🍽️ Mealbot — A Discord Food Tracking Bot

A Discord bot to help you log meals, track your weekly nutrition goals, and get AI-powered insights—all powered by Google Gemini AI and SQLite. Whether you drop an image or type a description, Mealbot breaks down your food groups, stores your data, and gives you personalized tips and recommendations. 🥗🍎

---

🔗 Invite the Bot!

Click here to invite Mealbot to your server!

---

🚀 Key Features

• Meal Logging (Server & DM)
  • Upload an image or send a description of your meal in the #food channel or DM the bot directly.
  • Bot responds with a breakdown of how that meal contributes to your weekly nutrition goals (as decimal fractions).

• AI-Powered Classification
  • Uses Google Gemini (“gemini-1.5-flash-8b”) to classify each meal into six food groups:

  • Fruits
  • Vegetables
  • Grains
  • Protein
  • Dairy
  • Oils
    • Returns a JSON-only response indicating the fractional contribution toward each’s weekly target.

• Personalized Meal Tip
  • After logging a meal, Mealbot crafts a short, friendly tip that:

  1. Mentions the specific food you logged (e.g., “Given that you had spaghetti…”).
  2. Praises healthy aspects (e.g., high vegetable content).
  3. Suggests a complementary action (e.g., “Consider drinking water…”).

• Weekly Progress Report
  • Use !foodreport (in #food or via DM) to get a detailed, visual progress report.
  • Progress bars indicate what percentage of each weekly goal you’ve achieved—aligned uniformly for readability.

• Food Recommendations
  • Use !recommend to receive AI-generated suggestions on how to balance any incomplete food-group targets.
  • Recommendations are delivered as a simple JSON array converted to a friendly message.

• SQLite Database
  • All meal logs (timestamps, user ID, image URL, description, classification JSON) are stored locally in foodbot.db.
  • Data persists across restarts, ensuring you never lose your progress.

---

📂 Project Structure

.
├── bot.py                 # Main bot logic (classification, tips, reports, commands)
├── requirements.txt       # Python dependencies
├── Dockerfile             # Docker configuration
├── docker-compose.yaml    # Docker Compose setup
├── .env.bak               # Example environment variables
├── .gitignore             # Git ignore rules
└── README.md              # This documentation

---

🐳 Running with Docker

1. Build & Run

   docker-compose up --build -d

This will:

• Build a Docker image named mealbot.
• Launch the bot container in detached mode.

2. Stop the Bot

   docker-compose down

---

🐳 Docker Setup (Development)

Build Image

docker build -t mealbot .

Force Rebuild (no cache)

docker build --no-cache -t mealbot .

---

💻 Local Setup (Development)

1. Clone Repository

   git clone https://github.com/PhysCorp/Mealbot.git
   cd Mealbot

2. Environment Variables Create a .env file in the project root:

   DISCORD_TOKEN=your_discord_bot_token
   GEMINI_API_KEY=your_google_gemini_api_key

3. Install Dependencies

   pip install -r requirements.txt

4. Run Locally

   python bot.py

---

🔧 Available Commands

Use these commands either in the #food channel or via a Direct Message (DM) to the bot.

Command	Description
!foodreport	View your weekly progress across all food groups (with uniform progress bars).
!recommend	Get AI-powered suggestions on how to balance incomplete food-group targets (JSON → friendly).

---

📝 How It Works

1. Log a Meal

   • In #food: Upload an image or type a description.
   • Via DM: Send a message with an image attachment or a text description.

2. AI Classification & Tip

   • Bot sends typing indicator, then:

     1. Breaks down “fractions of weekly goals” for each food group.
     2. Crafts a personalized tip that explicitly mentions your food (e.g., “Your kale salad is an excellent vegetable choice…”).
     3. Stores data in foodbot.db.

3. Weekly Progress

   • Use !foodreport to see progress bars for: Fruits, Vegetables, Grains, Protein, Dairy, Oils.
   • Progress bars are left-aligned for consistent display regardless of label length.

4. Recommendations

   • Use !recommend to receive a set of tailored food ideas to complete any incomplete group targets.

---

🌟 Example Interaction

User (in #food): Today I ate a bowl of kale salad
Bot (breakdown):
📝 Current Meal Breakdown (fractions of weekly goals):
- Fruits      : 0.00
- Vegetables  : 0.12
- Grains      : 0.00
- Protein     : 0.02
- Dairy       : 0.00
- Oils        : 0.01

Bot (tip):
💡 **Meal Tip:** Your kale salad is packed with vegetables and fiber—great job! Consider drinking a glass of water afterward to stay hydrated.

Bot (weekly progress):
🍽️ **Weekly Progress:**
- **Fruits     **: [░░░░░░░░░░] 0% of weekly goal
- **Vegetables **: [███░░░░░░░] 30% of weekly goal
- **Grains     **: [░░░░░░░░░░] 0% of weekly goal
- **Protein    **: [█░░░░░░░░░] 10% of weekly goal
- **Dairy      **: [░░░░░░░░░░] 0% of weekly goal
- **Oils       **: [░░░░░░░░░░] 0% of weekly goal

---

📦 Dependencies

• discord.py
• python-dotenv
• google-genai
• requests
• SQLite3 (bundled with Python)

---

🔐 Configuration & Secrets

Place a .env file in the project root with:

DISCORD_TOKEN=your_discord_bot_token
GEMINI_API_KEY=your_google_gemini_api_key

Note: Never commit your actual tokens or API keys to source control.

---

📝 License

This project is licensed under the MIT License. See the LICENSE file for details.

---