Skip to content

shantanudeshp/dnd-assistant

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

D&D Assistant

Overview

D&D Assistant is an AI-powered tool designed to enhance the Dungeons & Dragons gaming experience. Utilizing the Llama 3.1 model and the Cerebras API for ultra high-speed inference, it acts as a dynamic Dungeon Master, managing game states, responding to player actions, and facilitating an interactive storytelling environment.

Features

  • Dynamic Game State Management: Tracks player stats, enemies, items, and locations.
  • AI-Powered Responses: Generates contextual and coherent responses based on the current game state.
  • Extensible Data Structures: Easily add new enemies, items, and locations via JSON files.
  • Interactive CLI: Simple command-line interface for engaging with the game.
  • Game Reset Functionality: Reset the game to its initial state at any time for consistent starting points.

Prerequisites

  • Python 3.9+
  • Git
  • A Cerebras API key

Sample Use

Narrative: Varric the Vigilant stands at the edge of a dense forest clearing, his rusty sword gleaming faintly in the sunlight. The air is thick with the scent of pine and earth. Suddenly, a Goblin Grunt emerges from the shadows, eyeing Varric warily.

Response: attack

Narrative: Varric the Vigilant swings his rusty sword at the Goblin Grunt, striking it with force. The goblin recoils, losing 5 HP.

Response: drink potion

Narrative: Varric the Vigilant drinks the health potion from his inventory, feeling a warm sensation spread through his body. His health increases by 10 HP.

Response: investigate cave

Narrative: Varric the Vigilant moves cautiously towards the dark cave. The sounds of dripping water grow louder as he approaches, and the shadows seem to dance ominously.

About

project for the 2024 Cerebras Fellowship Program https://cerebras.ai/fellows

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages