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

# Docs MCP Server

> Connect Claude, ChatGPT, Cursor, VS Code and other AI tools to the Junction docs through our MCP server, so they can search and read the docs while you build.

The Junction docs are available as an [MCP](https://modelcontextprotocol.io) server. Connect it to your AI assistant or coding agent and it can search and read these docs directly, instead of relying on what it remembers about the Junction API.

The server URL is:

```
https://docs.junction.com/mcp
```

It's read-only, covers the public docs and API reference, and needs no API key.

## Setup

**Claude Code**

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

**Claude.ai / Claude Desktop**

1. Go to **Settings → Connectors**.
2. Click **Add custom connector**.
3. Name it `Junction Docs` and set the URL to `https://docs.junction.com/mcp`.

**Codex**

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

Or add it to `~/.codex/config.toml`:

```toml theme={null}
[mcp_servers.junction-docs]
url = "https://docs.junction.com/mcp"
```

**ChatGPT**

1. Go to **Settings → Apps & Connectors → Advanced settings** and turn on **Developer mode**.
2. Back in **Apps & Connectors**, click **Create**.
3. Name it `Junction Docs`, set the URL to `https://docs.junction.com/mcp` and set authentication to **No authentication**.

Custom connectors depend on your ChatGPT plan and workspace settings.

**OpenAI API**

Pass the server as a remote MCP tool in the Responses API:

```python theme={null}
from openai import OpenAI

client = OpenAI()

response = client.responses.create(
    model="gpt-5",
    tools=[
        {
            "type": "mcp",
            "server_label": "junction_docs",
            "server_url": "https://docs.junction.com/mcp",
            "require_approval": "never",
        }
    ],
    input="How do I order an at-home phlebotomy test with Junction?",
)

print(response.output_text)
```

**Cursor**

Add to `~/.cursor/mcp.json` (or `.cursor/mcp.json` in your project):

```json theme={null}
{
  "mcpServers": {
    "junction-docs": {
      "url": "https://docs.junction.com/mcp"
    }
  }
}
```

**VS Code**

Add to `.vscode/mcp.json` in your project:

```json theme={null}
{
  "servers": {
    "junction-docs": {
      "type": "http",
      "url": "https://docs.junction.com/mcp"
    }
  }
}
```

Any other MCP client that supports Streamable HTTP can connect using the same URL.

## Example prompts

Once connected, ask your assistant things like:

* "Using the Junction docs, write a function that creates a user and orders a walk-in lab test."
* "Which webhook events does Junction send when a lab result is ready?"
* "What's the difference between Junction's at-home phlebotomy service tiers?"
* "How do I sync Apple Health data using the Junction Mobile SDK?"

If your assistant doesn't pick up the docs on its own, mention "Junction docs" in the prompt.

## Reporting docs issues

The server also lets your assistant submit feedback when it finds a page that's wrong, outdated or unclear. You can also reach us through [Getting Support](https://docs.junction.com/home/getting-support.md).


## Related topics

- [Rate Limiting](https://docs.junction.com/api-details/rate_limiting.md)
- [Link Demo provider](https://docs.junction.com/api-reference/link/link-demo-provider.md)
- [Fitbit](https://docs.junction.com/wearables/guides/fitbit.md)
- [Update user](https://docs.junction.com/api-reference/user/patch-user.md)
- [API](https://docs.junction.com/changelog/wearables/api.md)


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