setup

Provision GBrain databases and inject the brain-first lookup protocol into AGENTS.md.

1|Updated May 12, 2026
One-click install
npx skills add https://github.com/ltdrew/Agent-brain-2 --skill setup-ltdrew
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: setup
Source: https://github.com/ltdrew/Agent-brain-2/tree/main/skills/setup
Command: npx skills add https://github.com/ltdrew/Agent-brain-2 --skill setup-ltdrew

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Sets up GBrain from scratch by provisioning the database (Supabase or PGLite), injecting the brain-first lookup protocol into AGENTS.md, and performing the initial import, enabling a working brain in minutes.

Core Features & Use Cases

  • Auto-provisioning through Supabase or BYO Postgres with pgvector.
  • Injects the brain-first lookup protocol into AGENTS.md to enable fast, structured lookups.
  • Verifies setup with gbrain doctor --json and records state in ~/.gbrain/update-state.json for future upgrades.

Quick Start

Run the setup to provision the brain, inject the brain-first protocol, and perform the first import.

Frequently Asked Questions about setup

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

FAQPage Schema
How do I provision a Supabase database for GBrain setup?

GBrain setup automates Supabase database provisioning by initializing the required schema, handling the secure database connection, and verifying the deployment with gbrain doctor --json to ensure a working brain in minutes.

Can I use a Bring Your Own Postgres instance with pgvector for GBrain?

Yes, GBrain setup supports BYO Postgres with pgvector. You can use the auto-provisioning options to connect your existing Postgres database, initialize the brain schema, and perform the first import.

How does GBrain inject the brain-first lookup protocol into AGENTS.md?

GBrain setup injects the brain-first lookup protocol directly into your AGENTS.md file during initialization, enabling fast, structured lookups for your agents and recording the schema state in ~/.gbrain/update-state.json.

What is the best way to configure live-sync for a new GBrain deployment?

The best way to configure live-sync is running the GBrain setup, which handles initial provisioning, performs the first brain import, and establishes the live-sync configuration automatically for new deployments.

Do I need PGLite for local GBrain initialization?

PGLite is one of the auto-provisioning options available for local GBrain initialization. You can use PGLite or Supabase to provision the brain, inject the AGENTS.md protocol, and verify the setup.