--- name: afrexai-sprint-retro description: "Sprint Retrospective Facilitator" --- # Sprint Retrospective Facilitator Run structured sprint retrospectives that produce real action items — not vague "we should communicate better" fluff. ## What It Does 1. **Collects input** from the team across 4 categories: What went well, What didn't, What confused us, What should we try next 2. **Groups and prioritizes** themes by frequency and impact 3. **Generates action items** with owners, deadlines, and success criteria 4. **Produces a retro report** ready to share with stakeholders 5. **Tracks patterns** across sprints to surface recurring issues ## How to Use Tell the agent: - `Run a retrospective for Sprint 24` — starts the full facilitation flow - `Quick retro: [paste team feedback]` — processes raw feedback into structured output - `Compare last 3 retros` — finds recurring themes and stalled action items ## Frameworks Supported - **Start/Stop/Continue** — classic, works for any team size - **4Ls** — Liked, Learned, Lacked, Longed For - **Mad/Sad/Glad** — emotional temperature check - **Sailboat** — Wind (helps), Anchors (blockers), Rocks (risks), Island (goals) - **DAKI** — Drop, Add, Keep, Improve ## Output Format ```markdown # Sprint [N] Retrospective — [Date] ## Top Themes 1. [Theme] — mentioned by X people, impact: HIGH/MED/LOW ## Action Items | # | Action | Owner | Due | Success Criteria | |---|--------|-------|-----|------------------| | 1 | ... | ... | ... | ... | ## Patterns (vs. previous sprints) - [Recurring issue] — appeared in 3/5 last retros - [Improvement] — resolved since Sprint N-2 ``` ## Tips - Run retros within 24 hours of sprint end while memory is fresh - Limit to 3-5 action items per sprint — more than that and nothing gets done - Assign every action item to ONE person (shared ownership = no ownership) - Review previous retro actions at the start of each new retro --- Built by [AfrexAI](https://afrexai-cto.github.io/context-packs/) — AI context packs for business teams. Browse all 10 industry packs at $47 each, or grab the full bundle for $197.