yazio-exporter

Export all your Yazio nutrition, body, and exercise data to JSON, CSV, or SQLite.

Features

• Full data coverage — daily diary, consumed items (products + recipes), weight & body measurements, exercises, water intake, micronutrients (40+ vitamins & minerals), and user profile
• Three output formats — JSON (pretty-printed), CSV (one file per data type), SQLite (normalized with foreign keys and indexes)
• Built-in analytics — calorie stats, macro ratios, meal distribution, weight trend, calorie-weight correlation, top products
• LLM-ready analysis — generates a compact prompt you can paste into ChatGPT, Claude, DeepSeek, or Gemini for personalized nutrition insights
• Auto-discovery — scans your history automatically, no date range needed
• Concurrent fetching — parallel API calls (up to 10 workers) for fast exports
• One-shot export — export-all dumps everything in a single command

Requirements

• Python 3.11+
• A Yazio account with email/password login

Installation

From PyPI (recommended)

pip install yazio-exporter

With uv

uv tool install yazio-exporter

With pipx

pipx install yazio-exporter

From source

git clone https://github.com/aleksandr-bogdanov/yazio-exporter.git
cd yazio-exporter
uv sync

Quick start

# One command — exports everything to output/
yazio-exporter export-all you@email.com yourpassword

# Regenerate reports from existing exports
yazio-exporter report -d output/

# Or step by step:
yazio-exporter login you@email.com yourpassword
yazio-exporter days
yazio-exporter weight
yazio-exporter nutrients
yazio-exporter products
yazio-exporter summary -w weight.json
yazio-exporter report

All commands have sensible defaults — login saves to token.txt, other commands read from it, and each writes to its natural filename (days.json, weight.json, etc.).

CLI reference

login — Authenticate and save token

yazio-exporter login EMAIL PASSWORD [-o token.txt]

Token file is created with 0600 permissions.

profile — Export user profile

yazio-exporter profile [-t token.txt] [-o profile.json] [--format json]

days — Export daily diary data

yazio-exporter days [-t token.txt] [-o days.json] [-f 2024-01-01] [-e 2024-12-31]

Without --from-date/--end-date, auto-discovers all months with data.

weight — Export weight and body measurements

yazio-exporter weight [-t token.txt] [-o weight.json] [-f 2024-01-01] [-e 2024-12-31]

nutrients — Export nutrient history

yazio-exporter nutrients [-t token.txt] [-o nutrients.json] [-n vitamin.d,mineral.iron]

Available nutrients:

• Vitamins: vitamin.a, vitamin.b1, vitamin.b2, vitamin.b3, vitamin.b5, vitamin.b6, vitamin.b7, vitamin.b11, vitamin.b12, vitamin.c, vitamin.d, vitamin.e, vitamin.k
• Minerals: mineral.calcium, mineral.iron, mineral.potassium, mineral.magnesium, mineral.phosphorus, mineral.zinc, mineral.copper, mineral.manganese, mineral.selenium, mineral.iodine, mineral.fluoride, mineral.chlorine, mineral.choline

products — Resolve product details from days export

yazio-exporter products [-t token.txt] [-f days.json] [-o products.json]

summary — Generate analytics and statistics

yazio-exporter summary [-f days.json] [-w weight.json] [-p products.json]

Outputs to stdout.

export-all — Complete export pipeline

yazio-exporter export-all EMAIL PASSWORD [-o output/]

Creates the output directory and writes: profile.json, days.json, weight.json, nutrients.json, products.json, summary.txt, analysis.md, llm_prompt.txt.

report — Generate analysis and LLM prompt from existing exports

yazio-exporter report [-d output/] [--start 2024-01-01] [--end 2024-12-31]

Reads days.json, weight.json, products.json, and profile.json from the directory and generates analysis.md (pre-computed statistics) and llm_prompt.txt (paste-ready LLM prompt).

LLM analysis

The export-all and report commands generate an llm_prompt.txt file — a compact, paste-ready prompt containing your nutrition data formatted for any LLM. Copy the contents and paste into ChatGPT, Claude, DeepSeek, or Gemini to get personalized insights:

# Generate from existing exports
yazio-exporter report -d output/

# Filter to a specific date range
yazio-exporter report -d output/ --start 2024-06-01 --end 2024-12-31

# Then paste llm_prompt.txt into your LLM of choice
cat output/llm_prompt.txt | pbcopy  # macOS

The prompt instructs the LLM to act as a sports nutritionist and analyze binge-recovery cycles, plateau mechanics, TDEE, trigger patterns, protein adequacy, food crutches, and more — all backed by your actual data.

An analysis.md file is also generated with pre-computed statistics (calorie averages, macro splits, weight trends, top foods) that you can reference directly or include alongside the LLM prompt for richer context.

Output examples

JSON (days)

{
  "2024-01-15": {
    "consumed": {
      "products": [
        {"product_id": "abc-123", "amount": 150, "daytime": "breakfast"}
      ],
      "recipe_portions": [],
      "simple_products": []
    },
    "goals": {"data": {"energy": 2000, "protein": 120}},
    "exercises": {"training": [], "custom_training": [], "activity": null},
    "water": {"water_intake": 2500},
    "daily_summary": {"meals": {"breakfast": {"nutrients": {"energy.energy": 450}}}}
  }
}

CSV (nutrients)

date,nutrient_id,value
2024-01-15,vitamin.d,15.2
2024-01-15,mineral.iron,8.5
2024-01-16,vitamin.d,12.0

SQLite schema

days(date, energy, carb, protein, fat, energy_goal)
consumed_items(id, date, product_id, amount, energy)
products(product_id, name, category, energy_per_100g, carb_per_100g, protein_per_100g, fat_per_100g)
recipes(recipe_id, name, portion_count, energy_per_portion, ...)
weight_log(date, weight, body_fat, waist, hip, chest)
nutrient_daily(id, date, nutrient_id, value)
exercises(id, date, exercise_type, duration_minutes, calories_burned)
water_intake(date, water_ml)
goals(date, energy_goal, protein_goal, fat_goal, carb_goal, water_goal, ...)
daily_summary(date, total_energy, breakfast_energy, lunch_energy, dinner_energy, snack_energy)
users(id, email, start_weight, current_weight, goal, sex, activity_degree)

Development

just setup          # uv sync --all-extras
just test           # run all tests
just test -v        # verbose
just lint           # ruff check src/ tests/
just fmt            # ruff format src/ tests/
just build          # uv build
just clean          # remove build artifacts

License

MIT