> ## Documentation Index
> Fetch the complete documentation index at: https://scoutlayer.io/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Codex

> Connect Codex CLI, the IDE extension, or ChatGPT desktop to ScoutLayer's remote MCP server for live web data.

<Note>Works with Codex CLI, the Codex IDE extension, or ChatGPT desktop.</Note>

<Steps>
  <Step title="Add the server">
    Add ScoutLayer as a Streamable HTTP server in `config.toml`:

    ```toml ~/.codex/config.toml theme={null}
    [mcp_servers.scoutlayer]
    url = "https://scoutlayer.io/mcp"
    ```
  </Step>

  <Step title="Restart">
    Restart the client after saving the configuration.
  </Step>

  <Step title="Authenticate">
    Select **Authenticate**, or run `codex mcp login scoutlayer` from the CLI.
  </Step>

  <Step title="Confirm">
    Open `/mcp` to confirm the ScoutLayer tools are available.
  </Step>
</Steps>

## Try it

Once connected, ask Codex a natural-language question and it calls ScoutLayer for you:

> Find the best contact email and LinkedIn for the company behind stripe.com.

Codex gets live data straight from the company's own site — public emails, phones, social profiles, and company details — each fact linked back to the page it came from, so it can cite a real source instead of guessing.

See OpenAI's [Codex MCP documentation](https://developers.openai.com/codex/mcp/) for more detail.