功能描述

Personal health data management and intelligent diet recommendation system. Use when users need to: - Record and track body metrics (weight, height, BMI, bod...

使用说明 (SKILL.md)

NutriCoach Skill

Name: nutricoach 营养专家
Author: rainbowlion0320

A comprehensive personal health management system for tracking body metrics, logging meals, analyzing nutrition, managing pantry inventory, and receiving intelligent diet recommendations.

Quick Start

1. Initialize Database

python3 scripts/init_db.py --user \x3Cusername>

Creates isolated SQLite database for the user with all required tables.

2. Set Up User Profile

python3 scripts/user_profile.py --user \x3Cusername> set \
  --name "Robert" \
  --gender male \
  --birth-date 1994-05-15 \
  --height-cm 175 \
  --target-weight 70 \
  --activity-level moderate \
  --goal-type maintain

3. Log Daily Weight

python3 scripts/body_metrics.py --user \x3Cusername> log-weight --weight 72.5

4. Log a Meal

python3 scripts/meal_logger.py --user \x3Cusername> log \
  --meal-type lunch \
  --foods "米饭:150g, 鸡胸肉:100g, 西兰花:100g"

5. Get Diet Recommendation

python3 scripts/diet_recommender.py --user \x3Cusername> recommend --meal-type dinner

6. Launch Web Dashboard

python3 scripts/launch_dashboard.py --user \x3Cusername>

Opens browser at http://127.0.0.1:5000 with visual health data overview.

Core Features

📊 Body Metrics Tracking

Weight logging with automatic BMI calculation
Body fat percentage tracking
7/30-day trend analysis
TDEE (Total Daily Energy Expenditure) calculation

🍽️ Meal Logging

Natural language food entry: 米饭:150g, 鸡胸肉:100g
Automatic nutrition calculation
Daily summary with remaining calories vs TDEE
569 built-in Chinese foods database

📷 OCR Food Recognition

Scan food packaging with camera
Extract nutrition facts automatically
Barcode matching for existing products
Support for Kimi Vision and macOS Vision engines

🥬 Pantry Management

Track inventory by storage location (fridge/freezer/dry goods/counter)
Expiration date tracking with alerts
Automatic quantity deduction when logging meals
Category-based organization (protein/vegetable/carb/fruit/dairy)

🍳 Smart Recipe Recommendations

Generate recipes based on available pantry items
Prioritize ingredients nearing expiration
Match recipes to nutritional gaps
Dynamic threshold adjustment based on inventory levels

📈 Web Dashboard (v2)

Tabbed interface: Overview / Pantry / Weight History
Visual charts for weight and nutrition trends
Interactive pantry management
One-click recipe generation

💾 Data Management

JSON/CSV export
Automatic database backups (keeps last 10)
Easy restore from backup
Multi-user data isolation

Architecture

See references/ARCHITECTURE.md for system design and data flow.

Database Schema

See references/DATABASE_SCHEMA.md for complete ER diagram and table definitions.

Feature Guide

See references/FEATURE_GUIDE.md for detailed user-facing documentation including:

Complete feature descriptions
OCR food recognition workflow
Pantry management guide
Smart recipe recommendations
Web dashboard usage
Quick command reference

Developer Guide

See references/DEVELOPER_GUIDE.md for:

Complete API reference for all scripts
Food database structure
Database schema details
Extension guidelines

Changelog

See CHANGELOG.md for version history and breaking changes.

Available Scripts

Script	Purpose
`init_db.py`	Initialize user database
`user_profile.py`	Manage user profile and goals
`body_metrics.py`	Log weight and track trends
`meal_logger.py`	Log meals and view daily summary
`diet_recommender.py`	Get meal recommendations
`food_analyzer.py`	Search foods, OCR scan, add custom
`pantry_manager.py`	Manage ingredient inventory
`smart_recipe.py`	Generate recipes from pantry
`launch_dashboard.py`	Start web dashboard
`report_generator.py`	Generate weekly/nutrition reports
`export_data.py`	Export to JSON/CSV
`backup_db.py`	Backup and restore database
`food_ocr.py`	OCR engine wrapper
`food_matcher.py`	Match OCR results to database
`migrate_db.py`	Database migration tool

Data Storage

Each user has an isolated SQLite database stored at:

~/.openclaw/workspace/skills/nutricoach/data/\x3Cusername>.db

Backups are stored in:

~/.openclaw/workspace/skills/nutricoach/data/backups/

Common Workflows

Daily Logging Workflow

Morning: Log weight with body_metrics.py log-weight
After meals: Log food with meal_logger.py log
Evening: Review daily nutrition summary

Weekly Review Workflow

Generate weekly report with report_generator.py weekly
Check weight trend
Adjust targets if needed with user_profile.py update

Photo Food Recognition Workflow

Save food photo to accessible path
Run: python3 scripts/food_analyzer.py --user \x3Cusername> scan --image \x3Cpath>
Review identified foods and confirm/edit quantities
Log confirmed foods with meal_logger.py log

Pantry Management Workflow

Add groceries: pantry_manager.py add --food "鸡胸肉" --quantity 500
Log meals (auto-deducts from pantry)
Check expiring items: pantry_manager.py list --expiring 3
Get recipe ideas: smart_recipe.py --count 3

Web Dashboard Workflow

Launch: launch_dashboard.py --user \x3Cusername>
Browse Overview tab for trends
Manage Pantry tab for inventory
Use smart recipe buttons directly from pantry view

Configuration

OCR Engine Setup

OCR 食品包装识别支持多种方式，零配置即可使用。

方式一：AI 助手识别（推荐，零配置）

如果你使用 AI 助手（如 OpenClaw），直接发送食品包装图片：

[发送图片]
"录入这个食材，生产日期 2025-01-15"

助手会直接识别图片并录入，无需任何配置。

方式二：本地识别（命令行，免费）

使用 macOS 内置 Vision 框架，无需配置：

python3 scripts/food_analyzer.py --user \x3Cname> scan --image food.jpg --engine macos

方式三：云端识别（命令行，可选）

如需更高识别精度，配置云端 Vision API：

步骤 1：创建配置文件

cp data/user_config.example.yaml data/user_config.yaml

步骤 2：填入 API key

vision:
  api_key: "your-api-key"
  base_url: "https://api.moonshot.cn/v1"  # 或其他兼容 OpenAI 的 API
  model: "kimi-k2.5"

步骤 3：使用

python3 scripts/food_analyzer.py --user \x3Cname> scan --image food.jpg

支持的 API 提供商：

Moonshot (Kimi)
OpenAI
阿里 DashScope
其他兼容 OpenAI 接口的服务

说明：

user_config.yaml 已被加入 .gitignore，不会意外提交
云端 OCR 仅在命令行模式下需要配置
AI 助手模式始终优先使用助手内置的视觉能力

Food Database

Built-in database includes 569 Chinese foods across categories:

Proteins: 鸡胸肉, 牛肉, 鸡蛋, 豆腐, etc.
Vegetables: 西兰花, 菠菜, 西红柿, etc.
Carbs: 米饭, 燕麦, 红薯, etc.
Fruits: 苹果, 香蕉, 橙子, etc.
Dairy: 牛奶, 酸奶, 奶酪, etc.

Add custom foods via OCR scan or manual entry.

Last updated: 2026-03-28 (v2.0: Added pantry management, smart recipes, OCR, web dashboard)

安全使用建议

This skill appears to implement what it claims (local meal/pantry tracking, recipe suggestions, OCR wrappers and a local dashboard). Before installing and running: - If you want to use OCR cloud services, be aware that you will need to provide an API key (e.g., OPENAI_API_KEY or a Kimi Vision key) — only enable cloud OCR if you trust the external provider and accept that images/nutrition data may be sent off-device. The skill's metadata does not declare those env vars explicitly. - The code includes a HARDCODING_AUDIT pointing out many hardcoded row indices and other brittle code; this is an engineering quality issue (risk of bugs/corrupted exports) but not malicious. Review/patch db_schema usage before relying on it in production. - The web dashboard template pulls Chart.js from a CDN and the UI may request external assets; if you need an offline deployment, host those assets locally. - Data is stored in the skill's data/backups directories (relative to the skill). Backups are local copies of the DB — ensure you store backups securely (they contain sensitive health data). - Recommended: run the code in an isolated environment (virtualenv/container), review any optional OCR integration code paths you plan to enable, and inspect or run the test suite included to validate behaviour on your platform. If you want, I can: (a) scan specific files for network calls and exact locations where OPENAI_API_KEY or other env vars are read, (b) list all code locations that call subprocess/network, or (c) point out the highest-risk lines that should be reviewed/changed before deployment.

能力评估

✓ Purpose & Capability

Name, description, SKILL.md and the included scripts (meal logging, pantry, OCR wrappers, web dashboard, backups, export) are coherent — the code implements the advertised features (body metrics, meals, pantry, OCR, recipe recommendations, web UI). There are no obvious unrelated permissions or credentials requested.

ℹ Instruction Scope

SKILL.md instructs running the included scripts and launching a local web dashboard; the scripts operate on per-user SQLite DBs under the skill's data folder. A few instructions and scripts reference optional cloud OCR and advise setting OPENAI_API_KEY or using a data/user_config.yaml — that environment credential is not declared in the skill metadata. The skill will run subprocesses (calls to other included scripts) and read/write local files (data and backups) — all consistent with its function, but the optional OCR path can involve networked OCR services if configured.

✓ Install Mechanism

There is no install spec (the package is distributed as source scripts). Dependencies are declared in requirements.txt (Flask, PyYAML, optional OpenAI/Tesseract commented). No remote archives or download URLs are used by the install process shown. This is low install risk, but running pip install -r requirements.txt will pull packages from PyPI if used.

ℹ Credentials

The skill declares no required environment variables, which aligns with local-only operation. However, the code and SKILL.md mention optional cloud OCR (Kimi Vision / OpenAI) and reference OPENAI_API_KEY and data/user_config.yaml in error messages — these are optional but if you enable cloud OCR the skill will use API keys not declared up-front. No unrelated system credentials (AWS, GitHub tokens, etc.) are requested.

✓ Persistence & Privilege

always:false and user-invocable:true. The skill stores per-user SQLite DBs and keeps backups under its data/backups directory; it does not request system-wide configuration changes or claim permanent always-on privileges. Backups are local copies of the DB and the skill does not modify other skills' configuration.

版本历史

v1.0.1

- Added configuration examples and explanation for local and cloud OCR recognition, including user_config.yaml workflow. - Updated the configuration section with clear options for zero-config AI assistant, local (macOS), and cloud (API) OCR modes. - .gitignore and Clawhub project files were added for improved version control and project management. - Removed deprecated static CSS and archive dashboard template files, streamlining the codebase. - Documentation now better clarifies cloud OCR setup steps and safety regarding API key confidentiality.

v1.0.0

Major upgrade: NutriCoach evolves from a simple health tracker to an all-in-one personal health and nutrition management platform. - Comprehensive tracking for body metrics, daily meals, and nutrition history - Intelligent diet recommendations and recipe suggestions based on pantry items, goals, and nutrition gaps - Pantry inventory management with expiration tracking and automated deduction during meal logging - OCR food recognition from photos and support for barcode/packaging scan - Multi-user database isolation and full-featured web dashboard with luxury refined UI - 570 Chinese foods database with complete nutrition data - Data export, automatic backups, and detailed usage guides

元数据

Slug nutricoach

版本 1.0.1

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 2

常见问题

nutricoach 营养专家是什么？

Personal health data management and intelligent diet recommendation system. Use when users need to: - Record and track body metrics (weight, height, BMI, bod... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 127 次。

如何安装 nutricoach 营养专家？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install nutricoach」即可一键安装，无需额外配置。

nutricoach 营养专家是免费的吗？

是的，nutricoach 营养专家完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

nutricoach 营养专家支持哪些平台？

nutricoach 营养专家跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 nutricoach 营养专家？

由 RainbowLion0320（@rainbowlion0320）开发并维护，当前版本 v1.0.1。

nutricoach 营养专家