cheese-brain
par mhugo22
Search & ResearchRésumé
DuckDB-powered knowledge management system for fast retrieval across 22+ entity types (projects, contacts, tools.
Description
Ce que fait ce skill
DuckDB-powered knowledge base for storing and retrieving entities across 22+ categories including projects, contacts, tools, workflows, and decisions. Supports sub-millisecond keyword search and BM25 full-text search with relevance ranking. Persists context across sessions via a local database.
Unlike in-context memory that resets each session, Cheese Brain persists structured knowledge in a local DuckDB database queryable in under 1ms.
Quand l'utiliser
- Look up a past project's repo path and run command
- Find a contact's calendar feed URL
- Retrieve the rationale behind a past technical decision
- Search for a workflow's step-by-step restore procedure
- List all shipped projects filtered by tag
Exemple de workflow
User asks: How do I restore my config backup?
- 1
Run
cheese-brain search "backup config"to find relevant entities - 2
Review returned table of matches (backup script, restore script, workflow)
- 3
Run
cheese-brain get <entity-id>on the restore workflow entity - 4
Read the step-by-step instructions and related tool references from the data field
Full restore workflow with tool paths, commands, and related entity links
Prérequis
- cheese-brain Python package installed from https://github.com/mhugo22/cheese-brain
Installation
clawhub install mhugo22/cheese-brainExécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.