By popoffvg
Persistent knowledge management — automatic insight extraction, context injection, and memory search/save across sessions.
This skill should be used when the user says "context check", "check for insights", "anything worth saving", "review session", or wants to analyze the current conversation for valuable learnings worth persisting — debugging discoveries, architectural decisions, corrections, workflow patterns.
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.
Search and retrieve insights from persistent memory. This skill should be used when the user says "context find", "recall", "what do I know about", "find my notes on", "search memory", "do I have anything on", or wants to retrieve previously saved knowledge. Always invoke to check memory before answering any topic-based question.
This skill should be used when the user says "research", "look into", "look up", "find out about", "search for", or asks about a topic needing investigation. Checks persistent memory first, performs web research if insufficient, persists results.
This skill should be used when the user says "context save", "remember this", "save this", "note this for later", "keep this", or wants to persist any knowledge, pattern, decision, or learning from the current session to persistent memory.
Admin access level
Server config contains admin-level keywords
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Persistent knowledge management plugin for Claude Code. Automatically extracts insights from sessions, injects relevant context on startup, and provides search/save commands for cross-session memory.
claude plugin install popoffvg/claude-plugin-memory-keeper
Create ~/.claude/memory-keeper.local.md with your configuration:
---
insights_root: ~/my/insights/path
log_level: DEBUG
---
This file is required. The plugin will not function without it — hooks will skip and skills will prompt you to create it.
mkdir -p ~/my/insights/path
Start a new Claude Code session. The SessionStart hook should load without errors. Run /context find — it should read your insights root.
| Setting | Required | Description |
|---|---|---|
insights_root | Yes | Root directory for all saved knowledge (e.g. ~/ctx/insights) |
log_level | No | Logging verbosity for SessionStart hook (DEBUG, INFO, WARN). Default: DEBUG |
| Command | Description |
|---|---|
/context find <query> | Search saved knowledge by keyword or topic |
/context find | Show full knowledge index |
/context save <note> | Persist a specific insight or fact |
/context check | Analyze current session for insights worth saving |
/context research <topic> | Search memory + web, then persist results |
/context scan [timeframe] | Scan recent session logs for missed insights (fallback for failed Stop hooks) |
/context done [task-name] | Complete active task — summarize insights and distribute to repo folders |
To enable automatic scanning for missed insights during a session, run:
/context scan cron
This sets up a session-scoped hourly cron that runs /context scan automatically. The cron lives only in the current Claude session and auto-expires after 3 days.
insights_root from ~/.claude/memory-keeper.local.md<insights_root>/_summary.md (or INDEX.md fallback) into session contextProject name is derived from the git repo name (git rev-parse --show-toplevel | basename), falling back to the working directory basename.
insights_root from settings — skips if not configured_tasks/pending.mdinsight | task | agent_edit | none<insights_root>/_tasks/<task-slug>/notes.md (tagged with repo name)<insights_root>/<project>/insights.md<insights_root>/_tasks/pending.md + creates _tasks/<task-slug>/ directory<insights_root>/claude-config/behavior.md/context done)_tasks/<task-slug>/notes.md<insights_root>/<repo>/insights.mddone in pending.mdmcp__qmd__search (keyword) on ctx collectionmcp__qmd__deep_search (semantic) if results are sparsez-core collection (Obsidian vault)<insights_root>/
INDEX.md # Global knowledge index
<project>/ # project = git repo name or folder name
_summary.md # Project summary (loaded on session start)
insights.md # Auto-captured insights + task summaries
_tasks/
pending.md # Task list (status: active | done)
<task-slug>/ # Per-task directory
notes.md # Insights collected during task (tagged with repo names)
claude-config/
behavior.md # Agent behavior corrections
MIT
npx claudepluginhub popoffvg/claude-plugin-memory-keeper --plugin memory-keeperWork lifecycle management — start, track, recall, complete, and PR creation for tasks.
Extended Second Brain that learns as you work in Claude Code
Personal AI Infrastructure — persistent memory, session continuity, and knowledge graph for Claude Code
Persistent memory across Claude Code sessions using Cognis
The memory layer Claude Code doesn't have. A persistent knowledge graph that learns from your conversations — your AI assistant never starts a session blind.
Persistent memory system for Claude Code with enforced context
Persistent memory for AI coding agents. Survives across sessions and compactions.