Anyquery MCP

Name: Anyquery MCP
Author: julien040

by julien040

Query 40+ apps with SQL — unify Notion, Airtable, Google Sheets, and more into one interface

data Other Intermediate Self-hostable No API key

⭐ 3k stars 📅 Updated: 1mo ago

View on GitHub

Description

MCP server that lets AI assistants query data from over 40 applications using standard SQL syntax. Anyquery bridges the gap between siloed SaaS data and AI workflows by providing a unified SQL interface to platforms like Notion, Airtable, Google Sheets, GitHub, Spotify, and many more. Simply connect your data sources and write SQL queries — no custom API integrations needed. Written in Go for high performance, it works as a standalone binary with no runtime dependencies. Ideal for data exploration, cross-platform reporting, and building AI-powered data pipelines.

✅ Best for

Teams and developers who need to query and combine data from multiple SaaS apps through a single SQL interface

⏭️ Skip if

You only work with a single data source that already has a dedicated MCP server

💡 Use cases

Querying and joining data across multiple SaaS platforms using SQL
Building cross-platform reports combining Notion, Sheets, and Airtable data
Exploring app data through natural language converted to SQL by AI assistants
Automating data workflows that span multiple services

👍 Pros

✓ Supports 40+ data sources with a single unified SQL interface
✓ No runtime dependencies — single Go binary, easy to install via brew or download
✓ Standard SQL syntax makes it accessible to anyone who knows SQL
✓ Active development with growing plugin ecosystem

👎 Cons

✗ AGPL-3.0 license may be restrictive for commercial embedded use
✗ Some data sources require individual authentication setup
✗ Complex joins across multiple sources can be slow on large datasets

🔧 Exposed tools (4 tools)

Tool	Category	Description
list_sources	management	List all available and connected data sources
connect_source	management	Connect a new data source to the SQL engine
execute_sql	query	Execute a SQL query across connected data sources
describe_table	query	Describe the schema of a virtual table

⚡ Installation

Check Claude Code documentation to configure this MCP server.

💡 Tips & tricks

Start by running `anyquery` and connecting your most-used data source. Use the plugin system to add new sources — each plugin maps an app's data to virtual SQL tables.

🔄 Alternatives

dbhub

MCP server — editorial content pending

Similar functionality

db-mcp