From qe-framework
Facilitates structured team retrospectives (Start/Stop/Continue, 4Ls, Sailboat), pre-mortem risk analysis, and release notes generation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/qe-framework:Qpm-retroThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**Distinct from:** Qlesson-learned (git history analysis). This skill facilitates real-time team reflection sessions using structured formats (Start/Stop/Continue, 4Ls, Sailboat) and conducts pre-mortem risk analysis. It does not analyze historical git commits or extract lessons from code changes (see /Qlesson-learned). Retros are forward-looking; lessons-learned extracts insights from what was...
Distinct from: Qlesson-learned (git history analysis). This skill facilitates real-time team reflection sessions using structured formats (Start/Stop/Continue, 4Ls, Sailboat) and conducts pre-mortem risk analysis. It does not analyze historical git commits or extract lessons from code changes (see /Qlesson-learned). Retros are forward-looking; lessons-learned extracts insights from what was built.
Run structured retrospectives, conduct pre-mortem risk analysis, and generate release notes. Provides multiple facilitation formats so teams can rotate approaches and avoid retro fatigue.
| Format | Best For |
|---|---|
| Start / Stop / Continue | Teams new to retros or short timebox (30 min) |
| 4Ls (Liked, Learned, Lacked, Longed for) | Reflective teams that want deeper insight |
| Mad / Sad / Glad | Surfacing emotional undercurrents the team avoids |
| Sailboat (Wind, Anchor, Rock, Island) | Visual thinkers and metaphor-driven discussion |
See references/retro-format-templates.md for all format templates and facilitation tips.
"Imagine it is [deadline]. The project has failed. What went wrong?"
Step 1: Failure Brainstorm (10 min) Each participant silently writes failure scenarios — the more specific, the better.
Step 2: Risk Categorization (10 min) Group failures into categories: Technical, People, Process, External, Scope.
Step 3: Probability/Impact Assessment (10 min)
| Risk | Category | Probability (1-5) | Impact (1-5) | Score | Mitigation |
|---|---|---|---|---|---|
| P x I |
Step 4: Mitigation Planning (15 min) For risks with score >= 9, define concrete mitigation actions.
## Pre-mortem: [Project Name]
**Date:** [date] | **Target deadline:** [date]
### Project Context
- Goal: [one-line summary]
- Team: [names/roles]
- Timeline: [start] to [deadline]
### Failure Scenarios
1. [specific failure scenario]
2. [specific failure scenario]
3. ...
### Risk Matrix
| # | Risk | Category | P (1-5) | I (1-5) | Score | Mitigation |
|---|------|----------|---------|---------|-------|------------|
| 1 | | | | | | |
| 2 | | | | | | |
### Top Risks & Mitigations (Score >= 9)
| Risk | Mitigation | Owner | Checkpoint |
|------|------------|-------|------------|
| | | | |
### Assumptions to Validate
- [ ] [assumption that, if wrong, changes the plan]
# [Product Name] v[X.Y.Z] Release Notes
**Release date:** [date]
## New Features
- **[Feature name]** — [one-line user benefit]. [Optional: brief how-to]
## Improvements
- **[Area]** — [what changed and why it matters to users]
## Bug Fixes
- Fixed [symptom users experienced] ([issue ref])
## Breaking Changes
- **[What changed]** — [what users need to do]. See [migration guide link].
## Deprecations
- [Feature] will be removed in v[X+1]. Use [alternative] instead.
# [Product Name] v[X.Y.Z] — Internal Release Summary
**Release date:** [date] | **Release manager:** [name]
## Changes Included
| Type | Description | PR | Author |
|------|-------------|----|--------|
| feat | | | |
| fix | | | |
| refactor | | | |
## Deployment Notes
- [ ] Database migrations: [yes/no, details]
- [ ] Config changes: [env vars, feature flags]
- [ ] Rollback plan: [steps]
## Known Issues
- [issue description] — [workaround if any]
## Metrics to Watch
- [metric to monitor post-deploy]
What went as planned, What didn't, Actions, Shoutouts.
Best for: end-of-project or end-of-quarter wrap-ups (longer than a sprint retro).
## Post-Project Review: [Project Name]
**Date:** [date] | **Duration:** [project duration]
### What Went As-planned
- [thing that worked according to plan and why]
### What Didn't Go As-planned
- [thing that diverged from plan]
- **Root cause:** [why]
- **Impact:** [what it cost in time/scope/quality]
### Actions for Next Time
| Action | Category | Priority |
|--------|----------|----------|
| | Process / Technical / People | High / Med / Low |
### Shoutouts
- [person] — [specific contribution worth recognizing]
### Key Metrics
| Metric | Planned | Actual | Delta |
|--------|---------|--------|-------|
| Timeline | | | |
| Scope | | | |
| Quality (bugs found post-launch) | | | |
User: Run a retrospective for our last sprint
User: Run a retrospective for this sprint
User: Do a pre-mortem for this project
User: Write release notes for this version
User: Help me write release notes from these PRs
User: Help me with a post-mortem for this project
Credits: Frameworks adapted from phuryn/pm-skills (MIT)
npx claudepluginhub inho-team/qe-framework --plugin qe-frameworkCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.