Nutrition MCP

Nutrition MCP is a fitness tool. A free, open-source MCP server that lets Claude, ChatGPT and other MCP clients log meals, macros, water, weight and body measurements from plain-language chat, photos or barcodes, with goals, trends and CSV export.

Nutrition MCP is a free, open-source MCP server that turns an AI chat client into a calorie and macro tracker. Instead of a separate app with its own interface, the user connects the server's URL to Claude, ChatGPT or another client that supports remote MCP servers, then describes meals in ordinary language. The assistant estimates calories, protein, carbs, fat, fiber, sugar, added sugar and caffeine, and saves each entry to the user's account. The tool suits people who already spend time in an AI assistant and want food logging without learning another dashboard. Input is flexible: typed descriptions, meal photos, screenshots from delivery apps, or a barcode photo whose digits are looked up in Open Food Facts. When a portion size or meal type is missing, the assistant asks before guessing, and photo logging waits for the user to approve a summary. Past logs are searched to fill in details a photo cannot show, such as a repeat restaurant order. Beyond logging, the server exposes 41 tools covering water, weight and nine body measurements, goals and limits, alcohol tracking in US standard drinks or UK units, and reporting. Users can set calorie, macro and water targets, then ask for daily progress, 7, 14 or 30 day trend cards, streaks, weight trend smoothing, and comparisons of their own habits, such as days with and without breakfast. Patterns are reported as correlations rather than causes. All logged data can be exported as CSV files at no cost. Setup takes about a minute: sign in through OAuth with Google or an email and password, then paste the server URL into the client's connector settings. Configuration snippets are provided for Cursor, VS Code, Claude Code and Windsurf. Compared with standalone calorie counters, it trades a polished native interface for conversational flexibility and open source transparency. Compared with generic chatbots, it adds persistent storage, structured targets and reusable history, so answers draw on what was actually logged.