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

# Introduction

> Connect AI agents to Footstep

<Frame>
  <img src="https://mintcdn.com/footstep/QNknRZ9DgHsEJ89b/images/mcp-introduction.webp?fit=max&auto=format&n=QNknRZ9DgHsEJ89b&q=85&s=4c2f4f027394a98b48f778f78e7d346b" alt="Footstep MCP server" width="1408" height="768" data-path="images/mcp-introduction.webp" />
</Frame>

The Footstep MCP server lets AI agents call routing and geocoding tools directly. It implements the [Model Context Protocol](https://modelcontextprotocol.io), the open standard for connecting LLMs to external tools.

## Connect

Point your MCP client at the hosted server. No installation required.

```
https://mcp.footstep.ai
```

Pass your API key as a header or in the URL, depending on your client. See the [configuration guide](/mcp/configuration) for setup examples.

## Supported clients

MCP is an open standard. The Footstep MCP server works with any compatible client:

| Client                 | Type                      |
| ---------------------- | ------------------------- |
| Claude Desktop / Code  | AI assistant / CLI        |
| Cursor                 | IDE                       |
| Windsurf               | IDE                       |
| Cline / Continue       | VS Code extensions        |
| OpenAI Agents SDK      | Python framework          |
| Gemini (via LangChain) | Python framework          |
| Vercel AI SDK          | TypeScript framework      |
| Any MCP client         | Custom agents, frameworks |

## Next steps

<Columns cols={2}>
  <Card title="Configuration" icon="gear" href="/mcp/configuration">
    Set up the MCP server with your client.
  </Card>

  <Card title="Tools" icon="wrench" href="/mcp/tools">
    See what tools are available to your agent.
  </Card>
</Columns>