FyloCore

Real-time, collaborative knowledge-graph engine powering the next wave of open science and collective intelligence.

Fylo Core is the open-source heart of Fylogenesis—our mission to give researchers a platform that connects ideas across disciplines, augments human insight with AI, and keeps collaboration genuinely live.

We’re releasing the core so anyone can self-host, extend, and help shape a future where knowledge grows as a shared graph instead of siloed PDFs.

<p align="center"> <img src="https://fylo.io/fylo_banner_light.png" alt="FyloLabs"> </p> <div align="center">

</div>

✨ Key capabilities

• Live collaboration – multi-cursor editing, conflict-free merges, and chat built on Socket.IO
• AI-assisted ingestion – documents chunked, embedded, and summarised via Anthropic/OpenAI pipelines
• Schema-aware graphs – Scientific-discourse schema out of the box; bring your own JSON-Schema to extend
• Batteries included – Next.js UI, TypeScript API, PostgreSQL + pgvector, Docker & Helm charts
• Open-core – permissive Apache-2.0 license; enterprise add-ons (SSO, RBAC, analytics) stay separate

🚀 Quick Start for Developers

1. Clone the Repository

git clone https://github.com/fylo-io/FyloCore.git
cd FyloCore

2. Set Up Environment Variables

Copy environment templates

# Copy environment templates
cp core-server/.env.example core-server/.env
cp core-ui/.env.example core-ui/.env

Edit core-server/.env with your configuration:

# Required: Database connection
POSTGRES_HOST=localhost
POSTGRES_PORT=5432
POSTGRES_DB=fylocore
POSTGRES_USER=postgres
POSTGRES_PASSWORD=your_postgres_password

# Required: Anthropic API key for AI functionality  
ANTHROPIC_API_KEY=your_anthropic_api_key

# Optional: JWT secret for authentication (auto-generated if not provided)
JWT_SECRET=your_jwt_secret

# Optional: Email service
RESEND_API_KEY=your_resend_api_key_for_emails

# Optional: OpenAI API (if using OpenAI features)
OPENAI_API_KEY=your_openai_api_key

# Application URLs
APP_URL=http://localhost:3000

Edit core-ui/.env with your configuration:

# Required: API connection to backend
NEXT_PUBLIC_API_URL=http://localhost:8000

# NextAuth configuration
NEXTAUTH_URL=http://localhost:3000
NEXTAUTH_SECRET=your_nextauth_secret_here

# Optional: Google OAuth
GOOGLE_CLIENT_ID=your_google_client_id
GOOGLE_CLIENT_SECRET=your_google_client_secret

# Optional: Zotero integration
ZOTERO_CLIENT_KEY=your_zotero_client_key
ZOTERO_CLIENT_SECRET=your_zotero_client_secret
ZOTERO_CALLBACK_URL=http://localhost:3000/api/zotero/callback

Note: You need a valid Anthropic API key and PostgreSQL database for the application to work properly.

Getting API Keys

1. Anthropic: Get your API key from console.anthropic.com
2. PostgreSQL: Use the provided Docker setup or your own PostgreSQL instance

3. Run with Docker

Quick Start (Recommended)

cd deploy
./start-docker.sh

Manual Docker Compose

cd deploy
docker-compose up --build

4. Access the Application

• Frontend: http://localhost:3000
• Backend API: http://localhost:8000
• API Documentation: http://localhost:8000/api-docs
• PostgreSQL: localhost:5433 (external port)

5. Stop the Application

cd deploy
./stop-docker.sh
# OR
docker-compose down

🛠️ What Gets Built

• PostgreSQL Database: Persistent data storage on port 5433
• Core Server (Backend): Node.js/TypeScript API on port 8000
• Core UI (Frontend): Next.js application on port 3000

📋 Prerequisites

• Docker and Docker Compose installed
• API key for Anthropic Claude
• PostgreSQL database (included in Docker setup)
• Internet connection for downloading dependencies

🔧 Development Notes

• Environment variables are loaded from core-server/.env and core-ui/.env
• Logs are saved to core-server/logs/
• Database data is persisted in Docker volumes
• The PostgreSQL database is automatically initialized with required extensions

📖 Documentation

• Detailed Docker Guide
• API Documentation
• Project Documentation

🛠️ Troubleshooting

Common Issues:

• Port conflicts: Ensure ports 3000, 8000, and 5433 are available
• Docker not running: Start Docker Desktop or Docker daemon
• Build failures: Try docker system prune to clean up Docker cache
• API connection errors: Verify your .env files have correct API keys

Need Help?

• Check the Detailed Docker Guide for more setup options
• Look at example environment variables in .env.example files