PokeDex IRL is an open-source application that identifies real-world animals, insects, and sea life from images and generates creative, Pokédex-style entries using AI and LangGraph agents. It features a modern React UI and a robust FastAPI backend.

• Image Upload & Validation: Upload images of animals, insects, or sea life for instant identification.
• AI-Powered Identification: Uses advanced LLMs and LangChain agents to identify creatures and generate detailed, Pokédex-style entries.
• REST API: FastAPI-based backend with endpoints for image upload, creature CRUD, search and identification.
• Database Storage: Stores creature data in SQLite using SQLModel/SQLAlchemy.
• Modern Web UI: Vite/React TypeScript frontend for a seamless Pokédex experience.
• Extensible Agents: Modular agent architecture for scanning and explanation, powered by LangGraph and LangChain.

• Backend: Python 3.12+, FastAPI, SQLModel, SQLAlchemy, Pydantic, Uvicorn, Loguru, LangChain, LangGraph
• Frontend: React 19+, TypeScript, Vite
• Database: SQLite
• AI/LLM: OpenAI-compatible LLMs (configurable)
• Containerization: Docker, Docker Compose

pokedex-irl/
├── docker/                  # Docker-related files
│   ├── nginx.conf           # Nginx config for reverse proxy
│   └── start.sh             # Entrypoint script for Docker container
├── pokedex-service/         # FastAPI backend, AI/LLM logic, database, API
│   ├── src/pokedex/
│   │   ├── agent/           # LangChain/LangGraph agents (scanner, explainer)
│   │   ├── creature/        # Creature models, CRUD, dependencies, utils
│   │   ├── config.py        # App configuration
│   │   ├── database.py      # DB setup and session
│   │   ├── llm.py           # LLM integration
│   │   └── main.py          # FastAPI app entrypoint
│   ├── pokedex.db           # SQLite database
│   └── pyproject.toml       # Python project config & dependencies
├── pokedex-ui/              # Vite/React TypeScript frontend
│   ├── src/
│   │   ├── App.tsx, CreatureList.tsx, CreatureCard.tsx, NavBar.tsx, ScanPopup.tsx, ...
│   └── package.json, vite.config.ts, tsconfig.json, ...
├── screenshots/             # Project screenshots
├── docker-compose.yml       # Docker Compose setup for backend + frontend
├── Dockerfile               # Multi-stage Dockerfile for building and running the app
├── .env.example             # Example environment variables
├── static/uploads/          # Uploaded images
├── README.md                # This file
└── LICENSE                  # Apache 2.0 License

• Python 3.12+
• Node.js 18+

cd pokedex-service
python -m venv venv
source venv/bin/activate  # On Windows: .\venv\Scripts\activate
pip install .
# Or, for development:
pip install -e .[test]

Create required directories:

mkdir -p static/uploads

Run the development server:

uvicorn src.pokedex.main:app --reload

API Documentation:

• Swagger UI: http://localhost:8000/api/v1/docs
• ReDoc: http://localhost:8000/api/v1/redoc

cd pokedex-ui
npm install
npm run dev

Open http://localhost:5173 in your browser.

Pokedex IRL can be run in a single Docker container using Docker Compose.

Ensure Docker and Docker Compose are installed. Modify the docker-compose.yml with your environment variables as needed (or create a .env file). Then run:

docker-compose up --build -d

This will start both the backend (FastAPI) and frontend (Vite/React) services in a single container. The application will be accessible at http://localhost:8080.

1. Start the backend and frontend servers as above.
2. Open the frontend in your browser (default: http://localhost:5173).
3. Upload an image to receive a Pokédex-style entry and detailed explanation.

• POST /api/v1/creature/identify — Upload an image to identify a creature and generate a Pokédex entry.
• GET /api/v1/creature/ — List all creatures in the database.
• GET /api/v1/creature/{id} — Get details for a specific creature.
• GET /api/v1/creature/search — Search creatures by name or description.

See Swagger UI for full API details.

Environment variables can be set in a .env file in pokedex-service/ (see src/pokedex/config.py for all options):

• MODEL_NAME, MODEL_API_KEY, MODEL_ENDPOINT — LLM configuration
• IMAGE_MODEL_NAME, IMAGE_MODEL_ENDPOINT, IMAGE_MODEL_API_KEY — Image model config

Contributions are welcome! Please open issues or pull requests for bug fixes, features, or improvements.

Apache License 2.0 — See LICENSE for details.