Reviews completed activities step by step to extract and persist lessons learned.
npx skills add https://github.com/isizono/calm --skill postmortem-isizono
Please help me install this Agent Skill. Skill: postmortem Source: https://github.com/isizono/calm/tree/main/skills/postmortem Command: npx skills add https://github.com/isizono/calm --skill postmortem-isizono
High-intent search queries and answers about installing and using this skill.
Invoke the skill with an activity_id argument, or let it list up to 15 completed activities via get_activities so you can pick one. It then creates a dedicated retrospective activity and walks through each step interactively.
The review collects logs and decisions tied to related topics, explores linked entities with get_map, and checks stored materials. If logs and decisions are too sparse, it asks you to narrate what happened to build the steps manually.
Lessons can be persisted as tag notes, topic decisions, Claude Code auto memory feedback, CLAUDE.md project rules, or new intent tags. The skill presents the options, but you decide where each lesson goes.
Yes. Progress is logged frequently during the review, so the retrospective can pause and resume across sessions without losing context, since all records live in the persistent memory database.
Do not use it for simply marking an activity complete or syncing memory records; those are handled by activity-finish and sync-memory. It is designed only for reflective review of already completed work.
thiansitAppend plain-text observations and decisions to psi/memory/logs/activity.log.
adobeCapture session corrections and lessons into .ai/memory/lessons.md.
electroheadfxSummarize project phases, plans, requirements, git history, and timeline status.
sunchenddSave conversation history to timestamped session log files.
