run-retrolisted
Install: claude install-skill mike-arbuzov365/engineering-intelligence-framework
# Skill: Run Retro
<!-- Knowledge source: GENERALIZE of the private EI's retro SKILL.md. Thin
pointer to playbooks/run-retro.md - one canonical source per D-007. -->
Full workflow: [`playbooks/run-retro.md`](../../playbooks/run-retro.md).
## When this applies
After an execution packet closeout, after a significant incident, or on a
2-4 week cadence for an active project instance. **Not** after every
session: a retro over one session has no repetition to find.
## Process
1. Derive the period's activity from the repository (`git log`,
`git shortlog`, merged Knowledge Deltas, session and packet closeouts) -
not from recollection.
2. Look for repetition: repeated failure signatures, blind-fix cycles,
skipped evidence, skipped retrieval, policy-induced failures, hotspots.
3. Route each genuinely recurring pattern through
[`knowledge-ingest.md`](../../playbooks/knowledge-ingest.md), then
[`knowledge-lint.md`](../../playbooks/knowledge-lint.md) if it is a
promotion candidate.
4. Write the artifact from
[`templates/retro.md`](../../templates/retro.md).
5. Apply any new or changed rule to the persistent agent-instruction file
in this same retro.
## Output
```markdown
Retro: <packet | incident | period> Period: <start> - <end>
Recurring patterns: <N> (each with occurrence count and where observed)
New artifacts: <paths, or "none">
Promotion decisions: <promoted / stays local, per artifact>
Entrypoint updates applied: <yes, what | none>
```
## Boundari