---
title: AI agents
meta:
   description: "Interact with Tinybird via AI agents."
---

# AI agents

Apart from manually creating your projects and integrating them in your app, you can use AI agents to build and consume Tinybird projects.

## Build Tinybird projects with agents

Use agents to generate, refine, and test Tinybird projects as code.

- [Agent skills](/classic/analytics-agents/agent-skills) add Tinybird-specific guidance to your coding agents.

## Query Tinybird data with agents

Every Tinybird workspace is a fully managed remote MCP server you can connect to.

The Tinybird remote MCP server enables AI agents to connect directly to your workspace to use endpoints as tools or execute queries. The [Model Context Protocol](https://modelcontextprotocol.io) gives AI assistants access to your analytics APIs, data sources, and endpoints through a standardized interface.

Use this integration when you want AI agents to autonomously query your data, call your analytics API Endpoints, or build data-based apps without requiring manual API integration.

{% callout type="info" %}
The Tinybird MCP server only supports Streamable HTTP as the transport protocol. If your MCP client doesn't support it, use the `mcp-remote` package as a bridge.
{% /callout %}

For practical examples and guardrails, see [MCP examples](/classic/analytics-agents/mcp-server-snippets) and [MCP best practices](/classic/analytics-agents/best-practices).


## Next steps

- Explore the [MCP server](/classic/analytics-agents/mcp) tools.
- Review the [example snippets](/classic/analytics-agents/mcp-server-snippets).
- Follow [best practices](/classic/analytics-agents/best-practices) for using the MCP server with AI agents.