reviews-retros-reflection — Claude Code skill: setup and use
reviews-retros-reflection structures retrospectives, incident post-mortems, and periodic check-ins. Helps teams identify root causes and turn feedback into SMART action items without blame.
no license 113 installs ★ 203 updated 2026-09-01
npx skills add lyndonkl/claude --skill reviews-retros-reflection -a claude-codeWhat reviews-retros-reflection does
The skill walks teams through a five-step retrospective: setting stage context, gathering metrics and events, uncovering root causes, agreeing on fixes, and tracking follow-through. It uses techniques like the 5 Whys and timeline analysis to examine broken processes rather than targeting individual contributors.
Several workshop models are provided depending on available time and team needs. A thirty-minute Start/Stop/Continue session suits routine sprint reviews, while Mad/Sad/Glad helps unpack difficult emotional periods. For quarterly milestones or complex project shutdowns, the methodology suggests the 4Ls framework or the visual Sailboat exercise.
Piling up more than five action items guarantees none will get done. Because of this, the workflow insists on capping follow-ups at three SMART tasks and aiming for an eighty percent completion rate before the next retro. Final meeting quality is verified against resources/evaluators/rubric_reviews_retros_reflection.json with a minimum required score of 3.5.
What it helps with
- facilitate a sprint retrospective using the Start/Stop/Continue model
- conduct a blameless post-mortem using timeline and 5 Whys analysis
- turn retro discussions into one to three assigned SMART actions
- lead an end-of-project debrief with the 4Ls framework
Example prompts for the agent
- Help me facilitate a 45-minute sprint retrospective using Mad/Sad/Glad.
- Let's analyze yesterday's production outage and trace root causes with 5 Whys.
- Structure our quarterly reflection using the 4Ls approach.
How to install reviews-retros-reflection
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill reviews-retros-reflection -a claude-code2Manually: put the files in place
~/.claude/skills/reviews-retros-reflection/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/reviews-retros-reflection/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill reviews-retros-reflection -a codex2Manually: put the files in place
.agents/skills/reviews-retros-reflection/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill reviews-retros-reflection -a cursor2Manually: put the files in place
.cursor/skills/reviews-retros-reflection/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lyndonkl/claude skills/reviews-retros-reflection2Manually: put the files in place
.github/skills/reviews-retros-reflection/SKILL.md1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill reviews-retros-reflection -a gemini-cli2Platform CLI
gemini skills install https://github.com/lyndonkl/claude --path skills/reviews-retros-reflectionManually: put the files in place
.devin/skills/reviews-retros-reflection/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill reviews-retros-reflection -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about reviews-retros-reflection
How do I install reviews-retros-reflection in Claude Code?
Run npx skills add lyndonkl/claude --skill reviews-retros-reflection -a claude-code. The file installs locally to ~/.claude/skills/reviews-retros-reflection/SKILL.md.
What is reviews-retros-reflection used for?
It guides engineering teams through sprint retrospectives, post-mortems, and recurring reviews. The skill keeps discussions structured around root cause analysis while maintaining psychological safety.
How many action items should a retro produce?
The skill enforces a limit of one to three SMART actions with explicit owners. Tracking requires completing over 80% of these tasks before the next review session.