MCP Server

What is MCP Server?

The MCP Server provides a comprehensive framework for AI-powered command execution, plugin-based tools, and advanced features including synchronous script execution, secure Python evaluation, and knowledge management. It can be run as a standalone service or embedded in other projects to expose a consistent API for invoking tools and managing tasks.

How to use MCP Server?

Install dependencies with uv sync or pip install -e ., create a .env file from the template, then start the server with uv run server/main.py and connect to the SSE endpoint at http://0.0.0.0:8000/sse. Docker builds are also supported via the included Dockerfile.

Key features of MCP Server

• Run-to-completion synchronous script execution with output capture
• Safe Python expression evaluation using RestrictedPython
• Responsive web interface for data analysis and visualization
• Modular plugin system for tools and integrations
• Browser automation via Playwright integration
• Dynamic tool definitions from YAML files

Use cases of MCP Server

• Execute and capture output of shell scripts synchronously
• Run secure Python expressions for data analysis
• Browse and visualize data frames via web interface
• Integrate with Azure DevOps, Git, Kusto, and CircleCI
• Manage knowledge graphs and vector stores

FAQ from MCP Server

How to install MCP Server?

Use uv sync or pip install -e . to install all dependencies.

How to configure MCP Server?

Copy config/env.template to .env and edit variables such as GIT_ROOT, AZREPO_ORG, AZREPO_PROJECT, and AZREPO_REPO.

How to run MCP Server?

Run uv run server/main.py and connect to the SSE endpoint at http://0.0.0.0:8000/sse.

How to add custom plugins?

Declare plugins in plugin_config.yaml with a plugin_repo entry, then run the mcp_admin tool with the refresh_plugins operation.

Can I use MCP Server with an editor like Cursor or VSCode?

Yes, add the SSE URL to your editor's MCP server settings (e.g., `"url": "http://0.0.0.0