Agent skill
code-review
Perform structured code reviews focusing on correctness, readability, security, and maintainability. Use this skill when reviewing pull requests, evaluating code changes, or establishing review standards for a team.
Install this agent skill to your Project
npx add-skill https://github.com/s-hiraoku/synapse-a2a/tree/main/.claude/skills/code-review
SKILL.md
Code Review
Perform structured, actionable code reviews.
When to Use
- Reviewing a pull request or set of changes
- Evaluating code quality before merging
- Auditing a module for technical debt
- Establishing review standards or checklists
Review Dimensions
Review code across these dimensions, in priority order:
1. Correctness
- Does the code do what it claims?
- Are edge cases handled? (null, empty, overflow, concurrency)
- Are error paths tested?
- Do tests cover the changed behavior?
2. Security
- Input validation at system boundaries
- No secrets in code (API keys, passwords, tokens)
- SQL/command injection prevention
- Proper authentication and authorization checks
- See
security-auditskill for deeper analysis
3. Readability
- Clear naming (variables, functions, classes)
- Functions do one thing
- No deep nesting (max 3 levels)
- Comments explain "why", not "what"
- Consistent style with the surrounding codebase
4. Maintainability
- No unnecessary abstractions
- DRY without over-abstraction (rule of three)
- Dependencies are justified
- Breaking changes are flagged
5. Performance
- Only flag when there is a real concern (hot path, large data, N+1 queries)
- Do not micro-optimize unless the context demands it
Review Output Format
Structure feedback as:
## Review: <PR title or file>
### Must Fix
- [ ] **file.py:42** — [Correctness] Description of the issue and suggested fix
### Should Fix
- [ ] **file.py:78** — [Readability] Description and suggestion
### Consider
- [ ] **file.py:100** — [Performance] Optional improvement
### Positive
- file.py:15 — Good use of context manager for resource cleanup
Severity levels:
| Level | Meaning | Merge? |
|---|---|---|
| Must Fix | Bug, security issue, or broken contract | Block |
| Should Fix | Significant readability/maintainability concern | Request changes |
| Consider | Optional improvement, style preference | Approve with comment |
| Positive | Good patterns worth highlighting | - |
Guidelines
- Be specific - Point to exact lines, suggest concrete alternatives
- Explain why - "This could cause X because Y", not just "change this"
- Separate style from substance - Automate style (linters); review logic manually
- Limit scope - Review what changed, not the entire file (unless asked)
- Acknowledge good work - Include at least one positive observation
- Propose, don't impose - "Consider using X" not "You must use X" (unless it's a Must Fix)
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
s-hiraoku/synapse-a2a
task-planner
Guide for decomposing large tasks into a structured plan with dependency chains, managing priorities, and distributing work across agents. Outputs plan cards or delegation messages as the team contract; TodoList for personal micro-steps.
s-hiraoku/synapse-a2a
react-performance
Comprehensive React and Next.js performance optimization guide. Covers waterfall elimination, bundle size reduction, server-side optimization, re-render prevention, and rendering performance. Use when building, reviewing, or optimizing React/Next.js applications for speed.
s-hiraoku/synapse-a2a
release
Update version in pyproject.toml, plugin.json, and add changelog entry. This skill should be used when the user wants to bump the version number and update CHANGELOG.md. Triggered by /release or /version commands.
s-hiraoku/synapse-a2a
api-design
Guide API design for REST, GraphQL, gRPC, and CLI interfaces. Use this skill when designing new APIs, reviewing existing API contracts, or establishing API conventions for a project. Produces consistent, well-documented API specifications.
s-hiraoku/synapse-a2a
pr-guardian
Continuously monitor a GitHub PR for merge conflicts, CI failures, and CodeRabbit review comments, then automatically fix any issues found. Polls every 5 minutes and loops until every check is green. Use this skill whenever a PR has just been created or code has been pushed to a PR branch — it should be the default follow-up action after any PR creation or push. Also trigger on: "watch this PR", "guard this PR", "monitor CI", "keep fixing until green", "PRを監視して", "CIが通るまで 直して", /pr-guardian. When a PostToolUse hook reports that a push or PR creation just happened, proactively invoke this skill to start monitoring without waiting for the user to ask.
s-hiraoku/synapse-a2a
post-impl2
Workflow: Test workflow with non-existent agent target. . Triggered by /post-impl2 command.
Didn't find tool you were looking for?