Skip to content

pre-commit-review: cluster title line carries its rule ids #92

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
buzzdan merged 2 commits into from
Oct 18, 2026
Merged

pre-commit-review: cluster title line carries its rule ids #92

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a5364f8
pre-commit-review: cluster title line carries its rule ids
buzzdan Oct 14, 2026
0972a3c
Merge remote-tracking branch 'origin/main' into claude/cluster-header…
buzzdan Oct 18, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 8 additions & 6 deletions core/skills/pre-commit-review/reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ primitive, R11's the duplicated switch, R2's the ownerless validation: one disea
four jurisdictions). Render each cluster as a first-class entry above the categories:

```
🔗 CLUSTER: Alert.Channel
🔗 CLUSTER: Alert.Channel — R1, R11, R2, R7
Convergence: 4 findings — R1, R11, R2, R7
Hypothesis: missing domain concept — a Channel type wants to exist
Skeptic: CONFIRMED (score 6: switched in 3 files, +2 unrepresentable — the
Expand All @@ -170,12 +170,14 @@ script excluded renders as its heading and the script's reason. Clusters never
cross a group: a Go type and a Python class are different anchors. `BROADER CONTEXT`
and the closing `Caller decides` line come once, after the last group.

Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title is the
Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor> — <rules>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title names the
anchor itself — the type, field, function or package name the findings share
(`🔗 CLUSTER: ProcessHeartbeat`, `🔗 CLUSTER: Catalog.Find`,
`🔗 CLUSTER: internal/common, internal/utils`), never a description of the problem;
the description is the Hypothesis line beneath it. When the skeptic
(`🔗 CLUSTER: ProcessHeartbeat — R3, R11, R12`, `🔗 CLUSTER: Catalog.Find — R1, R2`,
`🔗 CLUSTER: internal/common, internal/utils — R4, R5`), never a description of the problem;
the description is the Hypothesis line beneath it. The rules after the dash are the ids
of the rules that converged on the anchor, on the title line itself, never on the
lines below. When the skeptic
reviewed an extraction at that anchor, the entry carries its verdict — a CONFIRMED
type routes design-first as above; a REFUTED one keeps the cluster (the convergence
is still real) and routes to the cheaper alternative, which ships as 🟢 Polish. When no
Expand Down
14 changes: 8 additions & 6 deletions go-linter-driven-development/skills/pre-commit-review/reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ primitive, R11's the duplicated switch, R2's the ownerless validation: one disea
four jurisdictions). Render each cluster as a first-class entry above the categories:

```
🔗 CLUSTER: Alert.Channel
🔗 CLUSTER: Alert.Channel — R1, R11, R2, R7
Convergence: 4 findings — R1, R11, R2, R7
Hypothesis: missing domain concept — a Channel type wants to exist
Skeptic: CONFIRMED (score 6: switched in 3 files, +2 unrepresentable — the
Expand All @@ -170,12 +170,14 @@ script excluded renders as its heading and the script's reason. Clusters never
cross a group: a Go type and a Python class are different anchors. `BROADER CONTEXT`
and the closing `Caller decides` line come once, after the last group.

Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title is the
Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor> — <rules>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title names the
anchor itself — the type, field, function or package name the findings share
(`🔗 CLUSTER: ProcessHeartbeat`, `🔗 CLUSTER: Catalog.Find`,
`🔗 CLUSTER: internal/common, internal/utils`), never a description of the problem;
the description is the Hypothesis line beneath it. When the skeptic
(`🔗 CLUSTER: ProcessHeartbeat — R3, R11, R12`, `🔗 CLUSTER: Catalog.Find — R1, R2`,
`🔗 CLUSTER: internal/common, internal/utils — R4, R5`), never a description of the problem;
the description is the Hypothesis line beneath it. The rules after the dash are the ids
of the rules that converged on the anchor, on the title line itself, never on the
lines below. When the skeptic
reviewed an extraction at that anchor, the entry carries its verdict — a CONFIRMED
type routes design-first as above; a REFUTED one keeps the cluster (the convergence
is still real) and routes to the cheaper alternative, which ships as 🟢 Polish. When no
Expand Down
14 changes: 8 additions & 6 deletions linter-driven-development/skills/pre-commit-review/reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ primitive, R11's the duplicated switch, R2's the ownerless validation: one disea
four jurisdictions). Render each cluster as a first-class entry above the categories:

```
🔗 CLUSTER: Alert.Channel
🔗 CLUSTER: Alert.Channel — R1, R11, R2, R7
Convergence: 4 findings — R1, R11, R2, R7
Hypothesis: missing domain concept — a Channel type wants to exist
Skeptic: CONFIRMED (score 6: switched in 3 files, +2 unrepresentable — the
Expand All @@ -170,12 +170,14 @@ script excluded renders as its heading and the script's reason. Clusters never
cross a group: a Go type and a Python class are different anchors. `BROADER CONTEXT`
and the closing `Caller decides` line come once, after the last group.

Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title is the
Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor> — <rules>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title names the
anchor itself — the type, field, function or package name the findings share
(`🔗 CLUSTER: ProcessHeartbeat`, `🔗 CLUSTER: Catalog.Find`,
`🔗 CLUSTER: internal/common, internal/utils`), never a description of the problem;
the description is the Hypothesis line beneath it. When the skeptic
(`🔗 CLUSTER: ProcessHeartbeat — R3, R11, R12`, `🔗 CLUSTER: Catalog.Find — R1, R2`,
`🔗 CLUSTER: internal/common, internal/utils — R4, R5`), never a description of the problem;
the description is the Hypothesis line beneath it. The rules after the dash are the ids
of the rules that converged on the anchor, on the title line itself, never on the
lines below. When the skeptic
reviewed an extraction at that anchor, the entry carries its verdict — a CONFIRMED
type routes design-first as above; a REFUTED one keeps the cluster (the convergence
is still real) and routes to the cheaper alternative, which ships as 🟢 Polish. When no
Expand Down
14 changes: 8 additions & 6 deletions python-linter-driven-development/skills/pre-commit-review/reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ primitive, R11's the duplicated switch, R2's the ownerless validation: one disea
four jurisdictions). Render each cluster as a first-class entry above the categories:

```
🔗 CLUSTER: Alert.Channel
🔗 CLUSTER: Alert.Channel — R1, R11, R2, R7
Convergence: 4 findings — R1, R11, R2, R7
Hypothesis: missing domain concept — a Channel type wants to exist
Skeptic: CONFIRMED (score 6: switched in 3 files, +2 unrepresentable — the
Expand All @@ -170,12 +170,14 @@ script excluded renders as its heading and the script's reason. Clusters never
cross a group: a Go type and a Python class are different anchors. `BROADER CONTEXT`
and the closing `Caller decides` line come once, after the last group.

Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title is the
Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor> — <rules>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title names the
anchor itself — the type, field, function or package name the findings share
(`🔗 CLUSTER: ProcessHeartbeat`, `🔗 CLUSTER: Catalog.Find`,
`🔗 CLUSTER: internal/common, internal/utils`), never a description of the problem;
the description is the Hypothesis line beneath it. When the skeptic
(`🔗 CLUSTER: ProcessHeartbeat — R3, R11, R12`, `🔗 CLUSTER: Catalog.Find — R1, R2`,
`🔗 CLUSTER: internal/common, internal/utils — R4, R5`), never a description of the problem;
the description is the Hypothesis line beneath it. The rules after the dash are the ids
of the rules that converged on the anchor, on the title line itself, never on the
lines below. When the skeptic
reviewed an extraction at that anchor, the entry carries its verdict — a CONFIRMED
type routes design-first as above; a REFUTED one keeps the cluster (the convergence
is still real) and routes to the cheaper alternative, which ships as 🟢 Polish. When no
Expand Down
14 changes: 8 additions & 6 deletions ts-react-linter-driven-development/skills/pre-commit-review/reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ primitive, R11's the duplicated switch, R2's the ownerless validation: one disea
four jurisdictions). Render each cluster as a first-class entry above the categories:

```
🔗 CLUSTER: Alert.Channel
🔗 CLUSTER: Alert.Channel — R1, R11, R2, R7
Convergence: 4 findings — R1, R11, R2, R7
Hypothesis: missing domain concept — a Channel type wants to exist
Skeptic: CONFIRMED (score 6: switched in 3 files, +2 unrepresentable — the
Expand All @@ -170,12 +170,14 @@ script excluded renders as its heading and the script's reason. Clusters never
cross a group: a Go type and a Python class are different anchors. `BROADER CONTEXT`
and the closing `Caller decides` line come once, after the last group.

Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title is the
Render *every* cluster the pass finds, one `🔗 CLUSTER: <anchor> — <rules>` line per converged
anchor: two findings or twenty, the largest and the smallest alike. The title names the
anchor itself — the type, field, function or package name the findings share
(`🔗 CLUSTER: ProcessHeartbeat`, `🔗 CLUSTER: Catalog.Find`,
`🔗 CLUSTER: internal/common, internal/utils`), never a description of the problem;
the description is the Hypothesis line beneath it. When the skeptic
(`🔗 CLUSTER: ProcessHeartbeat — R3, R11, R12`, `🔗 CLUSTER: Catalog.Find — R1, R2`,
`🔗 CLUSTER: internal/common, internal/utils — R4, R5`), never a description of the problem;
the description is the Hypothesis line beneath it. The rules after the dash are the ids
of the rules that converged on the anchor, on the title line itself, never on the
lines below. When the skeptic
reviewed an extraction at that anchor, the entry carries its verdict — a CONFIRMED
type routes design-first as above; a REFUTED one keeps the cluster (the convergence
is still real) and routes to the cheaper alternative, which ships as 🟢 Polish. When no
Expand Down
Loading