patterns/ai-reviewed-commit/skills/ai-reviewed-commit/SKILL.md
Use when: committing with pre-commit review, authorship-aware fix-retry
npx skillsauth add ryderfreeman4logos/cli-sub-agent ai-reviewed-commitInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Check your initial prompt. If it contains the literal string "Use the ai-reviewed-commit skill", then:
YOU ARE THE EXECUTOR. Follow these rules:
../../PATTERN.md relative to this SKILL.md, and follow it step by step.csa run, csa review, csa debate, or ANY csa command. You must perform the work DIRECTLY. Running any csa command causes infinite recursion.Only if you are the main agent (Claude Code / human user):
Ensure every commit is reviewed by an independent model before creation. Uses authorship-aware strategy: self-authored code gets adversarial debate review (csa debate), while other-authored code gets standard csa review --diff --allow-fallback. The compiled weave workflow is a single-pass mechanical gate, while the orchestrator is responsible for driving additional rounds under the contract below. Includes an automated fix-and-retry loop with a 3-round hard cap, mandatory AGENTS.md compliance checking, and Conventional Commits message generation.
git diff --staged shows contentcsa run --sa-mode true --skill ai-reviewed-commit "Review and commit the staged changes"
When operating under SA mode (e.g., dispatched by /sa or any autonomous workflow),
ALL csa invocations MUST include --sa-mode true. This includes csa run,
csa review, csa debate, and any other execution commands. Omitting --sa-mode
at root depth causes a hard error; passing false when the caller is in SA mode
breaks prompt-guard propagation.
git add the target files.git diff --stat --staged. If >= 500 lines, consider splitting the commit.csa debate "Review my staged changes for correctness, security, and test gaps. Run 'git diff --staged' yourself."csa review --diff --allow-fallbackWHILE/GOTO.parallel-fix pattern (parallel RECON / serial EDIT) instead of a single-employee fix. See patterns/parallel-fix/skills/parallel-fix/SKILL.md.Timing/Race Scenarios, verify that matching regression tests exist and are named under Regression Tests Added; missing or mismatched tests are a blocking failure. Zero unchecked items before proceeding.csa run "Run 'git diff --staged' and generate a Conventional Commits message" (tier-1).git commit -m "${COMMIT_MSG}".| Command | Effect |
|---------|--------|
| /ai-reviewed-commit | Review staged changes and commit |
| /ai-reviewed-commit files=src/lib.rs | Stage, review, and commit specific file |
csa-review (for other-authored code), debate (for self-authored code)commit (Step 8 pre-commit review)git status confirms commit was created.development
Use when running a non-blocking CSA background code health scan that uses csa health and csa tokuin estimate to propose refactoring GitHub issues for files over token or complexity thresholds.
data-ai
Recover main-agent context after `/clear`, `/compact`, or lost local thread state by using `csa recall` against recorded Claude main sessions.
tools
Use when: merged PR had HIGH/CRITICAL findings that represent a bug class — extracts reusable coding rule
tools
Use when: review found 2+ independent findings in different files, fix phase can parallelize RECON