postmortem
End-of-session capture of failures, surprises, and lessons so agent behavior and repo config improve. Use after non-trivial work; skip trivial sessions unless something went wrong.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-01
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- postmortem
- description
- End-of-session capture of failures, surprises, and lessons so agent behavior and repo config improve. Use after non-trivial work; skip trivial sessions unless something went wrong.
03 SKILL.md 正文
Postmortem
Purpose
Record what broke, what was confusing, and what should change so the same mistake is less likely next time. Prefer small, concrete follow-ups (one rule, one hook, one skill tweak) over large rewrites.
When to use
- After debugging, CI/test failures, or repeated tool friction
- After non-obvious design or dependency decisions
- When you would want a one-paragraph “what we learned” for the next session
When to skip
- Trivial sessions (typo, one-line fix, format-only) with no unexpected failure or surprise
Steps
- Outcome: What shipped or what state was reached (one or two sentences).
- What went wrong or surprised you: Facts only — errors, false assumptions, wasted loops.
- Root cause (best guess): Missing doc, wrong default, flaky test, unclear rule, etc.
- Action items: Classify each as rule (
AGENTS.md/.cursor/rules/), hook (.claude/settings.json), skill (.claude/skills/), agent (.claude/agents/), or no change if noise. Prefer one high-value change over a list of maybes. - Optional: Add a dated one-line bullet under Recent learnings in root
CLAUDE.mdwhen the lesson is stable and repo-wide.
Use /improve-claude-config when the follow-up is primarily evolving .claude/ layout or automation.
04 同仓库更多 Skills
security-review
Perform a structured security review of the target code
test-and-fix
Run unit tests and automatically fix code failures, regression bugs, or test mismatches. Use when tests are failing, after implementing new features, or to repair "broken" tests.
setup-dev-env
Set up the development environment for the project. Use when starting work on the project, when dependencies are out of sync, or to fix environment setup failures.
security-vulnerability-audit
Workflow for auditing security vulnerabilities using Trunk (Trivy and OSV-scanner). Use when checking for project vulnerabilities, hard-coded secrets, or repairing security flaws.
security-scan
Scan for vulnerable dependencies and CVEs via pnpm lint:security (Trunk Trivy/OSV-scanner), optional pnpm security:grype or pnpm lint:all (Trunk-all plus Grype). Use for dependency CVE checks, security-scoped lint, or post-bump verification.
node-upgrade
Safely upgrade Node.js dependencies in pnpm workspaces. Use when asked to "upgrade dependencies", "update packages", "check for updates", or fix version mismatches.