Skip to content
Agentic Levels
Start hereLevelsTracksPromptsAbout
Get my levelGet my level

  • Start hereYour first useful task with AI.
  • Level quiz8 questions. No AI needed.
  • Fluency ScoreYour Level and a score, from your own AI.

  • LevelsThe 11 levels, in order.
  • LessonsShort lessons, filter by level.
  • TracksGuided paths by tool, role or goal.
  • Set up your AI tool
  • Starter packFree files that help AI remember.

  • PromptsReady prompts to copy and adapt.
  • ChatGPT vs Claude vs GeminiPick the right one for the job.
  • How-tos
  • ReferenceCommands for Claude Code, Codex, Antigravity.
  • ResourcesHand-picked reading and tools.

  • Your progress
  • Your next moveAn interview that picks your next skill.
  • Your next lessonsThe next lessons at your level.
  • Updates

  • Pricing
  • About
  • FAQ & Feedback
  • Work with us
  • How we handle your data
Sign inPreferences
AboutPricingHow-tosPromptsStarter packMy stackHow we handle your data

© 2026 Fuentes Studio

PrivacyTermsRefundsAccessibilityAI content policyCredits
yourCouncil
Ready to help
✦

What do you want to understand?

Ask anything about what you're learning.

←How-to guides
  1. Resources
  2. ›How-to guides

How to use gbrain

Give your AI agent a memory that remembers your whole project and tells you what it does not know yet.

Levels 6 to 8On GitHub

Your coding agent is amnesiac about everything that is not code. Every session it forgets the decisions, the people, the half-finished threads. Plain search makes it worse: ask a question and you get back ten pages you still have to read yourself. GBrain is the memory layer that fixes both. It synthesizes a cited answer instead of a page list, wires the people and projects it sees into a knowledge graph on its own, and ends every answer with an honest note on what it does not know yet. It runs on your own machine, your own database, your own keys, and connects to Claude Code as one more tool your agent can reach for.

Before you start

This one is for the run-your-own-agents crowd. You should be comfortable running Claude Code from a terminal and have bun installed. No server, no cloud account, and no API-key wrangling to get started.

  1. 1Let your agent install it
  2. 2Or wire it in yourself with two commands
  3. 3Feed it your project context
  4. 4Ask a real question and read the gap analysis
  5. 5Let it keep itself sharp

Steps

01

Let your agent install it

GBrain is built to be installed by an agent, not by hand. Open Claude Code in any project and paste the line below. Claude pulls GBrain's install protocol, runs it, and asks you for anything it needs as it goes. You answer questions, it does the work. About 30 minutes to a fully working brain, and the local database is ready in seconds.

Paste into Claude Code
Retrieve and follow the instructions at: https://raw.githubusercontent.com/garrytan/gbrain/master/INSTALL_FOR_AGENTS.md
02

Or wire it in yourself with two commands

Want to do it by hand instead? Spin up a local brain and connect it to Claude Code in two commands. The first builds a local database in about two seconds with no Docker and no server. The second registers GBrain as an MCP tool your agent can call. Zero server, zero token, zero tunnel. Swap claude for codex if you run Codex instead.

Run in your terminal
gbrain init --pglite claude mcp add gbrain -- gbrain serve
03

Feed it your project context

GBrain earns its keep on the stuff that lives outside your repo: meeting notes, decisions, the why behind a refactor, who owns what. Hand that context to your agent and let GBrain ingest it. As it writes each page it extracts the people, companies, and projects and links them with typed edges automatically, no extra LLM calls. That graph is what lets it answer questions plain search cannot.

Ask your agent
Ingest this into gbrain as project context. Pull out the people, decisions, and open threads so I can query them later. [YOUR_CONTEXT]
[YOUR_CONTEXT]Notes, decisions, or threads from outside your code e.g. Standup 2026-06-10: agreed to drop the Redis cache for now. Maya owns the migration, blocked on the schema review from the platform team. Revisit after launch.
04

Ask a real question and read the gap analysis

This is the part that changes how you work. Ask a question that spans several notes. GBrain returns a synthesized answer with a source behind each claim, then ends with what it does not know yet. That closing note is the point: it tells you where to go look instead of letting your agent confidently fill the gap with a guess.

Ask your agent
Using gbrain, answer this. Give me a synthesized answer with sources, then tell me plainly what the brain does not know yet about it. [YOUR_QUESTION]
[YOUR_QUESTION]A question that spans more than one note e.g. What is blocking the cache migration and who do I chase to unblock it?
05

Let it keep itself sharp

GBrain is a daemon, not a one-shot tool. It runs a dream cycle that ingests new material, enriches the people and companies it already knows, fixes its own citations, and consolidates memory while you are not looking. Set it up once and the brain gets sharper on its own. Everything stays on your hardware and your keys the whole time.

Ask your agent
Set up gbrain's dream cycle so it keeps ingesting, enriching, and consolidating on its own in the background. Confirm it is running locally on my machine.

How do you know it worked?

  • gbrain shows up as an MCP tool in Claude Code, backed by a local database on your own machine.
  • After you feed it project context, your agent confirms the entities and relationships were extracted into the graph.
  • A multi-note question returns a synthesized answer with a source behind each claim, not a list of pages.
  • Every answer ends with an explicit note on what the brain does not know yet.
  • The dream cycle is running in the background, enriching and consolidating without you prompting it.

If something looks off

The gap note is the feature, not a disclaimer
The 'here is what I do not know yet' line at the end of each answer is the whole reason to use GBrain over plain search. Read it every time. It is telling you where your agent would otherwise guess.
It only knows what you feed it
GBrain is not connected to your email or Slack by default. If an answer feels stale, it usually means an update came through a channel the brain never saw. The gap note will often say so. Feed it the missing context and ask again.
Local and yours by default
The two-command path runs entirely on your machine, your database, your keys. Nothing leaves your hardware. If you later point laptop agents at a shared remote brain instead, that is when access scoping matters, so decide who sees what before you go multi-person.

Skills & Commands Reference

Once the MCP tool is wired in, these are the calls you reach for most. Your agent invokes them; you just ask in plain language.

Command / SkillWhat it doesWhen to use it
gbrain thinkadvancedReturns a synthesized, cited answer plus a note on what it does not know yet, not a page list.Any question that spans more than one note or needs an actual answer.
gbrain init --pgliteBuilds a local brain in about two seconds with no server and no Docker.First-time setup when you want everything on your own machine.
gbrain serveadvancedRuns the brain as an MCP server your coding agent connects to.Wiring GBrain into Claude Code or Codex as a tool.

See also

  • Hermes Agent

What you'll do, 5 steps

  1. 1Let your agent install it
  2. 2Or wire it in yourself with two commands
  3. 3Feed it your project context
  4. 4Ask a real question and read the gap analysis
  5. 5Let it keep itself sharp

Was this helpful?