Skip to main content
traxy connects to Codex through a custom Streamable HTTP MCP server. No API key is required — the connection uses secure browser OAuth.

​
1 minute guide on how to connect traxy x ChatGPT/Codex!

1

Open the Codex integration in traxy

In the traxy sidebar, go to Integrations, find the Codex card, and click Configure.On the integration page, make sure the App & browser tab is selected (use the Terminal tab instead if you’re connecting from the Codex CLI).
2

Copy the traxy MCP URL

Click Copy MCP URL to copy the connector URL:
3

Open Codex settings

In the Codex app, open Settings, then under Integrations select Plugins.
4

Add the MCP server

On the Plugins page, switch to the MCPs tab, then click Add → Add MCP server.In the Connect to a custom MCP form:
  1. Enter traxy.ai as the Name.
  2. Set Type to Streamable HTTP.
  3. Paste the MCP URL — https://api.traxy.ai/api/v1/mcp — into the URL field.
  4. Leave Bearer token env var and Headers blank.
  5. Click Save.
5

Authenticate the connection

Back on the MCPs tab, traxy-ai now appears in your Servers list with an Authenticate button. Click it.You’ll be redirected to traxy’s authorization page:
  • Select the Workspace you want Codex to access.
  • Review the Requested access scopes (leads, analytics, ICP & Signal Brief, watchlist, integrations, agent tools, and more).
  • Click Approve.
6

Confirm the connection

You’ll be taken back to Codex settings, where traxy-ai now appears in the Servers list, toggled on with no pending authentication step — the connection is complete.

​
Next steps

Once connected, you can ask Codex things like “What’s my traxy lead pipeline look like this week?” or “Pull my top-performing posts from traxy” — Codex will call the relevant traxy tools directly. To disconnect at any time, go to Settings → Plugins → MCPs, find traxy-ai, and toggle it off or remove it via the gear icon.