SebastianGilPinzon/colab-mcp

What it does

SebastianGilPinzon/colab-mcp is an MCP (Model Context Protocol) server that enables AI agents to programmatically control Google Colab notebooks and manage GPU resource allocation. It allows agents to create notebooks, execute cells, and assign specific GPU types (T4, L4, A100) directly from conversational AI interfaces. This is an enhanced fork of Google's original colabmcp with improved visibility of all tools at startup, OAuth-based GPU control, and extended Windows platform support.

How to set up

Installation involves cloning the repository and configuring it as an MCP server. Users need to set up OAuth credentials for Google Colab access, configure their AI agent to connect to the MCP server endpoint, and specify preferred GPU types. Detailed setup instructions are available in the repository's README, with support for Python environments on macOS, Windows, Linux, and cloud platforms.

Key features

• Control Google Colab notebooks from any MCP-compatible AI agent
• GPU allocation and management (T4, L4, A100 options)
• OAuth-based secure authentication
• All available tools visible at startup
• Cross-platform support (Windows, macOS, Linux)
• Enhanced fork with improved stability and features