Console RPG — Text + Tactical Web Battle

Maintenance notice: development is paused for the near future; expect limited updates.

Requirements

• Python 3.12+
• Google Gemini API key with a model that has quota (set via environment)

Environment

Create a .env in the project root:

GEMINI_API_KEY=your_key_here

Structure

Console_RPG/
├── app.py                  # Text adventure server (Flask, port 8000)
├── webui.py                # Tactical battle UI server (Flask, port 5000)
├── DEF.py                  # Main D&D game logic and DM AI integration
├── gemini.py               # Gemini API wrapper with key rotation and retry logic
├── gemini_schema.py        # Pydantic models and JSON schemas for AI responses
├── room_manager.py         # Room state management for multiplayer sessions
├── config.py               # Game configuration (battlefield, player, enemies, rules)
├── battlefield_configs.py  # Predefined battle arenas and terrain configs
├── character_config.py     # Race/class stats, bonuses, and ability calculations
├── dnd_spells.py           # Spell definitions (level 1, 2, basic attacks)
├── prompts.py              # System prompts for the AI DM in multiple languages
├── translations.py         # Translation loading utilities
├── .env                    # Environment variables (GEMINI_API_KEY, etc.)
├── .env.example            # Example environment file
├── requirements.txt        # Python dependencies
├── pyproject.toml          # Project metadata and uv config
├── static/
│   ├── css/
│   │   ├── battle.css      # Tactical battle UI styles
│   │   ├── character.css   # Character creation styles
│   │   ├── dice.css        # Dice roll animations
│   │   ├── game.css        # Text adventure UI styles
│   │   ├── index.css       # Landing page styles
│   │   ├── style.css       # Common styles
│   │   └── styles.css      # Additional global styles
│   └── js/
│       ├── character.js    # Character creation logic
│       ├── combat.js       # Combat system frontend
│       ├── game.js         # Text adventure frontend
│       ├── hexgrid.js      # Hex grid pathfinding and rendering
│       ├── index.js        # Landing page logic
│       ├── notifications.js # Toast notifications system
│       ├── settings.js     # Settings panel
│       ├── translations.js # Frontend translation utilities
│       └── ui.js           # UI helper functions
├── templates/
│   ├── base.html           # Base template with common layout
│   ├── battle.html         # Tactical hex-grid battle view
│   ├── character.html      # Character sheet view
│   ├── create.html         # Character creation form
│   ├── game.html           # Text adventure chat interface
│   └── index.html          # Landing/room selection page
├── translations/
│   ├── en.json             # English translations
│   └── ru.json             # Russian translations
├── logs/                   # Application logs (auto-created)
├── saves/                  # Saved game files (auto-created)
└── README.md               # This file

Run servers

Text adventure (rooms, DM, combat handoff signal):

python app.py  # runs on http://localhost:8000

Tactical battle UI:

python webui.py  # runs on http://localhost:5000

How the handoff works

• During text play, when combat starts, the server responds with a redirect flag and enemy info. The frontend should navigate the player to the battle UI and seed the enemy (name, HP).
• After the tactical fight, the battle UI should POST the result back to the text server so the narrative can continue (enemy dead → story continues; player dead → offer character creation).

Playing the text adventure

• Choose language, create/join a room, and interact with the DM via text.
• Save/load via provided endpoints (see app UI).

Playing the tactical battle

• Starts when redirected from the text game or by visiting /battle in the battle app.
• Supports movement, attacks, spells, effects, and HP tracking for player/enemy.

Known limitations

• Development is currently paused; features and fixes may be delayed and not surely will be continued.
• Gemini responses depend on your model/quota; ensure your key has access to the configured model.