[antonwolfy]

Summary

Follow-up to three reviews. Each left a point open that a mutant shows on main:

• test(cli): flag registry, ratchet contract test, CLI snapshot (#94 F1a) #254 (review, body B1, item 3): dokimi.cli_flags builds two module-level tables at import, CANONICAL, BANNED = _index(), and exports both in __all__. Nothing in the package, the scripts, the tests or the docs reads them. The flag contract test builds its own index from CONCEPTS in violations().
• feat(context, repo): consistency of repeated runs; pass^k and path stability in repo compare #155 (review, body B3, item 1): perl -pi is pinned only where written() parses the in-place flag. No test has an Edit matcher meet a perl -pi Bash step, so writes_file can stop listing perl without any test noticing.
• fix(cli): run, compare, arms, tasks, explore and repo exit with a message on bad input or output #238 (thread and reply): the accepted costs added in reply are _SAFE_MAGNITUDE and -_SAFE_MAGNITUDE, so they move with the constant. Nothing accepts a fixed large value.

Mutant on main	Tests run	Result on main
_index() returns {}, {}	tests/test_cli_flags.py	37 passed
perl removed from _SHELL_WRITERS	context tests	443 passed
_SAFE_MAGNITUDE = 1e6	tests/test_repo_*.py, tests/test_run_compare_errors.py	all pass

Changes

• dokimi/dokimi/cli_flags.py: removes CANONICAL, BANNED, the _index() helper that built them (it had no other caller) and their two __all__ entries.
• tests/test_context_consistency.py: new test test_perl_in_place_is_an_edit_of_its_file, next to the sed -i / tee matcher cases. A contract whose edit action is {tool: Edit, file_path: src/a\.py$} is satisfied by the Bash step perl -pi -e 's/a/b/' src/a.py. perl -Mstrict -e 'print 1' src/a.py is not an edit (requires: edit).
• tests/test_repo_report_unreadable_cell.py: _NUMBERS_ACCEPTED gains a fixed cost of 1e100, far above ordinary costs and well below SCORE_BOUND (1e150). test_a_cost_at_the_bound_is_accepted now runs it through repo report --json. These accepted cases are not shared with the compare side, which keeps its own test at the bound.
• Changelog fragment (Removed).

Risk

Code that imported dokimi.cli_flags.CANONICAL or BANNED now gets an ImportError. Nothing in this repo does. No command, flag, default or output changes. The other two changes only add tests.

Testing

On a Linux review host, at 9b15398 (on main e97ff7a):

• make ci-local: 28 passed, 2 failed, 7 skipped. The two FAILs are ci.yml:static / Install gitleaks (curl: (22) The requested URL returned error: 504, four times) and ci.yml:static / Scan commits, which then found no gitleaks binary (exit 127). The gitleaks download could not be fetched, so the secret scan was not run here. Skipped: external-catalog-smoke (needs docker + network), test-latest, docs.yml:deploy and pip-audit.yml:report (job condition false on a PR), cli-surface / Package the result, conda-forge-deps (micromamba not on PATH), and semgrep secrets (offline).
  10653 tests passed, 4 skipped. Changed-line coverage: 0 measurable package lines changed (the package change is a deletion).
• The touched test files plus test_cli_flags.py, test_context_shell.py and the guard tests (test_docs_code_refs.py, test_growth_ratchet.py, test_audit_test_quality.py, test_import_architecture.py): 745 passed. ruff check on the changed files and mypy on the whole package are clean.
• Mutants. Each new case fails on the change it pins and passes without it:
  • perl removed from _SHELL_WRITERS: test_perl_in_place_is_an_edit_of_its_file fails (1 failed, 92 passed).
  • The perl in-place pattern loosened to any option (^-): the same test fails on its perl -Mstrict half.
  • _SAFE_MAGNITUDE = 1e6: test_a_cost_at_the_bound_is_accepted[large-but-within-the-bound] fails (1 failed, 155 passed). The two cases built from the constant still pass, which is the gap the review named.
• Merged together with the other follow-up branch (fix/repo-run-unreadable-result) on main e97ff7a: pytest -n auto 10656 passed, 4 skipped; mypy and ruff check . clean.

Revert

git revert the commit. It restores the two tables and drops the two test cases. Nothing is stored or versioned.

Checklist

• Tests added/updated and pytest passes
• ruff, mypy, and schema checks pass
• Docs updated (or N/A) — see the docs-drift checklist in CONTRIBUTING.md
• Self-reviewed per "Before opening a PR" in AGENTS.md
• No ad-hoc run artifacts committed
• New files carry the SPDX license header (or are exempt in .github/.licenserc.yaml)

[Copilot]

🟢 Approval recommended

No active review issues remain for the unused CLI table cleanup and regression-test additions.

No open issues 🎉

What changed in this PR

0 new findings; nothing to fix.

This follow-up to #254, #155 and #238 removes unused CLI registry tables and strengthens regression tests without changing command behavior.

Changes:

• Removes CANONICAL, BANNED and their unused _index() helper.
• Tests that perl -pi satisfies an Edit contract while a non-editing Perl command does not.
• Adds a fixed 1e100 cost case to guard against accidentally lowering the accepted magnitude.

File	Description
dokimi/​tests/​test_repo_report_unreadable_cell.py	Adds a fixed large-cost acceptance case.
dokimi/​tests/​test_context_consistency.py	Tests Perl in-place editing against an Edit contract.
dokimi/​dokimi/​cli_flags.py	Removes unused lookup tables, exports and helper.
dokimi/​changelog.d/​20261030_120000_antonwolfy_cli-flags-unused-tables.md	Documents removal of the exported tables.

_{Review effort: Balanced}

[exai-bot]

✅ OpenCodeReview: Review complete: 0 finding(s) across 4 selected item(s).

Review details

• Commit reviewed: 9b15398 against merge base e97ff7a
• Coverage: 4 files in 1 groups
• Engine: OpenCodeReview v1.13.3, us.openai.gpt-6.1-sol
• Usage: 84k tokens (58k from cache), 15s
• Tool calls: 11 (7 file_read, 4 code_search)
• Remove unused CLI table flags and update consistency tests: 4 files

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.55%. Comparing base (e97ff7a) to head (9b15398).
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #303   +/-   ##
=======================================
  Coverage   93.54%   93.55%           
=======================================
  Files         275      275           
  Lines       35701    35691   -10     
  Branches     6053     6050    -3     
=======================================
- Hits        33398    33392    -6     
+ Misses       1486     1483    -3     
+ Partials      817      816    -1     

Files with missing lines	Coverage Δ
dokimi/dokimi/cli_flags.py	100.00% <ø> (ø)

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.