Databricks MCP Server Nm7

🚀 Databricks MCP Server

The Databricks MCP Server is a tool for interacting with the Databricks platform, implementing the Model Completion Protocol (MCP). It allows access to Databricks features via REST API and supports direct use or communication through an MCP client.

🚀 Quick Start

The Databricks MCP Server is a powerful tool designed to interact seamlessly with the Databricks platform. It implements the Model Completion Protocol (MCP), enabling users to access various Databricks features via a REST API. Whether you're using it directly or through an MCP client, this server simplifies the process of working with Databricks.

✨ Features

Functionalities

• Implements the MCP protocol.
• Provides access to the Databricks REST API.
• Supports cluster management, job submission, and other platform features.
• Documentation is generated using Sphinx and located in the docs/api directory.

Code Standards

• Follows PEP 8 standards with a maximum line length of 100 characters.
• Uses 4 spaces for indentation (no tabs).
• Prefers double quotes for string literals.
• All classes, methods, and functions must include Google-style docstrings.
• All code, except test code, requires type hints.

Code Review

• Performs code checks using Pylint and Flake8 and runs Mypy for static type checking.

Testing Requirements

• All new features must come with test cases.
• Requires a minimum code coverage of 80%.
• Uses pytest with coverage reports for testing.

📦 Installation

System Requirements

• Python 3.7 or higher.

Installation Steps

1. Clone the Project

   git clone [Repository URL]
   cd [Project Directory]
   

2. Manage Dependencies Install dependencies using poetry or other tools.
3. Set Environment Variables (Optional)
   • Configure Databricks credentials:

     export DATABRICKS_HOST=[Databricks Service Address]
     export DATABRICKS_TOKEN=[Access Token]
     

4. Run the Server

   poetry install && poetry run start_server
   

📚 Documentation

Project Structure

.
├── src/
│   ├── main.py        # Main program entry point
│   └── cli/          # Command-line tool module
├── tests/
│   ├── test_clusters.py
│   └── test_jobs.py   # Test case files
├── examples/
│   └── example_usage.py
└── docs/
    └── api/           # Documentation generation directory

For more detailed information, refer to project_structure.md.

Development Guide

Contribution Process

1. Create a Branch

   git checkout -b feature/[Feature Name]
   

2. Commit Code Include clear commit messages when committing code.
3. Submit a Pull Request After completion, submit a pull request and wait for review.

Update Documentation

• Update relevant documentation in the docs/ directory.
• Generate new documentation using Sphinx.

📄 License

This project is licensed under the MIT License. For more details, see the LICENSE file.