-sitter-bash/tsdown.config.ts b/packages/tree-sitter-bash/tsdown.config.ts
new file mode 100644
index 0000000000..b27e00afd7
--- /dev/null
+++ b/packages/tree-sitter-bash/tsdown.config.ts
@@ -0,0 +1,13 @@
+import { defineConfig } from 'tsdown';
+
+export default defineConfig({
+  entry: ['./src/index.ts'],
+  format: ['esm'],
+  dts: false,
+  outDir: 'dist',
+  clean: true,
+  deps: {
+    alwaysBundle: [/^@moonshot-ai\//],
+    neverBundle: [],
+  },
+});
diff --git a/packages/tree-sitter-bash/vitest.config.ts b/packages/tree-sitter-bash/vitest.config.ts
new file mode 100644
index 0000000000..85c8b889e8
--- /dev/null
+++ b/packages/tree-sitter-bash/vitest.config.ts
@@ -0,0 +1,8 @@
+import { defineConfig } from 'vitest/config';
+
+export default defineConfig({
+  test: {
+    name: 'tree-sitter-bash',
+    include: ['test/**/*.test.ts'],
+  },
+});
diff --git a/pnpm-lock.yaml b/pnpm-lock.yaml
index 57ea9f82e7..b0cc621a24 100644
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -692,6 +692,9 @@ importers:
       '@moonshot-ai/protocol':
         specifier: workspace:^
         version: link:../protocol
+      '@moonshot-ai/tree-sitter-bash':
+        specifier: workspace:^
+        version: link:../tree-sitter-bash
       '@mozilla/readability':
         specifier: ^0.6.0
         version: 0.6.0
@@ -1018,6 +1021,15 @@ importers:
         specifier: 'catalog:'
         version: 4.3.6
 
+  packages/tree-sitter-bash:
+    devDependencies:
+      tree-sitter-bash:
+        specifier: 0.25.0
+        version: 0.25.0
+      web-tree-sitter:
+        specifier: 0.25.10
+        version: 0.25.10
+
 packages:
 
   '@agentclientprotocol/sdk@0.23.0':
@@ -7666,6 +7678,10 @@ packages:
   node-addon-api@7.1.1:
     resolution: {integrity: sha512-5m3bsyrjFWE1xf7nz7YXdN4udnVtXK6/Yfgn5qnahL6bCkf2yKt4k3nuTKAtT4r3IG8JNR2ncsIMdZuAzJjHQQ==}
 
+  node-addon-api@8.9.0:
+    resolution: {integrity: sha512-ekZMeaaIzSQTSpr7X2X3iJM7lTzgnx8ahAG9pJfT/7+14mlEM8ZYQ9cgCDvSSRbReFK0oHli3WrZdCiRsgAT9Q==}
+    engines: {node: ^18 || ^20 || >= 21}
+
   node-domexception@1.0.0:
     resolution: {integrity: sha512-/jKZoMpw0F8GRwl4/eLROPA3cfcXtLApP0QzLmUT/HuPCZWyB7IY9ZrMeKw2O/nFIqPQB3PVM9aYm0F312AXDQ==}
     engines: {node: '>=10.5.0'}
@@ -7688,6 +7704,10 @@ packages:
     resolution: {integrity: sha512-dRB78srN/l6gqWulah9SrxeYnxeddIG30+GOqK/9OlLVyLg3HPnr6SqOWTWOXKRwC2eGYCkZ59NNuSgvSrpgOA==}
     engines: {node: ^12.20.0 || ^14.13.1 || >=16.0.0}
 
+  node-gyp-build@4.8.4:
+    resolution: {integrity: sha512-LA4ZjwlnUblHVgq0oBF3Jl/6h/Nvs5fzBLwdEF4nuxnFdsfajde4WfxtJr3CaiH+F6ewcIB/q4jQ4UzPyid+CQ==}
+    hasBin: true
+
   node-pty@1.1.0:
     resolution: {integrity: sha512-20JqtutY6JPXTUnL0ij1uad7Qe1baT46lyolh2sSENDd4sTzKZ4nmAFkeAARDKwmlLjPx6XKRlwRUxwjOy+lUg==}
 
@@ -9155,6 +9175,14 @@ packages:
     resolution: {integrity: sha512-L0Orpi8qGpRG//Nd+H90vFB+3iHnue1zSSGmNOOCh1GLJ7rUKVwV2HvijphGQS2UmhUZewS9VgvxYIdgr+fG1A==}
     hasBin: true
 
+  tree-sitter-bash@0.25.0:
+    resolution: {integrity: sha512-gZtlj9+qFS81qKxpLfD6H0UssQ3QBc/F0nKkPsiFDyfQF2YBqYvglFJUzchrPpVhZe9kLZTrJ9n2J6lmka69Vg==}
+    peerDependencies:
+      tree-sitter: ^0.25.0
+    peerDependenciesMeta:
+      tree-sitter:
+        optional: true
+
   trim-lines@3.0.1:
     resolution: {integrity: sha512-kRj8B+YHZCc9kQYdWfJB2/oUl9rA99qbowYYBtr4ui4mZyAQ2JpvVBd/6U2YloATfqBhBTSMhTpgBHtU0Mf3Rg==}
 
@@ -9714,6 +9742,14 @@ packages:
     resolution: {integrity: sha512-d2JWLCivmZYTSIoge9MsgFCZrt571BikcWGYkjC1khllbTeDlGqZ2D8vD8E/lJa8WGWbb7Plm8/XJYV7IJHZZw==}
     engines: {node: '>= 8'}
 
+  web-tree-sitter@0.25.10:
+    resolution: {integrity: sha512-Y09sF44/13XvgVKgO2cNDw5rGk6s26MgoZPXLESvMXeefBf7i6/73eFurre0IsTW6E14Y0ArIzhUMmjoc7xyzA==}
+    peerDependencies:
+      '@types/emscripten': ^1.40.0
+    peerDependenciesMeta:
+      '@types/emscripten':
+        optional: true
+
   webidl-conversions@3.0.1:
     resolution: {integrity: sha512-2JAn3z8AR6rjK8Sm8orRC0h/bcl/DqL7tRPdGZ4I1CjdF+EaMLmYxBHyXuKL849eucPFhvBoxMsflfOb8kxaeQ==}
 
@@ -17187,6 +17223,8 @@ snapshots:
 
   node-addon-api@7.1.1: {}
 
+  node-addon-api@8.9.0: {}
+
   node-domexception@1.0.0: {}
 
   node-emoji@2.2.0:
@@ -17208,6 +17246,8 @@ snapshots:
       fetch-blob: 3.2.0
       formdata-polyfill: 4.0.10
 
+  node-gyp-build@4.8.4: {}
+
   node-pty@1.1.0:
     dependencies:
       node-addon-api: 7.1.1
@@ -19004,6 +19044,11 @@ snapshots:
 
   tree-kill@1.2.2: {}
 
+  tree-sitter-bash@0.25.0:
+    dependencies:
+      node-addon-api: 8.9.0
+      node-gyp-build: 4.8.4
+
   trim-lines@3.0.1: {}
 
   trough@2.2.0: {}
@@ -19547,6 +19592,8 @@ snapshots:
 
   web-streams-polyfill@3.3.3: {}
 
+  web-tree-sitter@0.25.10: {}
+
   webidl-conversions@3.0.1: {}
 
   webidl-conversions@7.0.0: