retro-runner
Skill1 starsChanged 7 months agoMITView on GitHub
Facilitate structured retrospectives for sprints, launches, experiments, or incidents. Use when asked to run a retro, post-mortem, or reflect on what happened.
In this repository26 other files it gives its agents.
- AGENTS.mdAGENTS.md
- app-store-analyzerSkill
- brainstormingSkill
- competitor-trackerSkill
- continual-learningSkill
- decision-loggerSkill
- exec-communicatorSkill
- experiment-designerSkill
- experiment-writeupSkill
- growth-loop-mapperSkill
- jtbd-extractorSkill
- knowledge-updaterSkill
- launch-postSkill
- launch-readinessSkill
- meeting-to-actionsSkill
- metric-storytellerSkill
- one-pagerSkill
- pptx-creatorSkill
- prd-writerSkill
- prioritization-scorerSkill
- stakeholder-updateSkill
- strategy-connectorSkill
- user-interview-synthSkill
- what-ifSkill
- working-backwardsSkill
- writing-clearlySkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
After any milestone, sprint, or incident
- Distinctive: line 17
When the team feels stuck in repeat patterns
- Distinctive: line 23
Ask (one at a time):
- Very specific: line 54
Read memory/decisions/ for related past decisions
- Distinctive: line 55
Check if the same issues recurred — flag repeat offenders
- Very specific: line 71
Write the retro output to memory/decisions/YYYY-MM-DD_retro-[topic].md so future retros can reference it.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Retro Runner
Facilitate structured retrospectives that produce concrete improvements — not just venting sessions. Works for sprints, launches, experiments, incidents, or any milestone worth reflecting on.
When to Use
- "Run a retro on..."
- "What went well / badly with..."
- "Post-mortem for..."
- "Let's reflect on the launch"
- After any milestone, sprint, or incident
- When the team feels stuck in repeat patterns
Process
1. Set the Scope
Ask (one at a time):
- What are we retro-ing? (sprint, launch, experiment, incident, quarter)
- Time window? (last 2 weeks, last month, since launch)
- Who was involved? (just you, your squad, cross-functional team)
2. Gather Signals
Prompt the user for input in four categories:
What worked — things to keep doing
- What felt smooth?
- What results exceeded expectations?
- What should we do more of?
What didn't work — things to stop or change
- Where did we feel friction?
- What took longer than expected?
- What would we do differently?
What surprised us — things we didn't predict
- Unexpected wins or failures
- Customer behavior we didn't anticipate
- Dependencies or blockers we missed
What we learned — durable insights
- New knowledge about our users, product, or process
- Patterns across multiple retros (reference
memory/for prior retros)
3. Identify Patterns
Cross-reference with prior retros if available:
- Read
memory/decisions/for related past decisions - Check if the same issues recurred — flag repeat offenders
- Note if a previous action item would have prevented this issue
4. Extract Action Items
For each issue identified:
- Action — specific, verb-first improvement
- Owner — who will drive it
- Type — process change, tool fix, communication fix, skill gap
- Reversibility — is this easy to try and revert?
- Due — when should this be done
Prioritize ruthlessly: max 3 action items. More than 3 means nothing gets done.
5. Log the Decision
Write the retro output to memory/decisions/YYYY-MM-DD_retro-[topic].md so future retros can reference it.
Output
## Retro: [Topic]
**Date:** YYYY-MM-DD
**Scope:** [what was retro'd]
**Participants:** [who was involved]
### What Worked
- [item]
### What Didn't Work
- [item]
### Surprises
- [item]
### Patterns
- [recurring theme or first-time issue]
### Action Items (max 3)
| # | Action | Owner | Type | Due |
|---|--------|-------|------|-----|
| 1 | [specific action] | [name] | [type] | [date] |
| 2 | [specific action] | [name] | [type] | [date] |
| 3 | [specific action] | [name] | [type] | [date] |
### Key Takeaway
[One sentence: the most important thing we learned]Principles
- Blameless — focus on systems, not people
- Pattern-aware — connect to prior retros
- Action-capped — max 3 items, or nothing sticks
- Forward-looking — retros exist to improve the future, not relitigate the past
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

