Review Implementation — Security skill for Claude Code
Independent implementation audit of uncommitted code by Codex CLI against a plan doc, then apply agreed fixes to the code by default.
How to install Review Implementation
Installs to ~/.claude/skills/philadamson93-research-skills-review-implementation/SKILL.md
mkdir -p ~/.claude/skills/philadamson93-research-skills-review-implementation && curl -fsSL https://raw.githubusercontent.com/philadamson93/research-skills/HEAD/commands/review-implementation.md -o ~/.claude/skills/philadamson93-research-skills-review-implementation/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Review Implementation does
name: review-implementation description: Independent implementation audit of uncommitted code by Codex CLI against a plan doc, then *apply* agreed fixes to the code by default. Use after a non-trivial implementation completes, BEFORE invoking /commit-review. Invoked explicitly via `/review-implementation []`. Also offered proactively after substantial implementation work — gated by AskUserQuestion since not every change warrants a Codex audit. Runs `codex exec` ov
Alternatives in Security
- Sovereign Skills — 20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session man 125 ★
- Mobile Resume — Audit an unfinished mobile app, create a completion plan, then implement remaining work after approval 107 ★
- Test — Prove it works 42 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Pause After Planning
Standing instruction: finish planning and verification (including plan-gate and plan-auditor), then stop so yo
Gaia Audit
Audit memory, wiki, and auto-loaded files for duplication, conflicting instructions, and stale content. The de
Orchestrate Consult
Run an independent external audit of a completed ZOdyssey run. Hands the plan + full git diff to the external
Pre Ship Performance Audit
Agent skill that audits and fixes web app performance before launch: four read-only subagents check images, fo
Cleanfiles
Sweep the repo's accumulating working files: audit first, apply only after operator confirmation, one reviewab
UI Refine
Iterative UI refinement: audit existing UI, select issues to fix, apply fixes, re-audit. Provide file paths or
Related Agents
Principal Code Reviewer
Review non-trivial Kotlin/Java changes, crash/file/auth/navigation/lifecycle/security fixes, or multi-file dif
Static Analysis
Use after code-generator completes to run static analysis tools and report findings. Runs in parallel with sec
Codex Adversarial Reviewer
Adversarial reviewer that calls Codex CLI for an independent second-opinion review of changed files, then adju