Agent Skills
› langchain-ai/deepagents
› code-review
code-review
GitHub提供结构化的代码审查流程,涵盖正确性、质量、测试及安全维度。通过运行测试套件和lint工具辅助检查,确保代码符合规范并解决潜在问题。
Trigger Scenarios
实现功能变更后的自我验证
提交代码前的质量检查
Install
npx skills add langchain-ai/deepagents --skill code-review -g -y
SKILL.md
Frontmatter
{
"name": "code-review",
"description": "Perform a structured code review of changes, checking for correctness, style, tests, and potential issues."
}
Code Review Skill
Use this skill after implementing changes to validate your work before delivering.
Review Checklist
1. Correctness
- Changes solve the original issue/task
- No unintended side effects on existing functionality
- Edge cases are handled
- Error handling is appropriate (not excessive)
2. Code Quality
- Code matches existing style and patterns
- No unnecessary complexity or abstraction
- Variable and function names are clear
- No dead code, commented-out code, or TODOs left behind
3. Tests
- New functionality has test coverage
- Existing tests still pass
- Tests cover both happy path and error cases
- Tests are not brittle (don't test implementation details)
4. Safety
- No hardcoded secrets or credentials
- User input is validated at boundaries
- No SQL injection, XSS, or command injection vectors
- File operations use safe paths
Process
- Read each modified file end-to-end (not just the diff)
- Run the test suite:
execute("python -m pytest -v") - Run linters if available:
execute("ruff check .") - Run the bundled lint check:
execute("python /skills/code-review/lint_check.py .") - Check against each item in the review checklist
- If any issues found, fix them and re-review
- When everything passes, the review is complete
Helper Scripts
/skills/code-review/lint_check.py— Scans Python files for missing docstrings, long functions (>50 lines), and bareexcept:clauses. Run it viaexecute("python /skills/code-review/lint_check.py [path ...]").
Version History
- 1e4f507 Current 2026-08-20 16:32


