Sherlock Holmes: AI-powered Interactive Adventure

Welcome to Sherlock Holmes: Interactive Adventure, an interactive text-based adventure game powered by AI! Step into the shoes of the world's greatest detective and solve mysteries generated dynamically by a Large Language Model (LLM) accessed via the Nebius API.

Experience Victorian London, gather clues, interrogate suspects, and use your deductive prowess – all narrated in the first-person perspective as if you are Sherlock Holmes.

Features ✨

• Play as Sherlock Holmes: Immerse yourself in the role with a unique first-person narrative style.
• AI-Generated Mysteries: Each case is dynamically created by an LLM, offering unique scenarios.
• Genre Selection: Choose from various genres like Mystery, Murder, Supernatural, Sci-Fi, and more for diverse cases.
• Interactive Gameplay: Your typed actions and dialogue directly influence the story's progression.
• Atmospheric Setting: Explore the foggy streets and gaslit alleys of Victorian London.
• Save & Load: Save your progress at any time and resume your investigation later.
• Case Tracking: The game automatically extracts and remembers key clues, suspects, locations, and items.
• In-Game Commands: Use /save, /quit, /delete, and /help for game management.

Live-Demo

https://suenkler-ai.de/sherlock

Requirements 📋

• Python 3.7+
• A Nebius AI API Key (obtainable from Nebius AI) or another OpenAI-compatible API.

Installation ⚙️

1. Clone the repository:

   git clone https://github.com./your-username/sherlock-terminal-case.git # Replace with your repo URL
   cd sherlock-terminal-case

2. Create a virtual environment (Recommended):

   • On macOS/Linux:

     python3 -m venv venv
     source venv/bin/activate

   • On Windows:

     python -m venv venv
     .\venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

Configuration 🔑

1. Create a .env file in the root directory of the project (sherlock-terminal-case).

2. Add your Nebius API key to the .env file like this:

   # .env file
   NEBIUS_API_KEY="YOUR_NEBIUS_API_KEY_HERE"

   Replace "YOUR_NEBIUS_API_KEY_HERE" with your actual key.

3. Security: Ensure your .env file is listed in your .gitignore file to prevent accidentally committing your API key to version control. A basic .gitignore should include:

   # .gitignore
   venv/
   __pycache__/
   *.pyc
   *.log
   .env
   data/saved_games/*.tmp # Ignore temp save files

Running the Game ▶️

1. Make sure you are in the project's root directory (sherlock-terminal-case) in your terminal and that your virtual environment (if created) is active.
2. Run the main script:

   python sherlock.py

3. Follow the on-screen prompts in the main menu to start a new case or load a saved one.

Gameplay 🎮

• The game will present narrative scenarios from Holmes' first-person perspective.
• When prompted with >, type what you (as Holmes) want to do or say. Examples:
  • Examine the muddy footprints near the window.
  • Ask Mrs. Hudson if she saw anyone suspicious enter the building.
  • Go to Scotland Yard.
  • "My dear Watson, observe the distinct lack of dust on this mantelpiece!"
• Use the following commands during gameplay:
  • /save: Saves your current game progress.
  • /quit: Quits the current game (prompts to save) and returns to the main menu.
  • /delete: Permanently deletes the current game's save file and returns to the main menu.
  • /help: Displays the available commands.
• The story progresses based on your inputs and the LLM's responses. New clues, suspects, locations, and items will be highlighted.

File Structure 📁

sherlock-terminal-case/
├── .env # Your API key (You create this)
├── .gitignore # Specifies intentionally untracked files
├── data/
│ └── saved_games/ # Stores saved game state (.json files)
├── logs/
│ └── terminal_game.log # Log file for debugging
├── prompts/
│ └── sherlock_system_prompt.txt # Core instructions for the LLM persona
├── requirements.txt # Python dependencies
├── sherlock.py # Main game application script
└── sherlock_llm_handler.py # Module for handling LLM communication

Customization 🔧

You can tailor the AI's behavior by editing the prompts/sherlock_system_prompt.txt file. This file contains the core instructions given to the LLM regarding its role, narrative style, required elements, and constraints.

Contributing 🤝

Contributions are welcome! If you have suggestions for improvements or find bugs, please feel free to open an issue or submit a pull request.

License 📄

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