30
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill lyndonkl/claude --skill reviews-retros-reflection- SKILL.md8.5 KB
Overview
This skill guides teams and individuals through structured retrospectives, post-mortems, and reflection sessions to surface lessons, decide improvements, and track outcomes. It emphasizes psychological safety, root-cause thinking, and delivering a small set of SMART actions that drive measurable progress. Use it to build a repeatable continuous-improvement habit across sprints, projects, incidents, and reviews.
How this skill works
It walks you through a five-step workflow: set the stage, gather data, generate insights, decide actions, and close with follow-up. Choose a suitable retro format (Start/Stop/Continue, Mad/Sad/Glad, 4Ls, Sailboat, Timeline), collect facts and sentiment, apply root-cause techniques, vote on priorities, and convert outcomes into 1–3 SMART actions. Track completion and iterate on recurring issues to ensure real improvement.
When to use it
- Sprint or iteration retrospectives (Agile/Scrum)
- Project post-mortems and after-action reviews
- Weekly team reviews or monthly check-ins
- Incident reviews and blameless root-cause analysis
- Quarterly/annual reflections and process improvement sessions
- When the team says “retro”, “what went well”, or “lessons learned”
Best practices
- Start with a clear scope and prime directive to establish psychological safety
- Limit to 1–3 high-impact SMART actions; avoid >5 actions
- Rotate the facilitator and time-box each section to keep momentum
- Use data plus sentiment (metrics, events, team mood) to inform insights
- Vote (dot voting) to prioritize focus and ensure shared ownership
- Track action completion (>80% target) and review previous actions at the next retro
Example use cases
- A two-week sprint retro using Start/Stop/Continue to reduce scope churn
- A post-launch 4Ls session to capture what to replicate and what to document
- A 60-minute incident retrospective using timeline + 5 Whys to close reliability gaps
- A weekly 15-minute team review to celebrate wins and remove blockers
- A quarterly Sailboat session for strategic alignment and risk spotting
FAQ
Aim for 1–3 SMART actions. Too many items dilute focus and reduce completion rates.
What if the same issue recurs every retro?
Escalate persistent problems: assign an owner, define measurable success criteria, and involve leadership if systemic blockers persist.
How do we keep retrospectives blameless?
Use the Prime Directive, focus on process and systems rather than individuals, encourage dissenting views, and make confidentiality expectations explicit.