Self Review — Security skill for Claude Code
Run a full self-review-and-fix loop on the current changes.
How to install Self Review
Installs to ~/.claude/commands/kooleklabs-agentic-setup-self-review.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/kooleklabs/agentic-setup/HEAD/.claude/commands/self-review.md -o ~/.claude/commands/kooleklabs-agentic-setup-self-review.md Restart Claude Code, or start a new session, for it to be picked up.
What Self Review does
description: Run a full self-review-and-fix loop on the current changes. Checks tests, lint, and security — fixing each issue before moving to the next. Run before every commit or PR.
Run a self-review-and-fix loop on the current working changes:
Step 1 — Tests
- Detect and run the project test suite (check package.json, go.mod, Cargo.toml, pytest)
- If tests fail:
- Read each failure carefully
- Fix the root cause (not the test)
- Re-run until all pass
- If no test suit
Alternatives in Security
- Skill Coverage Audit — Trace codepaths in diffs, map against tests, auto-generate missing coverage — use before shipping PRs 2.8k ★
- Brooks Audit — Run a Brooks-Lint architecture audit 1.5k ★
- Engage.Threatmodel — Materialize, lint, and drift-check the engagement threat model 348 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Engineering Fix
Applies engineering corrections from the last /engineering-audit run. Creates an isolated branch, applies chan
Dx Report
Generate a DX report (format/lint/docs/types/tests/build/deps/audit/knip), summarize results, and optionally c
Maintainer Audit
Proactively review a skill or plugin against source-of-truth and the open-issue signal, logging gaps as issues
Audit Assist
Flow B — AI-assisted iterative audit with a human in the loop. Same lifecycle as audit-cycle, but pauses at ch
Engineering Audit
Full-spectrum engineering health audit: scores Maintainability, Security, Testing, Architecture, Documentation
Docsmith Fix Loop
Fix a confirmed docsmith skill/template/script finding, prove the fix with an eval + audit, and — only if it p
Related Agents
Issue Subagents Dev
Dev agent for issue-subagents skill. One-shot per invocation — initial spawn implements the spec via TDD again
Lane Implementer
Implementation lane with the full gate battery baked in - private target dir, all-targets clippy, targeted tes
Bug Fix
Use when diagnosing and fixing a bug. Chains reproduce → isolate → hypothesize → fix → verify → learn with hum