Code Review Pipeline
Testing & Qualitycode-review-pipelineCode Review Pipeline is an Agent Skill for testing & quality published in the a5c-ai/babysitter repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Multi-dimensional code review across correctness, security, performance, and maintainability with confidence-gated reporting and remediation loops.
QUICK START
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/a5c-ai/babysitter/blob/HEAD/library/methodologies/everything-claude-code/skills/code-review-pipeline/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/code-review-pipeline, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
- Logic errors and off-by-one mistakes
- Edge case handling (null, undefined, empty, boundary)
- Type safety (no implicit any, proper narrowing)
- Error handling completeness
- Floating promise detection
- Race condition analysis
Dimension 2: Security
- Injection vectors (SQL, XSS, command, template)
- Authentication and authorization gaps
- Data exposure (PII, credentials, internal state)
- Dependency vulnerabilities (known CVEs)
- Input validation completeness
Dimension 3: Performance
- Algorithmic complexity (O(n^2) detection)
- Memory leaks (event listeners, closures, caches)
- Unnecessary allocations in hot paths
- Database query optimization (N+1, missing indexes)
- Bundle size impact
Dimension 4: Maintainability
- Naming clarity and consistency
- Documentation completeness (JSDoc, inline comments)
- Test coverage adequacy
- Coupling analysis (afferent/efferent)
- File organization compliance
Confidence Gating
- Score each issue 0-100 on confidence
- Only report issues >= 80% confidence
- Prevents false positive noise
- Higher confidence for clear patterns, lower for heuristic matches
Remediation Loop
- Prioritize: critical > high > medium > low
- Apply fixes via refactor-cleaner agent
- Re-review after remediation
- Maximum 2 remediation cycles
- Exit when no critical/high issues remain
When to Use
- Post-implementation review
- Pre-merge PR review
- Security audit
- Technical debt assessment
Agents Used
code-reviewer(primary)refactor-cleaner(remediation)