> ## Documentation Index
> Fetch the complete documentation index at: https://levers9.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Docs in your AI tools

> Connect these docs to Claude Code, Cursor or any MCP client, and ask questions against them

This site hosts a Model Context Protocol server at `https://levers9.mintlify.site/mcp`. Point an AI tool at it
and the tool can search these docs and read any page while it works on your code, so it answers "how do I mount
a volume" from the real reference rather than from memory.

**Claude Code**

```bash theme={null}
claude mcp add --transport http levers9-docs https://levers9.mintlify.site/mcp
```

Then ask, for example, "using the levers9 docs, write a queue with retries that reads from the datasets volume".

**Cursor**

Open the command palette, search for **Open MCP settings**, and add:

```json theme={null}
{
  "mcpServers": {
    "levers9-docs": { "url": "https://levers9.mintlify.site/mcp" }
  }
}
```

**Any MCP client**

The server speaks streamable HTTP with no authentication. Discovery: `https://levers9.mintlify.site/.well-known/mcp`.

The server exposes three tools: search the docs, read a page from the docs filesystem, and submit feedback.
It indexes every page in the navigation.

> [!NOTE]
>
> This is the docs server. For skills (a SKILL.md per feature) and an MCP server over your own workspace
> (deployments, tasks, logs, GPU capacity, deploy, invoke), see [Skills and MCP for AI agents](https://levers9.mintlify.site/guides/agent-skills.md).
> To deploy an MCP server of your own on the cluster, with tools that run in
> containers, see [MCP server](https://levers9.mintlify.site/sdk/mcp-server.md).


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