Codex CLI

Connect OpenAI Codex CLI to the CoreClaw MCP Server so you can search for scrapers, run them, and retrieve data directly from your terminal conversations.

Prerequisites

Codex CLI installed (npm install -g @openai/codex)
A CoreClaw account with an API key — get it from Console → Settings → API & Integrations

Configuration

Codex CLI supports MCP via Streamable HTTP transport. You can configure it via a JSON config file.

Step 1: Create or edit the MCP configuration file

Create a file named codex-mcp.json in your project directory (or anywhere you prefer):

{
  "mcpServers": {
    "coreclaw": {
      "url": "https://mcp.coreclaw.com/mcp",
      "headers": {
        "api-key": "scraper_api_YOUR_KEY_HERE"
      }
    }
  }
}

Replace scraper_api_YOUR_KEY_HERE with your actual CoreClaw API key.

Step 2: Launch Codex with the MCP config

codex --mcp-config codex-mcp.json

Alternatively, you can place the config at ~/.config/codex/mcp.json (macOS/Linux) or %APPDATA%\codex\mcp.json (Windows) for automatic loading.

Verify the connection

Start Codex with the MCP configuration
Ask Codex: “Search for Amazon scrapers on CoreClaw”
Codex should invoke search_scrapers and return results

Example conversation

Once connected, you can perform scraping tasks directly from your terminal:

You: Find a Twitter scraper and extract the latest tweets from @elonmusk

Codex: I’ll search for a Twitter scraper and run it for you. [Calls search_scrapers → get_scraper_details → run_scraper → polls status → returns results]

Troubleshooting

MCP config not loading

Verify the JSON file path is correct when using --mcp-config
Ensure the JSON is valid (no trailing commas, proper quotes)
Check Codex CLI version — MCP support requires recent versions

Authentication errors

Ensure the api-key header value matches your CoreClaw API key exactly
Verify your key is active in the Console

Tools not available

Confirm Codex CLI supports Streamable HTTP transport
Try restarting the Codex session after config changes