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

# Codex

> Connect ScrapeBadger to OpenAI Codex (CLI, IDE extension and Codex app) via MCP with OAuth.

# Codex

The Codex CLI, IDE extension and Codex app share one MCP configuration, so you only set ScrapeBadger up once.

## Setup

**1. Add the server**

```bash theme={null}
codex mcp add scrapebadger --url https://mcp.scrapebadger.com/mcp
```

**2. Sign in**

```bash theme={null}
codex mcp login scrapebadger
```

Your browser opens the ScrapeBadger sign-in and consent screen. Approve it, and Codex stores the token. You don't need an API key.

> [!NOTE]
>
> Prefer an API key? Export it and point Codex at the variable instead of running `codex mcp login`:
>
> ```bash theme={null}
> export SCRAPEBADGER_API_KEY=sb_live_...
> codex mcp add scrapebadger --url https://mcp.scrapebadger.com/mcp --bearer-token-env-var SCRAPEBADGER_API_KEY
> ```

## What you can ask

Once connected, Codex can call any of ScrapeBadger's tools while it works. Try:

> "Pull the latest 20 tweets from @openai and write them to tweets.json"

> "Scrape [https://example.com/pricing](https://example.com/pricing) and turn the plans into a TypeScript type"


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.