菜单
Use when implementing any feature or bugfix, before writing implementation code. Enforces RED-GREEN-REFACTOR
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Debug frontend vanilla JS issues - console errors, rendering bugs, SPA navigation, WebSocket
Automates the process of setting up the first admin user or adding new admins via the bootstrap endpoint. Includes a Python script for easy execution.
Push to remote and wait for CI to pass. If CI fails, read logs, fix bugs, and re-push. Repeat until green.
Use when building or debugging LangGraph multi-agent systems - eval-first execution, task decomposition, model routing by complexity, and cost discipline
Use when AI agent modifies API routes or backend logic - catch systematic blind spots where the same model writes and reviews code
Use when making or recording significant architectural decisions - capture context, alternatives, and rationale as structured ADRs
| name | test-driven-development |
| description | Use when implementing any feature or bugfix, before writing implementation code. Enforces RED-GREEN-REFACTOR |
NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST
Write code before the test? Delete it. Start over. No exceptions.
# Good
def test_retries_failed_operations_3_times():
result = retry_operation(failing_operation)
assert result == 'success'
assert attempts == 3
pytest tests/path/test.py::test_name -v
Write simplest code to pass. Don't add features, refactor, or "improve" beyond the test.
pytest tests/path/test.py -v
After green only: remove duplication, improve names, extract helpers. Keep tests green.
| Excuse | Reality |
|---|---|
| "Too simple to test" | Simple code breaks. Test takes 30 seconds. |
| "I'll test after" | Tests passing immediately prove nothing. |
| "Deleting X hours is wasteful" | Sunk cost fallacy. Keeping unverified code is debt. |
| "TDD will slow me down" | TDD faster than debugging. |
| "Existing code has no tests" | Add tests for existing code you're improving. |