setup-gbrain

Install the gbrain CLI, bootstrap a PGLite or Supabase brain, and register MCP.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/AlejandroFigini/artist-portfolio --skill setup-gbrain-alejandrofigini
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: setup-gbrain
Source: https://github.com/AlejandroFigini/artist-portfolio/tree/main/.agent/skills/setup-gbrain
Command: npx skills add https://github.com/AlejandroFigini/artist-portfolio --skill setup-gbrain-alejandrofigini

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Streamlines the process of provisioning gbrain by installing the CLI, bootstrapping a local or remote brain, and registering MCP so the agent can securely persist context and trust policies across sessions.

Core Features & Use Cases

  • Install the gbrain CLI and verify environment prerequisites.
  • Initialize a local brain (PGLite) or link to a Supabase brain, plus register MCP for secure access.
  • Capture per-remote trust policy to enable consistent, safe agent memory across repos and machines.

Quick Start

Execute the setup to install the CLI, bootstrap a brain, and register MCP so the agent can begin using gbrain.

Frequently Asked Questions about setup-gbrain

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

FAQPage Schema
How do I install and configure persistent agent memory for Claude MCP?

To set up persistent agent memory, you need to install the gbrain CLI, bootstrap a local PGLite or remote Supabase brain, and register your MCP. This process securely persists context and trust policies across sessions.

What is the best way to bootstrap a local brain for secure agent context?

Bootstrapping a local brain uses PGLite to store agent context securely. You run the setup script to initialize the database and register MCP, enabling consistent memory across repositories and machines.

Can I use Supabase instead of PGLite for my agent memory database?

Yes, you can link to a Supabase brain instead of initializing a local PGLite instance. The setup adapts to your workflow by configuring either local or remote databases for secure, persistent agent memory.

Do I need Bash to set up per-remote trust policies for my agent?

Yes, Bash-powered scripts are required to execute the setup and capture per-remote trust policies. These policies ensure consistent, safe agent memory operations across different repositories and machines.

How does registering MCP enable secure agent memory across different repos?

Registering MCP connects your agent to the bootstrapped brain, applying per-remote trust policies to secure access. This enables consistent, safe agent memory and context persistence across multiple repositories and machines.