Gravitino
MCP server for Apache Gravitino
What is Gravitino?
Gravitino is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server for apache gravitino
MCP server for Apache Gravitino
This server falls under the Databases and Data Science & ML categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- MCP server for Apache Gravitino
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx gravitinoConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Gravitino
The Gravitino MCP server connects AI assistants to Apache Gravitino, the open-source unified metadata lake that manages catalogs, schemas, tables, tags, roles, and ML model versions across heterogeneous data sources. Through this server, Claude and other AI tools can list and explore metadata catalogs, retrieve column-level table details, manage user roles and permissions, apply tags to data assets, and browse registered ML model versions — all without writing Gravitino API calls directly. It is designed for data engineers and platform teams who want AI-assisted metadata governance and discovery across multi-cloud data estates.
Prerequisites
- Apache Gravitino server running and accessible (default: http://localhost:8090)
- Python 3.9 or later with uv installed (uv is used for dependency management)
- Git to clone the repository
- Gravitino credentials: either a JWT token (GRAVITINO_JWT_TOKEN) or username and password (GRAVITINO_USERNAME and GRAVITINO_PASSWORD)
- An MCP-compatible client such as Claude Desktop or Goose
Clone the repository
The server is not published to PyPI, so clone it from GitHub and enter the project directory.
git clone [email protected]:datastrato/mcp-server-gravitino.git
cd mcp-server-gravitinoCreate a virtual environment with uv
uv creates and manages a fast, reproducible virtual environment. Activate it before installing dependencies.
uv venv
source .venv/bin/activateInstall dependencies
Install the server and its dependencies into the virtual environment.
uv installSet environment variables
Configure the required connection variables. GRAVITINO_URI and GRAVITINO_METALAKE are always required. Choose either JWT or basic auth.
export GRAVITINO_URI=http://localhost:8090
export GRAVITINO_METALAKE=metalake_demo
# JWT authentication:
export GRAVITINO_JWT_TOKEN=your_jwt_token
# OR basic auth:
export GRAVITINO_USERNAME=admin
export GRAVITINO_PASSWORD=yourpasswordAdd to Claude Desktop config
Register the server in your MCP client configuration using uv run to invoke the server inside its virtual environment.
{
"mcpServers": {
"gravitino": {
"command": "uv",
"args": [
"--directory", "/absolute/path/to/mcp-server-gravitino",
"run", "--with", "fastmcp", "--with", "httpx",
"--with", "mcp-server-gravitino",
"python", "-m", "mcp_server_gravitino.server"
],
"env": {
"GRAVITINO_URI": "http://localhost:8090",
"GRAVITINO_METALAKE": "metalake_demo",
"GRAVITINO_USERNAME": "admin",
"GRAVITINO_PASSWORD": "yourpassword"
}
}
}
}Gravitino Examples
Client configuration
Complete claude_desktop_config.json entry for the Gravitino MCP server using uv run and basic authentication.
{
"mcpServers": {
"gravitino": {
"command": "uv",
"args": [
"--directory", "/Users/yourname/mcp-server-gravitino",
"run", "--with", "fastmcp", "--with", "httpx",
"--with", "mcp-server-gravitino",
"python", "-m", "mcp_server_gravitino.server"
],
"env": {
"GRAVITINO_URI": "http://localhost:8090",
"GRAVITINO_METALAKE": "metalake_demo",
"GRAVITINO_USERNAME": "admin",
"GRAVITINO_PASSWORD": "yourpassword"
}
}
}
}Prompts to try
Example queries you can ask Claude to run against your Gravitino metadata lake.
- "List all catalogs in the metalake_demo metalake"
- "Show me the schemas in the Hive catalog and describe the tables in the sales schema"
- "Get the column details for the orders table including data types and descriptions"
- "List all tags and show me which data assets are tagged as PII"
- "Show me the current user roles and who has admin permissions"
- "List the registered ML models and their latest versions"Troubleshooting Gravitino
Connection refused when the server tries to reach GRAVITINO_URI
Verify that the Apache Gravitino server is running: curl http://localhost:8090/api/version. If Gravitino is on a remote host, update GRAVITINO_URI to the correct hostname and port. Check firewall rules if running in a container or cloud environment.
Authentication fails with 401 Unauthorized
Ensure you are using only one authentication method at a time. If GRAVITINO_JWT_TOKEN is set, GRAVITINO_USERNAME and GRAVITINO_PASSWORD are ignored. Clear both sets of variables and set only the one your Gravitino instance expects.
uv run fails with 'package not found: mcp-server-gravitino'
The --with mcp-server-gravitino flag expects the package to be installable. If it has not been published to PyPI, replace --with mcp-server-gravitino with a direct path: --with /absolute/path/to/mcp-server-gravitino, or run 'uv pip install -e .' inside the project directory first.
Frequently Asked Questions about Gravitino
What is Gravitino?
Gravitino is a Model Context Protocol (MCP) server that mcp server for apache gravitino It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Gravitino?
Follow the installation instructions on the Gravitino GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Gravitino?
Gravitino works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Gravitino free to use?
Yes, Gravitino is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Gravitino Alternatives — Similar Databases Servers
Looking for alternatives to Gravitino? Here are other popular databases servers you can use with Claude, Cursor, and VS Code.
Excelize
★ 20.6kGo language library for reading and writing Microsoft Excel™ (XLAM / XLSM / XLSX / XLTM / XLTX) spreadsheets
MCP Toolbox for Databases
★ 15.3kOpen source MCP server specializing in easy, fast, and secure tools for Databases.
DBHub
★ 2.8kA universal database gateway MCP server that enables AI assistants to connect to and query multiple databases (PostgreSQL, MySQL, MariaDB, SQL Server, SQLite) with support for schema exploration, SQL execution, and secure connections via SSH tunnels.
Tabularis
★ 2.1kA lightweight, cross-platform database client for developers. Supports MySQL, PostgreSQL and SQLite. Hackable with plugins. Built for speed, security, and aesthetics.
Postgres AI Guide
★ 1.7kMCP server and Claude plugin for Postgres skills and documentation. Helps AI coding tools generate better PostgreSQL code.
Anyquery
★ 1.7k🏎️ 🏠 ☁️ - Query more than 40 apps with one binary using SQL. It can also connect to your PostgreSQL, MySQL, or SQLite compatible database. Local-first and private by design.
Browse More Databases MCP Servers
Explore all databases servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Gravitino in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Gravitino?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.