https://github.com/lightinn/protopolis
A multi-agent Ollama simulation in Rust
https://github.com/lightinn/protopolis
multi-agent-systems ollama-app rust simulation
Last synced: 3 months ago
JSON representation
A multi-agent Ollama simulation in Rust
- Host: GitHub
- URL: https://github.com/lightinn/protopolis
- Owner: LightInn
- Created: 2025-02-23T17:45:04.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-06-24T14:09:07.000Z (4 months ago)
- Last Synced: 2025-06-28T06:34:04.260Z (3 months ago)
- Topics: multi-agent-systems, ollama-app, rust, simulation
- Language: Rust
- Homepage:
- Size: 9.21 MB
- Stars: 7
- Watchers: 1
- Forks: 1
- Open Issues: 11
-
Metadata Files:
- Readme: README.md
- Contributing: .github/CONTRIBUTING.md
Awesome Lists containing this project
README
# 🤖 Protopolis
**A Rust-based AI multi-agent simulation framework (using Ollama)**
**Protopolis is a project that allows you to create and simulate interactions between AI agents in a colorful terminal environment. Agents can communicate with each other and with the user according to different states and energy levels.**
[](https://github.com/LightInn/protopolis/actions)
[](https://crates.io/crates/protopolis)
[](https://github.com/LightInn/protopolis)
[](https://crates.io/crates/protopolis)
![]()
> **⚠️ Warning:** This project is a work in progress (WIP) and will have frequent updates that may break various features.
## 📋 Features
- **🧠 Agent Simulation** with different states (Idle, Thinking, Speaking)
- **💬 Messaging System** between agents and with the user
- **🌈 Colorful Terminal Interface** for better visualization
- **⚡ Energy Management** for agents
- **🔄 Simple Commands** to control the simulation## 🛠️ Installation
> **⚠️ Note:** Make sure to have [Ollama](https://ollama.ai) installed with the `llama3.2` model before continuing.
```bash
git clone https://github.com/LightInn/protopolis
cd protopolis
cargo build --release
```## 🎮 Usage
To start the simulation:
```bash
cargo run --release
```### Available Commands
- `start` - Start the simulation
- `pause` - Pause the simulation
- `resume` - Resume the simulation
- `stop` - Stop the simulation
- `exit` - Exit the application
- `topic ` - Set a new discussion topic
- `msg ` - Send a message to a specific agent## 🏗️ Architecture
The project is built around several key components:
- **Agents** - Entities with states and behaviors
- **Messaging System** - Asynchronous communication between agents
- **User Interface** - Colorful display in the terminal
- **Simulation** - Orchestration of interactions## 🗺️ Roadmap
- [x] Basic terminal interface
- [x] Color system for agents
- [x] Communication between agents
- [ ] More complex agent personalities
- [ ] Save/load simulations
- [ ] Graphical visualization of interactions## 🤝 Contribution
Contributions are welcome! Feel free to open an issue or a pull request.
## 📚 Inspired By
- [TyniTroupe](https://github.com/microsoft/TinyTroupe) - LLM-powered multiagent persona simulation
## 🔌 Powered By
![]()
Ollama - Local LLM runner## License
MIT © 2025 Breval LE FLOCH