Swarm Intelligence Engine

The Swarm Intelligence Engine is a high-fidelity sociological simulation tool designed to model public opinion and societal reactions in Georgia. By leveraging custom-crafted, culturally authentic AI personas and Google's Gemini models, it simulates how different demographic cohorts react to political news, policy updates, or social events.

This is a local, self-hostable version with no payment restrictions or web3 dependencies. You only need to provide your own Google Gemini API key to run simulations and generate personas.

---

Key Features

• Swarm Simulation: Select from a diverse library of pre-seeded Georgian personas (representing different ages, locations, employment statuses, and political views) and see their realistic reactions.
• Semantic & Strategic Analysis: Generate an advanced strategic intelligence report detailing narrative clusters, demographic intersections, and sentiment analyses of the swarm's reactions.
• Persona Generator: Describe a cohort of people (e.g., "Tech students in Kutaisi" or "Pensioners in Kakheti") and generate highly customized, culturally authentic personas dynamically to expand your library.
• Local-First & Secure: Built using SvelteKit, Drizzle ORM, and SQLite. Your Gemini API key is stored locally in your browser cookies (or your server .env file) and sent directly to Google's API endpoints.

---

Getting Started

1. Prerequisites

You will need Bun installed on your machine.

2. Installation

Clone the repository and install the dependencies:

# Clone the repository
git clone https://github.com/your-username/SWARM.git
cd SWARM

# Install dependencies using Bun
bun install

3. Database Setup

The repository comes pre-packaged with a pre-seeded sqlite.db file in the root directory, which contains a rich set of initial Georgian personas.

If you ever need to recreate or reset the database schema, you can run Drizzle migrations:

bun x drizzle-kit push

4. Configure your Gemini API Key

You can configure your API key in one of two ways:

Option A: Via the Web UI (Recommended)

1. Start the application (see below).
2. Click the Settings (gear) icon in the top-right corner of the header.
3. Paste your Gemini API Key (AIzaSy...) and click Save Changes.
4. The key is securely saved in a browser cookie (gemini_api_key) and used for SvelteKit server-side API calls.

Option B: Via .env File

1. Copy .env.example to .env:

   cp .env.example .env

2. Open .env and paste your API key:

   GEMINI_API_KEY="your-gemini-api-key-here"

---

Running the Application

To run the engine in development mode:

bun run dev

Open your browser and navigate to http://localhost:5173.

Building for Production

To create a production-optimized build:

bun run build

Preview the production build locally:

bun run preview

---

Tech Stack

• Framework: SvelteKit (Svelte 5)
• Styling: Vanilla CSS with Tailwind CSS v4
• Database: SQLite with Drizzle ORM
• LLM Integration: Google GenAI SDK (@google/genai) using Gemini 3.1 models
• Icons: Lucide Svelte

---

License

This project is open-source and free to use. Set up your API key, run it locally, and start exploring collective intelligence simulations!