Nutrition MCP
A filesystem-based MCP server that turns any MCP-capable AI agent into a conversational calorie and protein tracker with natural-language estimates, confidence-aware logging, daily/weekly progress, food-history search, and export, working offline with local fallback data.
README
Nutrition MCP
A filesystem-based MCP server that turns any MCP-capable AI agent into a conversational calorie and protein tracker: natural-language estimates, confidence-aware logging, daily/weekly progress, food-history search, and export. Works offline with local fallback data — no API keys required.
Quickstart
The server runs straight from GitHub via npx — no clone, no manual build (it
self-builds on first fetch). You only need Node.js ≥ 20.
The one command every agent uses:
npx -y github:ronkommoji/nutrition-mcp
Pick your agent:
| Agent | Guide |
|---|---|
| Hermes Agent | docs/install/hermes.md |
| Claude Code | docs/install/claude.md |
| Codex | docs/install/codex.md |
| Cursor / Windsurf / Claude Desktop / other | docs/install/generic-mcp.md |
Claude Code users can install tools and the skill in one step:
/plugin marketplace add ronkommoji/nutrition-mcp
/plugin install nutrition-mcp
What's included
- MCP server — 8 tools + 2 resources (below).
- Skill —
skills/nutrition-tracking/SKILL.md: the estimate → confirm → log policy that makes the tools behave well. Auto-loaded by the Claude plugin; paste into system instructions / AGENTS.md for other agents.
Tools
setup_profile— create a user profile.update_profile— update goals, weight, goal type, or timezone.log_food— store a confirmed meal.undo_last_log— remove the most recent entry.get_daily_status— current day progress.get_weekly_summary— weekly averages and tracked-day metrics.search_food_history— search previous meals.export_logs— export logs as JSON or CSV.
Resources
nutrition://user_profilenutrition://daily_summary
Logging policy
The agent estimates calories and protein itself (its own knowledge plus web
search), shows its assumptions, and logs only after the user confirms.
log_food refuses any entry without userConfirmed: true.
Storage
Data is stored under ~/.nutrition-mcp/ by default (profile.json, logs/,
weekly/, cache/, settings.json). Override with NUTRITION_MCP_HOME.
No API keys
There are none. The agent's own model estimates calories and protein (its
knowledge plus web search), and the server only stores and reports them. The
single optional setting is NUTRITION_MCP_HOME (storage location, default
~/.nutrition-mcp).
Local development
npm install
npm run build # or: npm run dev (tsx watch)
npm start
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。