From memory-keeper
This skill should be used when the user says "work done", "context done", "task done", "finish task", "complete task", "close task", or indicates they have finished working on an active task. Summarizes task insights and distributes them to each repo's insight folder.
How this skill is triggered — by the user, by Claude, or both
Slash command
/memory-keeper:context-doneThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Finalize an active task: summarize collected insights and distribute to each related repo's insight folder.
Finalize an active task: summarize collected insights and distribute to each related repo's insight folder.
/context done — complete the current active task
/context done <task-name> — complete a specific task
Read insights_root from ~/.claude/memory-keeper.local.md YAML frontmatter. If the file is missing, stop and ask the user to create it with the required settings (see plugin README).
<insights_root>/_tasks/pending.md<insights_root>/_tasks/<task-slug>/notes.md_(repo: <name>)_)pending.mdCreate a concise task summary (3-5 sentences max):
For each unique repo in the task's Repos list:
<insights_root>/<repo>/insights.md first<insights_root>/<repo>/insights.md:
## <task-title> (task summary) — YYYY-MM-DD HH:MM
<summary focusing on insights relevant to this specific repo>
_(from task: <task-slug>)_
<insights_root>/_tasks/pending.md, change the task's status from active to done<insights_root>/_tasks/<task-slug>/ as-is (archive, don't delete)Report:
npx claudepluginhub popoffvg/claude-plugin-memory-keeper --plugin memory-keeperSync tracking documents based on current conversation results. Updates subtask, progress, findings, task_plan, project CLAUDE.md. Use when finishing a task or reaching a milestone.
Creates and manages task documentation including implementation plans, task IDs, and archiving completed work. Useful when starting a new feature or finishing one.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.