review-spec

📁 ferueda/agent-skills 📅 Jan 22, 2026
26
总安装量
26
周安装量
#14310
全站排名
安装命令
npx skills add https://github.com/ferueda/agent-skills --skill review-spec

Agent 安装分布

codex 21
opencode 21
gemini-cli 20
claude-code 18
github-copilot 15
cursor 14

Skill 文档

Review the given spec document by analyzing both the spec AND the referenced codebase.

Process

  1. Read the spec – Understand goals, phases, and proposed changes
  2. Explore the codebase – Read files and code mentioned in the spec
  3. Analyze patterns – Identify existing patterns, conventions, and architecture
  4. Validate the plan – Ensure proposals align with codebase reality
  5. Identify issues – Find gaps, risks, and improvement opportunities

Review Dimensions

Evaluate across these areas (focus on what’s relevant):

  • Architecture: Component boundaries, data flow, API contracts, separation of concerns
  • Feasibility: Implementation complexity, technology trade-offs, effort estimation
  • Reliability: Error handling, retries, idempotency, graceful degradation
  • Performance: Bottlenecks, caching, query patterns, scaling approach
  • Security: Auth, data protection, input validation, audit logging
  • Edge Cases: Null handling, limits, timeouts, race conditions, partial failures
  • Testing: Testability, integration strategy, rollback considerations

Output Format

For each finding:

### [Finding Title]

**Category**: Architecture | Feasibility | Reliability | Performance | Security | Edge Case | Testing
**Severity**: Critical | High | Medium | Low
**Section**: [Spec section or phase]

**Issue**: [Clear problem description]
**Recommendation**: [Specific, actionable change]
**Rationale**: [Technical justification]

Guidelines

  • Verify against code – Don’t trust the spec blindly; check actual implementations
  • Follow existing patterns – Recommendations should align with codebase conventions
  • Be specific – Reference exact files, functions, and line numbers
  • Prioritize – Order findings by severity and impact
  • Challenge assumptions – Question decisions that lack justification