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

# Documentation MCP

> Connect TheRundown API docs to Codex, Claude, Cursor, and other AI assistants.

TheRundown provides a [Model Context Protocol (MCP)](https://modelcontextprotocol.io) server that lets AI assistants search our documentation directly. Ask questions about endpoints, query parameters, response fields, market IDs, billing headers, and integration patterns without leaving your editor.

<Info>
  This MCP server is **documentation-only**. It helps your assistant understand the API, but it does not stream live sports data or make authenticated API calls on your behalf. For live data, your agent still needs to call the real API using your own TheRundown key.
</Info>

For authenticated sports and odds tools, run the [local data MCP scaffold](/data-mcp) with your own API key. Its [official public source repository](https://github.com/TheRundown/data-mcp) and versioned 0.2.1 source bundle provide six read-only tools: `list_sports`, `list_affiliates`, `list_markets`, `list_events`, `get_main_lines`, and `list_futures`. The local server also exposes `therundown://brief` for its scoped-data rules and first conversation. It is separate from this documentation server, which does not gain live-data tools when you install it. A hosted data MCP is not currently available.

## Server URL

```
https://docs.therundown.io/mcp
```

## Codex

Add the documentation server from your terminal:

```bash theme={null}
codex mcp add therundown-docs --url https://docs.therundown.io/mcp
```

This server searches TheRundown documentation only. It does not call the Product API or return live sports data. See [OpenAI's Codex MCP documentation](https://developers.openai.com/codex/mcp) for client configuration details.

## Claude Code

Add the server from your terminal:

```bash theme={null}
claude mcp add --transport http therundown https://docs.therundown.io/mcp
```

## Cursor

Open the command palette and search "Open MCP settings", then add to your `mcp.json`:

```json theme={null}
{
  "mcpServers": {
    "therundown": {
      "url": "https://docs.therundown.io/mcp"
    }
  }
}
```

## VS Code

Create or update `.vscode/mcp.json` in your project:

```json theme={null}
{
  "servers": {
    "therundown": {
      "type": "http",
      "url": "https://docs.therundown.io/mcp"
    }
  }
}
```

## Claude Desktop

Add this to your Claude Desktop config file:

* **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
* **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`

```json theme={null}
{
  "mcpServers": {
    "therundown": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://docs.therundown.io/mcp"
      ]
    }
  }
}
```

Restart Claude Desktop after saving.

## Windsurf

Add to your `~/.codeium/windsurf/mcp_config.json`:

```json theme={null}
{
  "mcpServers": {
    "therundown": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://docs.therundown.io/mcp"
      ]
    }
  }
}
```

## What you can ask

### Prepare the local data MCP before a data request

The documentation MCP cannot execute Product API requests. First install and connect the [local data MCP scaffold](/data-mcp) with your own API key. Then use this conversation with the local data MCP:

```text theme={null}
Use TheRundown to list current sports and affiliates. Find MLB (sport 3).
For today's UTC date, list events with market_ids [1,2,3]
and affiliate_ids [19,23]. Select an event ID from that response and
call get_main_lines with the same filters.
Show the source URL, each book's line value and price updated_at,
and the returned usage headers. Explain empty results without inventing odds.
```

The best workflow is:

1. Ask MCP which endpoint, filters, and market IDs you need
2. Ask it for a code example in your language
3. Have your agent call the real API with your own key
4. Use the returned billing headers to keep usage under control

Once connected, you can ask your AI assistant questions like:

* "What is the cheapest way to monitor NBA moneyline, spread, and total on DraftKings and FanDuel?"
* "Should I use market delta or WebSocket for live line movement?"
* "What headers tell me remaining data points on my key?"
* "What are the market IDs for player props in NBA?"
* "Show me a Python example for `GET /api/v2/sports/{sportID}/events/{date}` with `market_ids` and `affiliate_ids` filters"
* "What does the `0.0001` sentinel value mean and how should I handle it?"

The MCP server searches across the API reference, guides, SDK docs, and reference tables to find relevant answers and examples.