gbrain

Builds a PostgreSQL-native hybrid RAG knowledge brain from Markdown notes and conversations.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/escotilha/claude-public --skill gbrain
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gbrain
Source: https://github.com/escotilha/claude-public/tree/main/skills/gbrain
Command: npx skills add https://github.com/escotilha/claude-public --skill gbrain

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of scattered personal and team knowledge by compiling your notes into a searchable knowledge base that combines keyword search and semantic retrieval, so you can answer questions faster with better context.

Core Features & Use Cases

  • Hybrid RAG knowledge brain: stores compiled truth and append-only timelines in PostgreSQL to support accurate recall and change history.
  • Typed-graph entity linking: connects people, companies, concepts, and meetings with relationship types so queries stay grounded and navigable.
  • Operational workflow commands: supports setup, markdown import/sync, hybrid query, ingestion of new signals, and ongoing health/stats checks for continued quality.

Use case: You ingest meeting notes and scattered markdown files, then later ask what is known about a specific company or person and receive fused keyword+vector results grounded in stored pages and timelines.

Quick Start

Use the gbrain skill to query what you know about an entity by running the command: /gbrain query "What do we know about Company X?"

Frequently Asked Questions about gbrain

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How does hybrid RAG retrieval improve personal knowledge management?

Hybrid RAG retrieval enhances personal knowledge management by fusing tsvector keyword search with vector embeddings in Postgres, using RRF fusion to return accurate, context-rich results from compiled notes.

How do I import and sync markdown notes into a searchable knowledge base?

You import and sync markdown notes into a searchable knowledge base by using operational workflow commands that compile files into PostgreSQL, creating searchable pages with append-only timelines and typed entity relationships.

Does this knowledge brain require PostgreSQL to store embeddings and timelines?

Yes, this knowledge brain requires PostgreSQL storage to handle 1536-dimension embeddings, keyword search via tsvector, vector search, and append-only timelines for reliable hybrid RAG retrieval.

What is the best way to answer entity-focused questions from scattered meeting notes?

The best way to answer entity-focused questions from scattered meeting notes is using multi-query expansion and timeline-aware context against a typed entity graph, retrieving grounded facts about specific people, companies, and concepts.

Can I link people, companies, and concepts in a personal knowledge graph?

Yes, you can link people, companies, and concepts using typed-graph entity linking, which connects entities with specific relationship types so queries remain grounded and navigable within your knowledge base.