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

# Claude Agent SDK

> Integrate Caylex with the Claude Agent SDK for Python.

This guide shows you how to connect a Claude-powered agent to Caylex using the Claude Agent SDK.

## Prerequisites

* Python 3.10+
* A Caylex Navigator Instance with an API key ([create one here](/platform/navigators))
* Users authenticated with your project's servers ([set up auth links](/auth/auth-links))

## Installation

```bash theme={null}
pip install claude-agent-sdk
```

## Full example

```python main.py theme={null}
import asyncio
import os

from claude_agent_sdk import query, ClaudeAgentOptions, MCPServerStreamableHttp


async def main() -> None:
    api_key = os.environ["CAYLEX_API_KEY"]
    user_email = os.environ["CAYLEX_USER_EMAIL"]

    # Connect to Caylex MCP
    caylex_server = MCPServerStreamableHttp(
        name="Caylex",
        url="https://navigator.caylex.ai/mcp",
        headers={
            "x-api-key": api_key,
            "x-user-email": user_email,
        },
    )

    # Run the agent with Caylex tools
    async for message in query(
        prompt="Check my Linear issues, find any blockers, and post updates to the relevant Slack channels.",
        options=ClaudeAgentOptions(
            model="claude-opus-4-5",
            mcp_servers=[caylex_server],
        ),
    ):
        if hasattr(message, "result"):
            print(message.result)


asyncio.run(main())
```

## Step-by-step

<Steps>
  <Step title="Set environment variables">
    ```bash theme={null}
    export CAYLEX_API_KEY="ck_abc123.your-secret-key"
    export CAYLEX_USER_EMAIL="user@example.com"
    export ANTHROPIC_API_KEY="your-anthropic-key"
    ```
  </Step>

  <Step title="Create the MCP server connection">
    Use `MCPServerStreamableHttp` with the Caylex endpoint and authentication headers.

    ```python theme={null}
    caylex_server = MCPServerStreamableHttp(
        name="Caylex",
        url="https://navigator.caylex.ai/mcp",
        headers={
            "x-api-key": api_key,
            "x-user-email": user_email,
        },
    )
    ```
  </Step>

  <Step title="Run the agent with streaming">
    Use the `query` function with `ClaudeAgentOptions` to run the agent. The response streams as the agent works through its reasoning and tool calls.

    ```python theme={null}
    async for message in query(
        prompt="Your query here.",
        options=ClaudeAgentOptions(
            model="claude-opus-4-5",
            mcp_servers=[caylex_server],
        ),
    ):
        if hasattr(message, "result"):
            print(message.result)
    ```
  </Step>
</Steps>

## Further reading

* [Claude Agent SDK MCP documentation](https://platform.claude.com/docs/en/agent-sdk/mcp)
* [Connecting Your Agent](/integration/connecting) — details on how the Navigator works
* [Server Authentication](/auth/server-authentication) for user-level vs project-level auth