Digital Dungeon Master (DDM) is an AI-powered role-playing game assistant that brings your D&D campaigns to life. It acts as an intelligent, adaptive Dungeon Master that responds to player actions, narrates adventures, and manages game mechanics - all powered by large language models.
DDM creates an immersive D&D experience by providing an AI Dungeon Master that:
- Narrates rich, dynamic storylines that adapt to player choices
- Manages combat encounters and NPC interactions
- Handles game mechanics like dice rolls and skill checks
- Maintains consistent world state and character information
- Generates atmospheric descriptions and engaging dialogue
- Interactive Chat Interface: Communicate naturally with the AI Dungeon Master
- Context-Aware Responses: The AI maintains campaign history and character states
- Structured Game Data: Tracks locations, player stats, enemy information, and more
- Dice Rolling System: Fair and transparent dice mechanics
- Command System: Special commands for game mechanics and meta-actions
When playing with DDM, keep these guidelines in mind:
- Play in Good Faith: While the AI can be manipulated, the experience is most enjoyable when players engage honestly with the story
- Clear Communication: Be specific in your actions and intentions to help the AI understand
- Respect Game Mechanics: Use proper commands for dice rolls and game actions
- Stay in Character: Immerse yourself in role-playing for the best experience
- Address NPCs directly: Make it very clear to the AI who you are talking to
- Ollama installed and running
- NVIDIA GPU recommended for optimal performance
- Docker (for production deployment)
- Go for local development
-
Frontend:
- HTMX for dynamic interactions
- Bootstrap CSS for styling
-
Backend:
- Go server
- LLM integration via Ollama
-
AI/ML:
- Large Language Models for game mastering
-
Clone the repository:
git clone https://github.com/yourusername/digital-dungeon-master.git cd digital-dungeon-master -
Install dependencies:
# Backend dependencies go mod download -
Start Ollama and ensure it's accessible
-
Start the development server:
# Start backend
air
5. Visit `http://localhost:3000` in your browser
## Production Deployment
1. Build the Docker image:
```bash
make build
- Command overwrite system
- Dual LLM chat system for GM and mechanics
- Token usage tracking
- Enhanced dice mechanics
- Structured data extraction
- Llamafile integration
- HTMX improvements
- Game art generation
- Comfy UI integration
- Full containerization