food-database-query

Food Database Query Skill

Skill Name: Food Database Query Skill Type: Data Query and Analysis Creation Date: 2026-01-06 Version: v1.0

When to Use

Use when querying food nutritional components, comparing food differences, or performing nutritional calculations.
Tasks involve food database retrieval, food recommendations, portion conversion, or category filtering.
Use when analysis results need to be generated based on structured food data rather than free-text suggestions.

Skill Overview

This skill provides comprehensive nutritional food database query functionality, supporting food nutritional information queries, comparisons, recommendations, and automatic nutritional calculations.

Core Functions:

✅ Food Nutritional Information Query
✅ Food Comparison Analysis
✅ Smart Food Recommendations
✅ Automatic Nutritional Calculation
✅ Category Browsing and Search
✅ Portion Conversion and Estimation

Data Source

Main Database

File: data/food-database.json
Content: Detailed nutritional data for 50 common foods
Structure: Each food contains 30+ nutrient indicators

Classification System

File: data/food-categories.json
Category: 10 major categories, 30+ subcategories
Support: Browse and filter by category

Functional Modules

1. Food Query

1.1 Exact Query

Purpose: Query nutritional information by food name

Supported Input:

Chinese names: "Oats", "Broccoli", "Salmon"
English names: "Oats", "Broccoli", "Salmon"
Aliases: "Oatmeal", "broccoli", "Salmon meat"

Query Process:

Receive food name
Search for matching entries in the database
Support fuzzy matching and alias matching
Return complete nutritional information

Returned Information:

Basic information (name, category, standard portion)
Macronutrients (calories, protein, carbohydrates, fat, fiber)
Micronutrients (vitamins, minerals)
Special nutrients (Omega-3/6, choline, etc.)
Glycemic index data
Health tags and applicable populations
Common portions
Nutritional advantage description

Example:

# 用户输入: "燕麦"
# 返回:
{
  "name": "燕麦",
  "name_en": "Oats",
  "category": "谷物类",
  "nutrition_per_100g": {
    "calories": 389,
    "protein_g": 16.9,
    "carbs_g": 66.3,
    "fat_g": 6.9,
    "fiber_g": 10.6,
    # ... 更多营养素
  },
  "health_tags": ["高纤维", "低GI"],
  "glycemic_index": {"value": 55, "level": "低"}
}

1.2 Fuzzy Search

Purpose: Search for foods by nutritional characteristics

Search Criteria:

Nutrient content: "high protein", "high fiber", "low GI"
Nutrient combination: "high protein low calorie", "high fiber low GI"
Category filtering: "grains", "vegetables", "protein"
Applicable populations: "vegetarian-friendly", "hypertension", "diabetes"

Search Logic:

# 示例: 搜索"高蛋白 低卡路里"
def search_foods(criteria):
    results = []
    for food in database:
        protein = food.nutrition_per_100g.protein_g
        calories = food.nutrition_per_100g.calories

        # 定义阈值
        high_protein = protein >= 15  # 每100g≥15g蛋白质
        low_calorie = calories <= 150  # 每100g≤150卡

        if high_protein and low_calorie:
            results.append(food)

    return sorted(results, key=lambda x: x.protein_g, reverse=True)

Return Format:

Sort by match degree
Display key nutrients
Label matching tags

1.3 Category Browsing

Purpose: Browse all foods by food category

Category Hierarchy:

蛋白质来源
├── 肉类
├── 禽类
├── 鱼虾贝类
├── 蛋类
├── 豆类
├── 坚果种子
└── 乳制品

Browsing Mode:

List all foods under a category
Sort by nutrient
Sort by GI value
Filter by health tags

2. Food Comparison

2.1 Two-Food Comparison

Function: Compare the nutritional differences between two foods

Comparison Dimensions:

Macronutrients: Calories, protein, carbohydrates, fat, fiber
Micronutrients: Major vitamins and minerals
Glycemic Index: GI value, glycemic load
Nutrient Density: Overall score

Calculation Logic:

def compare_foods(food1, food2):
    comparison = {}

    # 宏量营养素差异
    for nutrient in ["calories", "protein_g", "fiber_g"]:
        val1 = food1.nutrition_per_100g[nutrient]
        val2 = food2.nutrition_per_100g[nutrient]
        diff = val1 - val2
        percent = (diff / val2) * 100

        comparison[nutrient] = {
            "food1": val1,
            "food2": val2,
            "difference": diff,
            "percent_change": percent,
            "better": "food1" if diff > 0 else "food2"
        }

    return comparison

Output Format:

Comparison table
Difference percentage
Advantage annotation
Recommendations

2.2 Multi-Dimensional Comparison

Supported Modes:

Comprehensive nutritional comparison
Compare only specific nutrients
Compare only GI value
Compare only specific health tags

Examples: /nutrition compare 三文鱼鸡胸肉营养素

3. Food Recommendation

3.1 Recommendation Based on Nutrients

Recommendation Logic:

def recommend_by_nutrient(nutrient, min_value=None, max_value=None):
    recommendations = []

    for food in database:
        value = food.nutrition_per_100g[nutrient]

        # 筛选符合条件
        if min_value and value < min_value:
            continue
        if max_value and value > max_value:
            continue

        recommendations.append({
            "food": food,
            "value": value,
            "rda_percent": (value / RDA[nutrient]) * 100
        })

    # 按含量排序
    return sorted(recommendations, key=lambda x: x["value"], reverse=True)

Recommendation Categories:

High protein: ≥15g/100g
High fiber: ≥5g/100g
Low GI: ≤55
Rich in vitamin C: ≥50mg/100g
Rich in Omega-3: ≥1g/100g
High calcium: ≥100mg/100g
High iron: ≥3mg/100g

3.2 Multi-Condition Recommendation

Supported combined conditions:

"High protein, low calorie"
"High fiber, low GI"
"Rich in iron, vegetarian-friendly"

Sorting Strategy:

Sort by first priority
Filter those meeting the second condition
Sort by overall score

3.3 Recommendation Based on Health Conditions

Hypertension (DASH diet):

Low-sodium foods
High-potassium foods
High-magnesium, high-calcium foods

Diabetes:

Low-GI foods
High-fiber foods
Low-carbohydrate

Hyperlipidemia:

High-Omega-3 foods
Low saturated fat
High-fiber foods

Osteoporosis:

High-calcium foods
Rich in vitamin D
High magnesium, high zinc

Anemia:

Rich in iron
Rich in folate
Rich in vitamin B12

4. Auto Nutrition Calculation

4.1 Food Recognition

Input Parsing:

def parse_food_input(text):
    # 示例: "燕麦粥 1杯 + 鸡蛋 1个 + 牛奶 250ml"

    foods = []
    portions = []

    # 识别食物名称
    for item in text.split("+"):
        food_name = extract_food_name(item)  # "燕麦粥"
        portion = extract_portion(item)      # "1杯"

        # 标准化食物名称
        standard_name = normalize_food_name(food_name)  # "燕麦"

        # 查询数据库
        food_data = query_database(standard_name)

        foods.append(food_data)
        portions.append(parse_portion(portion))

    return foods, portions

4.2 Portion Conversion

Common Portions:

"1 cup": 240ml (liquid) or weight depends on the food
"1 piece": egg 50g, apple 150g
"1 slice": bread 30g
"100g": use directly

Portion Database:

{
  "common_portions": [
    {
      "amount": 1,
      "unit": "个",
      "weight_g": 50,
      "description": "1个大号鸡蛋"
    },
    {
      "amount": 1,
      "unit": "杯",
      "weight_g": 240,
      "description": "1杯牛奶"
    }
  ]
}

4.3 Nutrition Calculation

Calculation Formula:

def calculate_nutrition(food, portion_grams):
    nutrition = {}

    for nutrient, value_per_100g in food.nutrition_per_100g.items():
        # 按100g比例计算
        nutrition[nutrient] = (value_per_100g * portion_grams) / 100

    return nutrition

4.4 Cooking Impact Correction

Factors Considered:

Weight change after cooking
Vitamin loss
Nutrient retention rate

Examples:

Raw oats: 100g → cooked: about 300g (3 times the weight)
Vitamin retention: 60-80% retained after cooking

5. Smart Search

5.1 Alias Matching

Supports synonyms:

"oats" = "oatmeal" = "oats" = "rolled oats"
"broccoli" = "broccoli" = "broccoli"

Matching Algorithm:

def find_food(name):
    # 1. 精确匹配主名称
    if name in database:
        return database[name]

    # 2. 匹配别名
    for food in database:
        if name in food.aliases:
            return food

    # 3. 模糊匹配
    matches = fuzzy_search(name)
    if matches:
        return matches[0]

    return None

5.2 Spelling Correction

Edit Distance Algorithm:

def fuzzy_search(name, max_distance=2):
    matches = []

    for food in database:
        # 计算编辑距离
        distance = levenshtein_distance(name, food.name)

        if distance <= max_distance:
            matches.append((food, distance))

    # 按距离排序
    return sorted(matches, key=lambda x: x[1])

Data Structure

Food Data Structure

{
  "id": "FD_001",
  "name": "燕麦",
  "name_en": "Oats",
  "aliases": ["燕麦片", "oats", "rolled oats"],
  "category": "grains",
  "subcategory": "whole_grains",

  "standard_portion": {
    "amount": 100,
    "unit": "g",
    "description": "100克"
  },

  "nutrition_per_100g": {
    "calories": 389,
    "protein_g": 16.9,
    "carbs_g": 66.3,
    "fat_g": 6.9,
    "fiber_g": 10.6,
    "sugar_g": 0.99,
    "saturated_fat_g": 1.4,
    "monounsaturated_fat_g": 2.5,
    "polyunsaturated_fat_g": 2.9,
    "trans_fat_g": 0,
    "water_g": 8.9,

    "vitamin_a_mcg": 0,
    "vitamin_c_mg": 0,
    "vitamin_d_mcg": 0,
    "vitamin_e_mg": 1.1,
    "vitamin_k_mcg": 1.9,
    "thiamine_mg": 0.763,
    "riboflavin_mg": 0.139,
    "niacin_mg": 6.921,
    "vitamin_b6_mg": 0.165,
    "folate_mcg": 56,
    "vitamin_b12_mcg": 0,
    "pantothenic_acid_mg": 1.349,
    "biotin_mcg": 0,

    "calcium_mg": 54,
    "iron_mg": 4.72,
    "magnesium_mg": 177,
    "phosphorus_mg": 523,
    "potassium_mg": 429,
    "sodium_mg": 2,
    "zinc_mg": 3.97,
    "copper_mg": 0.526,
    "manganese_mg": 4.916,
    "selenium_mcg": 2.8,
    "iodine_mcg": 0
  },

  "special_nutrients": {
    "omega_3_g": 0.685,
    "omega_6_g": 1.428,
    "choline_mg": 43.4,
    "beta_carotene_mcg": 0,
    "lutein_mcg": 0,
    "zeaxanthin_mcg": 0
  },

  "glycemic_index": {
    "value": 55,
    "level": "低",
    "glycemic_load": 11
  },

  "common_portions": [
    {
      "amount": 30,
      "unit": "g",
      "description": "1/4杯",
      "approximate_volume": "1/4 cup"
    },
    {
      "amount": 40,
      "unit": "g",
      "description": "1/3杯",
      "approximate_volume": "1/3 cup"
    },
    {
      "amount": 200,
      "unit": "ml",
      "description": "煮熟1杯",
      "notes": "煮熟后体积增加"
    }
  ],

  "cooking_effects": {
    "boiling": {
      "weight_change_percent": 200,
      "nutrient_changes": {
        "vitamin_c_retention": 0,
        "b_vitamins_retention": 60
      }
    }
  },

  "health_tags": ["高纤维", "低GI", "无麸质选项", "心脏健康"],

  "suitable_for": ["素食者", "高血压", "糖尿病", "高血脂"],

  "notes": "富含β-葡聚糖,有助于降低胆固醇"
}

RDA Reference Values

Adult Men (19-50 years)

RDA = {
  # 宏量营养素
  "calories": 2500,  # 中等活动水平
  "protein_g": 56,
  "carbs_g": 130,  # 最低值
  "fiber_g": 38,

  # 维生素
  "vitamin_a_mcg": 900,
  "vitamin_c_mg": 90,
  "vitamin_d_mcg": 15,
  "vitamin_e_mg": 15,
  "vitamin_k_mcg": 120,
  "thiamine_mg": 1.2,
  "riboflavin_mg": 1.3,
  "niacin_mg": 16,
  "vitamin_b6_mg": 1.3,
  "folate_mcg": 400,
  "vitamin_b12_mcg": 2.4,
  "pantothenic_acid_mg": 5,
  "biotin_mcg": 30,

  # 矿物质
  "calcium_mg": 1000,
  "iron_mg": 8,
  "magnesium_mg": 400,
  "phosphorus_mg": 700,
  "potassium_mg": 3400,
  "sodium_mg": 1500,  # 上限
  "zinc_mg": 11,
  "copper_mg": 0.9,
  "manganese_mg": 2.3,
  "selenium_mcg": 55
}

Adult Women (19-50 years)

RDA_FEMALE = {
  "calories": 2000,  # 中等活动水平
  "protein_g": 46,
  "fiber_g": 25,
  "iron_mg": 18,  # 育龄期
  # ... 其他略有差异
}

Integration Features

Integration with Nutrition Module

Log Diet: Automatically query nutrition data
Nutrition Analysis: Precise calculation based on database
Nutrition Advice: Data-driven food recommendations

Integration with Health Module

Hypertension: Recommend DASH diet-friendly foods
Diabetes: Filter low-GI foods
Hyperlipidemia: Recommend high Omega-3 foods

Integration with Exercise Module

Before and After Exercise: Recommend suitable foods
Muscle Gain: High-protein food recommendations
Fat Loss: Low-calorie high-protein foods

Usage Examples

Example 1: Log Breakfast

User Input:

/nutrition record breakfast 燕麦粥 1杯 + 鸡蛋 1个 + 牛奶 250ml

System Processing:

Recognize foods: oats, eggs, milk
Query nutrition data
Calculate nutrition per serving
Summarize nutrition for the whole meal
Log to diary

Return Result:

✅ 早餐已记录

**食物**: 燕麦粥(1杯) + 鸡蛋(1个) + 牛奶(250ml)

**营养汇总**:
- 卡路里: 417 卡
- 蛋白质: 25.1g
- 碳水化合物: 48.5g
- 脂肪: 15.2g
- 膳食纤维: 8.2g

**微量营养素亮点**:
- 维生素D: 3.1 μg (21% RDA)
- 钙: 332 mg (33% RDA)
- 维生素B12: 1.3 μg (54% RDA)

Example 2: Query Food

User Input:

/nutrition food 三文鱼

Return Result:

# 三文鱼 营养信息

## 基本信息
- **名称**: 三文鱼 (Salmon)
- **分类**: 蛋白质来源 > 鱼虾贝类
- **标准份量**: 100克

## 宏量营养素 (每100克)
- **卡路里**: 208 卡
- **蛋白质**: 20g ✅
- **碳水化合物**: 0g
- **脂肪**: 13g
- **Omega-3**: 2.5g ✅✅✅

## 营养亮点
- ✅✅✅ 富含Omega-3脂肪酸 (EPA+DHA)
- ✅✅ 高质量蛋白质
- ✅ 富含维生素D (11μg)
- ✅ 富含维生素B12 (3.2μg)

## 健康标签
- ✅ 高蛋白
- ✅ 富含Omega-3
- ✅ 心脏健康
- ✅ 大脑健康

## 推荐份量
- 100-150g/餐 (每周2-3次)

Example 3: Compare Foods

User Input:

/nutrition compare 鸡胸肉 三文鱼

Return Result:

# 食物比较: 鸡胸肉 vs 三文鱼

## 营养对比 (每100克)

| 营养素 | 鸡胸肉 | 三文鱼 | 差异 |
|--------|--------|--------|------|
| 卡路里 | 165 | 208 | +26% |
| 蛋白质 (g) | 31 | 20 | -35% ✅ |
| 脂肪 (g) | 3.6 | 13 | +261% |
| Omega-3 (g) | 0.1 | 2.5 | +2400% ✅✅✅ |

## 推荐建议

**选择鸡胸肉更适合**:
- ✅ 减脂期间 (低卡高蛋白)
- ✅ 控制脂肪摄入
- ✅ 蛋白质需求高

**选择三文鱼更适合**:
- ✅ 心脏健康 (高Omega-3)
- ✅ 大脑健康 (DHA)
- ✅ 抗炎需求

Expansion Plan

Short Term (1-2 months)

✅ Completed 50 common foods
⏳ Expand to 100 foods
⏳ Add more common servings
⏳ Optimize search algorithm

Medium Term (3-6 months)

⏳ Expand to 300 foods
⏳ Add branded foods
⏳ Support user-defined foods
⏳ Add food photos

Long Term (Ongoing)

⏳ Continuously update database
⏳ Add seasonal foods
⏳ Integrate barcode scanning
⏳ AI food recognition

Quality Assurance

Data Accuracy

Source: China Food Composition Tables (6th Edition) + USDA
Verification: Cross-validate multiple sources
Updates: Regularly update data

Functional Testing

Query accuracy testing
Calculation precision testing
Boundary condition testing
Performance testing

Notes

⚠️ Important Limitations

Data Scope: Currently covers only 50 common foods
Cooking Impact: Data is based on raw food/standard cooking
Individual Differences: Actual nutrient absorption varies from person to person
Regional differences: Food nutrition may differ by region

⚠️ Usage recommendations

Balanced diet: Do not rely on a single food
Diversified choices: Rotate different foods
Moderation principle: Even healthy foods should be consumed in moderation
Professional guidance: Consult a dietitian for special needs

Technical implementation

File location

Database: data/food-database.json
Category: data/food-categories.json
Command: .claude/commands/nutrition.md
Skill: .claude/skills/food-database-query/SKILL.md

Performance optimization

Database index (food name, category)
Cache frequently used queries
Fuzzy search optimization

Skill version: v1.0 Last updated: 2026-01-06 Maintainer: WellAlly Tech

Limitations

Use this skill only when the task clearly matches the scope described above.
Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

About this skill

When to Use

Skill Overview

Data Source

Main Database

Classification System

Functional Modules

1. Food Query

1.1 Exact Query

1.2 Fuzzy Search

1.3 Category Browsing

2. Food Comparison

2.1 Two-Food Comparison

2.2 Multi-Dimensional Comparison

3. Food Recommendation

3.1 Recommendation Based on Nutrients

3.2 Multi-Condition Recommendation

3.3 Recommendation Based on Health Conditions

4. Auto Nutrition Calculation

4.1 Food Recognition

4.2 Portion Conversion

4.3 Nutrition Calculation

4.4 Cooking Impact Correction

5. Smart Search

5.1 Alias Matching

5.2 Spelling Correction

Data Structure

Food Data Structure

RDA Reference Values

Adult Men (19-50 years)

Adult Women (19-50 years)

Integration Features

Integration with Nutrition Module

Integration with Health Module

Integration with Exercise Module

Usage Examples

Example 1: Log Breakfast

Example 2: Query Food

Example 3: Compare Foods

Expansion Plan

Short Term (1-2 months)

Medium Term (3-6 months)

Long Term (Ongoing)

Quality Assurance

Data Accuracy

Functional Testing

Notes

⚠️ Important Limitations

⚠️ Usage recommendations

Technical implementation

File location

Performance optimization

Limitations