Tinybird MCP

Name: Tinybird MCP
Author: Tinybird

by Tinybird

Real-time analytics on Tinybird serverless ClickHouse — query, manage pipes, and publish APIs

database Python Intermediate Verified

⭐ 150 stars 📅 Updated: 1mo ago

View on GitHub View package

Description

MCP server for Tinybird, the serverless real-time analytics platform built on ClickHouse. Enables LLMs to query data sources, manage pipes (transformation pipelines), push data, and publish API endpoints — all through natural language. Tinybird turns raw data into live API endpoints, and this MCP server gives your AI assistant full access to that workflow. Query billions of rows in milliseconds, explore your data pipeline architecture, and create new analytics endpoints without leaving your AI chat. Ideal for teams building real-time dashboards, product analytics, and data-driven APIs at scale.

✅ Best for

Teams using Tinybird for real-time analytics who want AI-assisted data exploration and API management

⏭️ Skip if

You need a generic ClickHouse connector without Tinybird's platform features

💡 Use cases

Querying real-time analytics data through conversational AI
Managing Tinybird pipes and data sources from AI assistants
Publishing and testing analytics API endpoints via natural language
Exploring data pipeline architecture and debugging transformation logic

👍 Pros

✓ Full access to Tinybird's real-time analytics platform from AI
✓ Sub-second query performance on billions of rows via ClickHouse engine
✓ Manage the complete data pipeline — ingest, transform, and publish APIs
✓ Simple token-based authentication with pip or uvx install

👎 Cons

✗ Requires a Tinybird account with active workspace and token
✗ ClickHouse SQL dialect has differences from standard SQL
✗ API endpoint management requires understanding Tinybird's pipe concept

🔧 Exposed tools (5 tools)

Tool	Category	Description
push_data	data	Push data into a Tinybird data source
list_pipes	discovery	List all pipes (transformation pipelines) in the workspace
list_datasources	discovery	List all data sources available in the workspace
publish_endpoint	management	Publish a pipe as a live API endpoint
query_data	query	Execute a SQL query against Tinybird data sources

⚡ Installation

Prerequisites:

• python v3.10+
• Tinybird authentication token
• API key required

Check Claude Code documentation to configure this MCP server.

💡 Tips & tricks

Set the TINYBIRD_TOKEN environment variable with your workspace admin token. Start by listing data sources and pipes to understand your workspace structure. Use query_data to explore data before building new pipes or endpoints.

🔄 Alternatives

ClickHouse MCP

Query ClickHouse databases with natural language for real-time analytics and OLAP

Similar functionality

MotherDuck MCP

Query and analyze data with MotherDuck serverless DuckDB and local files via SQL