Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
### Adds daily deduplication and weekly reconciliation jobs for a self-cleaning knowledge graph. - Introduced a daily deduplication cron job to remove near-duplicate facts based on vector similarity, keeping the knowledge base compact and accurate. - Added a weekly reconciliation cron job to prune stale, low-confidence facts, clean orphaned entities, and consolidate duplicates for ongoing graph health. - Updated setup instructions and documentation to reflect that 5 cron jobs (not 2) are now required for complete self-improving and self-cleaning operation. - Clarified each job’s purpose and scheduling in the documentation for easier maintenance and troubleshooting.
What This Skill Does
A knowledge graph memory system built on SurrealDB that stores semantic facts, episodic task histories, and working memory state for AI agents. Four scheduled background jobs extract facts from memory files, discover relationships between them, deduplicate, and prune stale data automatically. Each agent gets an isolated memory pool, with optional shared global facts.
Unlike static context files, the knowledge graph grows and refines itself on a schedule, so accumulated facts become progressively more connected and relevant without manual curation.
When to use it
- Retrieving decisions made in past coding sessions by semantic query
- Persisting project knowledge across separate agent conversations
- Running independent memory pools for multiple specialized agents
- Auto-injecting relevant context into prompts before each agent turn
- Calibrating fact confidence based on whether tasks succeeded or failed
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: What did we decide about the authentication approach last month?
- 1Calls knowledge_search with the user query against the SurrealDB vector index
- 2Filters results to the active agent's fact pool plus any global-scoped facts
- 3Traverses relates_to edges to surface connected facts from the knowledge graph
- 4Formats matched facts and episodic memories via memory_inject for prompt injection
Agent responds with stored decisions, related facts, and past task learnings pulled from the knowledge graph
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.