agents/skills/retro/SKILL.md
Run a structured retrospective or post-incident review with optional knowledge base capture. Use for running retrospectives, postmortems, or incident reviews.
npx skillsauth add drn/dots retroInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Structured retrospective for projects, sprints, or incidents. Captures learnings and optionally persists them to the knowledge base.
$ARGUMENTS - Optional: --incident for post-incident template, or a topic/project namegit log --oneline -20 2>/dev/null | head -20gh pr list --state merged --limit 10 2>/dev/null | head -10cat context/knowledge/index.md 2>/dev/null | head -30git branch --show-currentIF --incident is in $ARGUMENTS, use the Incident template.
OTHERWISE, use the Standard template.
Review the recent commits and merged PRs from the context above. Summarize the scope of work that this retro covers. If the user provided a topic, focus on that.
## Retrospective: <topic or date>
### What Went Well
- <positive outcomes, wins, good patterns>
### What Did Not Go Well
- <pain points, friction, mistakes>
### Action Items
- [ ] <concrete next step with owner if applicable>
### Key Learnings
- <insights to carry forward>
## Post-Incident Review: <incident name>
### Timeline
| Time | Event |
|------|-------|
| | <what happened> |
### Impact
- **Users affected:**
- **Duration:**
- **Severity:**
### Root Cause
<what actually caused the issue>
### What Went Well
- <effective response actions>
### What Did Not Go Well
- <gaps in detection, response, or prevention>
### Action Items
- [ ] <concrete remediation with owner and deadline>
### Follow-ups
- [ ] <longer-term improvements>
Present the template pre-filled with whatever you can infer from the git context. Ask the user to fill in, correct, or expand each section. Iterate until the user is satisfied.
IF a knowledge base exists (check context above for context/knowledge/index.md):
/knowledge (bold entity names, standalone facts, date-stamp volatile info)IF no knowledge base exists, skip this step entirely. Do not suggest creating one.
Offer to:
.context/retros/<date>-<slug>.mddevelopment
Walk every unresolved review thread on a PR, triage each one, reply with a rationale of whether or not the comment will be acted upon, make the code change if warranted, and mark the thread resolved. Use when the user asks to address only the open PR comments without re-running CI, respond to review feedback, resolve review threads, or clear bot comments on a PR.
tools
Iteratively run /rereview, fix the findings, and loop until reviewers approve clean. Use for iterative automated review, when you want /rereview to loop until clean, or for a paranoid pre-merge review that auto-addresses every blocker.
development
Generate self-contained HTML visualizations with Plannotator theming. Use for implementation plans, PR explainers, architecture diagrams, data tables, slide decks, and any visual explanation of technical concepts. Plans and PR explainers follow Plannotator's prescriptive approach; all other visual content delegates to nicobailon/visual-explainer.
development
Create reviewed Codex goal setup packages for long-running /goal work. Use when the user wants to turn an idea, backlog, project mission, or vague objective into durable goal files under a project goals slug folder, with Plannotator review gates for brief, narrative plan with acceptance criteria, verification, blockers, and the final /goal prompt.