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

# API and MCP

> Choose REST reads or an OAuth-authorized MCP connection.

Use the public REST API for application reads. Use MCP when an AI client should read workspace data and perform supported, authorized actions.

| Connection | Authentication | Workspace selection |
| - | - | - |
| REST and audience analytics | API key | The key's workspace |
| MCP | OAuth access token | The authorized default or an explicit `workspaceId` |

> **[Make your first REST request](https://docs.traxy.ai/api-reference/quickstart.md)**
>
> Create a key and read a small lead sample.

> **[Connect an AI client](https://docs.traxy.ai/integrations/ai-clients.md)**
>
> Authorize workspaces and supported actions through OAuth.

> **[MCP tool reference](https://docs.traxy.ai/api-reference/mcp/tools.md)**
>
> Review public tools and action requirements.

> **[Handle results and errors](https://docs.traxy.ai/api-reference/pagination-and-errors.md)**
>
> Use continuation cursors and distinguish error types.

## Endpoints

REST base URL:

```text theme={null}
https://api.traxy.ai/api/v1
```

MCP endpoint:

```text theme={null}
https://api.traxy.ai/api/v1/mcp
```

Endpoint reference pages describe the public contract. The in-app Agent has additional capabilities that are not necessarily available through public MCP. Use the public tool catalog rather than assuming every app action has an MCP equivalent.


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