Skip to main content
GBrain
Menu
Get Started

GBrain connects Codex to the team's memory and applications

Codex is OpenAI's coding agent, in ChatGPT, the editor and the terminal.

Codex does real engineering work, end to end

  • Features, refactors and migrations

    From routine pull requests to the hardest problems.

  • Several at once

    Worktrees and cloud environments let Codex work in parallel.

  • Work in the background

    Scheduled issue triage, alert monitoring and CI.

From OpenAI's Codex page.

Codex starts from what the team already knows

What one AI saves, every other AI the team uses can read.

A workspace's memory in GBrain. Each note is a file the team owns.

Codex can use every application GBrain connects

Connect an account once in GBrain. The AI never holds the key.

See every application

The same memory and applications in every AI the team uses

ChatGPT
Codex
Claude Code

GBrain

one memory

Cursor
OpenClaw
The terminal
Codex on the dark line, and a few of the AIs that connect the same way. Each one opens its own page.

See every AI GBrain works with

Codex connects with a few lines of settings

One address for every AI. The team decides what each one may do.

  1. In the GBrain workspace, make an access token for Codex. GBrain shows it once.

  2. Add GBrain to ~/.codex/config.toml, the settings file Codex reads in the terminal and in the ChatGPT desktop app.

    [mcp_servers.gbrain]
url = "https://gbrain.io/mcp"

[mcp_servers.gbrain.http_headers]
Authorization = "Bearer <token>"

Start a GBrain workspace

Sign in and name the workspace. Then connect Codex with the steps above.

Sign in