Stats
Actions
Tags
From memory-os
Use only after the user confirms they want a knowledge layer. Builds and maintains a typed-edge markdown wiki from raw captures.
How this skill is triggered — by the user, by Claude, or both
Slash command
/memory-os:knowledge-graphThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Structure under `.memory/knowledge/`:
Structure under .memory/knowledge/:
raw/ — unedited captures (never reorganised).wiki/ — organised nodes the agent writes; never hand-edited by the user.outputs/ — answers/briefings generated on request.index.md — written first; a short summary line per wiki node.Each wiki/*.md node carries frontmatter:
---
type: <concept|method|source|project>
relates-to: [other-node]
supersedes: [old-node]
sourced-from: [raw/<file>.md]
contradicts: [node]
---
Process:
raw/. Write index.md first, then one node per major topic, then fill typed edges..memory/knowledge/ as a read-only graph viewer. The agent remains the librarian; the user does not hand-organise.npx claudepluginhub tpop78/memory-os --plugin memory-osProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
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.