/ca-chore
Sanctioned lane for non-behavioral work — docs-only edits, dependency bumps, reverts. Type-scaled gates; no TDD demanded of prose.
$ npx -y skills add arbiterForge/codeArbiter --skill ca-chore --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/ca-chore
Context preview
The summary Claude sees to decide when to auto-load this skill.
Sanctioned lane for non-behavioral work — docs-only edits, dependency bumps, reverts. Type-scaled gates; no TDD demanded of prose.
SKILL.md
ca-chore.SKILL.mdname: ca-chore
description: Sanctioned lane for non-behavioral work — docs-only edits, dependency bumps, reverts. Type-scaled gates; no TDD demanded of prose.
argument-hint: "<docs|deps|revert> <description or SHA>"
$ca-chore — non-behavioral change lane
The sanctioned path for work that has no behavioral surface to test-drive. Three types, each with the gates that fit it — and nothing it doesn't need. Everything still exits through `commit-gate` (classification `docs` / `chore` / `revert`) and lands via branch/PR.
Types
**docs** — README, comments, CHANGELOG, `.codearbiter/` prose, typo fixes.
- Gates: secrets scan over the diff; diff review (no behavioral code change smuggled in);
anti-slop copy pass on any user-facing doc in the change; `commit-gate`.
- **Anti-slop copy pass** — for any user-facing doc the change touches (repo-root community docs
and `docs/**`; not codeArbiter's own framework bodies), load `${CLAUDE_PLUGIN_ROOT}/includes/anti-slop-design/INDEX.md`, then `core.md` and the `medium-documents` leaf (§7.A.1), and run the §3.A em-dash ban and the §3.B copy self-audit over the authored prose before `commit-gate`. The `H-13` PostToolUse reminder surfaces §3.A separator dashes as you write, so the pass is a confirmation, not a discovery.
- No `tdd` — prose has no failing test to write.
**deps** — bump an existing dependency's version (manifest + lockfile together, never one without the other).
- Gates: dispatch `dependency-reviewer` (same vetting as `$ca-add-dep` — license, provenance,
supply chain, changelog of the bump); full test suite green after the bump; `commit-gate`.
- A bump that requires code changes to adopt is not a chore — route the code change through
`$ca-feature` or `$ca-fix`.
**revert** — back out a named commit.
- Gates: `git revert <SHA>` (never hand-edited backout); full suite green after the revert; the
commit message references the reverted SHA and the reason; `commit-gate`.
- No new regression test demanded — the revert restores already-tested behavior. If the revert is
fixing a defect the suite missed, open `$ca-fix` afterward to pin it.
Flow
1. Classify `$ARGUMENTS` into one of the three types. Anything with a behavioral code change beyond a mechanical revert is NOT a chore — redirect to `$ca-feature` or `$ca-fix` and stop. 2. Run the type's gates above. 3. Exit through `commit-gate` with the matching classification, then `finishing-a-development-branch` as usual.
Hard gate
MUST reject a change containing behavioral code (beyond the revert itself) — that is `$ca-feature` or `$ca-fix` territory. MUST keep manifest and lockfile changes in the same commit for `deps`. MUST run the full suite for `deps` and `revert`. MUST run the anti-slop copy pass on any user-facing doc a `docs` change authors or edits. Never skips `commit-gate`.
When NOT to use
- Any new or changed behavior → `$ca-feature` / `$ca-fix`.
- Adding a NEW dependency → `$ca-add-dep`.
- Restructuring code → `$ca-refactor`.
Read more
name: ca-chore description: Sanctioned lane for non-behavioral work — docs-only edits, dependency bumps, reverts. Type-scaled gates; no TDD demanded of prose. argument-hint: "<docs|deps|revert> <description or SHA>"
$ca-chore — non-behavioral change lane
The sanctioned path for work that has no behavioral surface to test-drive. Three types, each with the gates that fit it — and nothing it doesn't need. Everything still exits through `commit-gate` (classification `docs` / `chore` / `revert`) and lands via branch/PR.
Types
**docs** — README, comments, CHANGELOG, `.codearbiter/` prose, typo fixes.
- Gates: secrets scan over the diff; diff review (no behavioral code change smuggled in);
anti-slop copy pass on any user-facing doc in the change; `commit-gate`.
- **Anti-slop copy pass** — for any user-facing doc the change touches (repo-root community docs
and `docs/**`; not codeArbiter's own framework bodies), load `${CLAUDE_PLUGIN_ROOT}/includes/anti-slop-design/INDEX.md`, then `core.md` and the `medium-documents` leaf (§7.A.1), and run the §3.A em-dash ban and the §3.B copy self-audit over the authored prose before `commit-gate`. The `H-13` PostToolUse reminder surfaces §3.A separator dashes as you write, so the pass is a confirmation, not a discovery.
- No `tdd` — prose has no failing test to write.
**deps** — bump an existing dependency's version (manifest + lockfile together, never one without the other).
- Gates: dispatch `dependency-reviewer` (same vetting as `$ca-add-dep` — license, provenance,
supply chain, changelog of the bump); full test suite green after the bump; `commit-gate`.
- A bump that requires code changes to adopt is not a chore — route the code change through
`$ca-feature` or `$ca-fix`.
**revert** — back out a named commit.
- Gates: `git revert <SHA>` (never hand-edited backout); full suite green after the revert; the
commit message references the reverted SHA and the reason; `commit-gate`.
- No new regression test demanded — the revert restores already-tested behavior. If the revert is
fixing a defect the suite missed, open `$ca-fix` afterward to pin it.
Flow
1. Classify `$ARGUMENTS` into one of the three types. Anything with a behavioral code change beyond a mechanical revert is NOT a chore — redirect to `$ca-feature` or `$ca-fix` and stop. 2. Run the type's gates above. 3. Exit through `commit-gate` with the matching classification, then `finishing-a-development-branch` as usual.
Hard gate
MUST reject a change containing behavioral code (beyond the revert itself) — that is `$ca-feature` or `$ca-fix` territory. MUST keep manifest and lockfile changes in the same commit for `deps`. MUST run the full suite for `deps` and `revert`. MUST run the anti-slop copy pass on any user-facing doc a `docs` change authors or edits. Never skips `commit-gate`.
When NOT to use
- Any new or changed behavior → `$ca-feature` / `$ca-fix`.
- Adding a NEW dependency → `$ca-add-dep`.
- Restructuring code → `$ca-refactor`.
When you can't trust yourself with your code base, trust Arbiter.
Repo: arbiterForge/codeArbiter
Other skills on codearbiter.
- /brainstorming
The Socratic spec-refinement front of /feature, and the planning front of /sprint. Routed to BEFORE any code — it takes a one-line idea and drives it to an approved, concrete spec with testable acceptance criteria. Four gated phases — frame, refine, write, approve. No
Open skill - /commit-gate
The only path to a commit. Routed to when the user invokes /commit or otherwise instructs codeArbiter to persist staged changes. Nine gated phases — permission, branch, classification, verification (test/lint/secrets), behavioral proof, diff review, selective stage, message,
Open skill - /context-check
Optional manual drift audit — report stale provenance-tracked docs (via _provenancelib drift detection across .codearbiter/.provenance/), then per stale doc offer re-scout / re-baseline / defer. Not the daily loop; commit-gate auto-heal owns routine maintenance.
Open skill - /context-creation
The brownfield back-fill. Routed to by /create-context, and by startup when .codearbiter/CONTEXT.md lacks the <!--INITIALIZED--> body marker but source code exists. Six gated phases — pre-flight, scout dispatch, synthesis, gap interview, write, lock. Reads the existing codebase
Open skill - /crypto-compliance
The banned-primitive gate. Routed to when changed code hashes, signs, encrypts, derives keys, generates security-relevant randomness, configures TLS, or imports a crypto library. Rejects broken primitives, disabled TLS verification, and home-rolled crypto; the approved-primitive
Open skill - /debug
Investigate-then-decide root-cause analysis for a defect whose cause is unknown (distinct from /fix, which assumes a known bug). Five gated phases: capture, hypothesize, gather, decide, hand off. Investigation only, no code changes; exits to /fix, /adr, or a no-action close.
Open skill

