Tinybird MCP Server

Data Science ToolsPython

Query and interact with Tinybird workspaces from any MCP client

Available Tools

list-data-sources

Lists all Data Sources in the Tinybird Workspace

list-pipes

Lists all Pipe Endpoints in the Tinybird Workspace

get-data-source

Gets the information of a Data Source given its name, including the schema

get-pipe

Gets the information of a Pipe Endpoint given its name, including its nodes and SQL transformation

request-pipe-data

Requests data from a Pipe Endpoint via an HTTP request, with optional parameters to filter the analytical data

run-select-query

Runs a select query over a Data Source to extract insights

append-insight

Adds a new business insight to the memo resource

llms-tinybird-docs

Provides access to the complete Tinybird product documentation

save-event

Sends an event to a Tinybird Data Source, useful for saving user-generated prompts

analyze-pipe

Uses the Tinybird analyze API to run a ClickHouse explain on the Pipe Endpoint query and suggest optimizations

push-datafile

Creates a remote Data Source or Pipe in the Tinybird Workspace from a local datafile

The Tinybird MCP server enables seamless interaction with Tinybird workspaces directly from MCP-compatible AI assistants. It allows you to query data sources, access API endpoints, and push data files without leaving your AI assistant interface. This integration bridges the gap between conversational AI and real-time analytics, making it possible to explore and analyze your data through natural language.

Overview

The Tinybird MCP server provides a powerful interface between AI assistants and your Tinybird workspace. It enables you to perform data analysis, run queries, and manage data directly through conversation with your AI assistant.

Prerequisites

Before setting up the Tinybird MCP server, you'll need:

A Tinybird account and workspace
An MCP-compatible AI assistant (like Claude Desktop)
uv package manager installed

Installation

Using MCP Package Managers

Option 1: Smithery

To install automatically via Smithery:

npx @smithery/cli install @tinybirdco/mcp-tinybird --client claude

Option 2: mcp-get

Install using mcp-get:

npx @michaellatman/mcp-get@latest install mcp-tinybird

Manual Installation

Install the package using uv:

uvx install mcp-tinybird

Configure your AI assistant to use the MCP server.

Configuration

Claude Desktop Configuration

Create a configuration file in the appropriate location for your operating system:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

Add the following configuration, replacing the placeholders with your actual Tinybird credentials:

{
    "mcpServers": {
        "mcp-tinybird": {
            "command": "uvx",
            "args": [
                "mcp-tinybird",
                "stdio"
            ],
            "env": {
                "TB_API_URL": "<TINYBIRD_API_URL>",
                "TB_ADMIN_TOKEN": "<TINYBIRD_ADMIN_TOKEN>"
            }
        }
    }
}

After saving the configuration, restart Claude Desktop.

SSE Mode

For web applications or other clients that support Server-Sent Events (SSE), you can run the MCP server in SSE mode:

uvx mcp-tinybird sse

Custom Prompts

The server comes with a default prompt for exploring data in Tinybird. You can configure additional prompt workflows by creating a prompts Data Source in your Tinybird workspace with the following schema:

SCHEMA >
    `name` String `json:$.name`,
    `description` String `json:$.description`,
    `timestamp` DateTime `json:$.timestamp`,
    `arguments` Array(String) `json:$.arguments[:]`,
    `prompt` String `json:$.prompt`

The MCP server loads these prompts on initialization, allowing you to customize the interaction flow.

Usage Examples

Analyzing Bluesky metrics: Example post and Claude transcript
Web analytics: Web analytics starter kit metrics with demo video

Troubleshooting

If you encounter issues:

Verify your Tinybird API URL and Admin Token are correct
Ensure Claude Desktop is properly configured and restarted
Check that the required dependencies are installed

For debugging, use the MCP Inspector:

npx @modelcontextprotocol/inspector uvx mcp-tinybird