STRADS-CLAUDE is an advanced simulation system inspired by Douglas Lenat's STRADS (STRategic Automatic Discovery System). This project aims to create a flexible, knowledge-based simulation of possible futures, allowing for the exploration of complex scenarios and agent interactions.
- Dynamic event generation and processing
- Multiple agents with unique personalities and behaviors
- Complex world state management
- Belief system for agents
- Customizable reactions and outcomes
- Logging system with multiple levels (debug, info, warning, error)
- Natural Language Generation (NLG) for creating narrative summaries of simulations
- SWI-Prolog (version 8.0 or higher recommended)
- Clone the repository:
git clone https://github.com/aindilis/strads-claude.git
- Navigate to the project directory:
cd strads-claude
To run the simulation, use the following command in your terminal:
swipl -s simulation.pl
This will start the simulation and output the results, including the debug information and the generated "Annals" at the end.
You can customize various aspects of the simulation by modifying the following components:
atomic_personality/2
: Define personalities for agentsatomic_reaction/2
: Specify possible reactions to eventsatomic_outcome/2
: Define outcomes for specific actionsatomic_affect/2
: Determine which actors are affected by specific eventsatomic_belief_update/2
: Specify how beliefs are updated based on actions
The simulation output includes:
- Detailed logs of event processing, world state changes, and agent interactions
- A summary of each simulation cycle, including world state and actor beliefs
- "The Annals of Our Simulation": A narrative summary of key events in simple English
Contributions to STRADS-CLAUDE are welcome! Please feel free to submit pull requests, create issues, or suggest improvements.
This project is licensed under the GPLv3 License - see the LICENSE.md file for details.
- Inspired by Douglas Lenat's original STRADS concept
- Developed with major assistance from Anthropic's Claude 3.5 Sonnet LLM