// Systematically evaluate code changes for security, correctness, performance, and spec alignment. Use when reviewing PRs, assessing code quality, or verifying implementation against requirements.
Defines required structure, frontmatter format, and best practices for SKILL.md files. Use BEFORE creating or editing any skill - this is the spec to follow, not optional reference.
Diagnose and fix bugs through systematic investigation, root cause analysis, and targeted validation. Use when something is broken, errors occur, performance degrades, or unexpected behavior manifests.
Maintain project documentation synchronized with code. Keep feature specs, API contracts, and README current with init-project standards. Use when updating docs after code changes, adding new features, or ensuring documentation completeness.
Systematically trace code flows, locate implementations, diagnose performance issues, and map system architecture. Use when understanding how existing systems work, researching concepts, exploring code structure, or answering "how/where/why is X implemented?" questions.
Coordinate concurrent task execution through agent delegation. Plan independent work, manage dependencies, and execute multiple agents simultaneously. Use when handling multiple unrelated tasks, research investigations, or layer-based implementations that can run concurrently.
Create structured implementation plans before coding. Use when breaking down complex features, refactors, or system changes. Validates requirements, analyzes codebase impact, and produces actionable task breakdowns with identified dependencies and risks.
| name | Reviewing Code |
| description | Systematically evaluate code changes for security, correctness, performance, and spec alignment. Use when reviewing PRs, assessing code quality, or verifying implementation against requirements. |
Evaluate code changes across security, correctness, spec alignment, performance, and maintainability. Apply sequential or parallel review based on scope.
Sequential (small PRs, <5 files):
Parallel (large PRs, >5 files):
Read documentation:
docs/feature-spec/F-##-*.md — Technical design and requirementsdocs/user-stories/US-###-*.md — Acceptance criteriadocs/api-contracts.yaml — Expected API signaturesdocs/data-plan.md — Event tracking requirements (if applicable)docs/design-spec.md — UI/UX requirements (if applicable)docs/system-design.md — Architecture patterns (if available)docs/plans/<slug>/plan.md — Original implementation plan (if available)Determine scope:
Security (/25)
Correctness (/25)
Spec Alignment (/20)
docs/api-contracts.yamldocs/data-plan.mddocs/design-spec.mdPerformance (/15)
Maintainability (/15)
Total: /100
Format:
Location: file.ts:123
Problem: [Description]
Impact: [Risk/consequence]
Fix: [Specific change needed]
Spec reference: [docs/api-contracts.yaml line X]
Highlight what was done well for learning
Best for <5 files, single concern:
Best for >5 files, multiple concerns:
Spawn specialized agents:
senior-engineer for vulnerability assessmentExplore for pattern complianceprogrammer for endpoint validationprogrammer for UI/UX and accessibilitydocumentor for comment quality and docsEach agent reviews specific quality dimension
Consolidate findings into single report
# Code Review: [Feature/PR]
## Summary
**Quality Score:** [X/100]
**Issues:** Critical: [N], Important: [N], Nice-to-have: [N]
**Assessment:** [APPROVE / NEEDS REVISION / MAJOR REWORK]
## Spec Compliance
- [ ] APIs match `docs/api-contracts.yaml`
- [ ] Events match `docs/data-plan.md`
- [ ] UI matches `docs/design-spec.md`
- [ ] Logic satisfies story AC
## Findings
### Critical Issues
[Issues with fix recommendations]
### Important Issues
[Issues that should be addressed]
### Nice-to-Have Suggestions
[Optional improvements]
### Good Practices
[What worked well]
## Recommendations
[Next steps: approval, revision needed, etc.]
Offer options:
Parallel fixes for large revisions:
Document format:
✅ FIXED: [Issue name]
File: [path:line]
Change: [what changed]
Verification: [how to test]
Check if specs need updates:
Always flag for user approval before modifying specs.