TrainTools

Overview

TrainTools is a Model Context Protocol (MCP) server: Recommend paper-backed diagnostics for PyTorch and Hugging Face training problems. In practice that means any MCP-compatible AI assistant can call TrainTools's tools directly — the model decides when to use them in a conversation or agent run.

It runs locally: the Python package traintools (via uvx) speaks MCP over stdio on your machine. No credentials are required — it works out of the box. The current release is v0.6.2, published under the MIT license, with source at AparajeetS/Traintools on GitHub.

TrainTools is listed under AI & Agents on mcp.site and works with any MCP client — Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed and the rest of the ecosystem — using the install snippets below. If you maintain TrainTools for AparajeetS, claim this listing to verify ownership, earn the Verified badge, and keep the details current.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "traintools": {
      "command": "uvx",
      "args": [
        "traintools"
      ]
    }
  }
}