Connecting gods, ancients, and modern humans through a single unbroken lineage graph — growing forever.

An autonomous, self-growing family tree that connects every recorded human, historical figure, and mythological being into one continuous graph. The AI agent runs 24/7, discovering ancestors and descendants from Wikipedia and Wikidata, scoring confidence on every relationship, and resolving "genesis blocks" as more connections are found.

Key concepts:

• Genesis Blocks — Every new unconnected person gets a temporary root (G1, G2...). As the agent finds their ancestry, genesis blocks dissolve into the main tree
• Confidence Scoring — Every parent-child link has a 0–100% confidence score. Conflicting sources create branches, all shown with their %
• Auto-merge — When confidence ≥ 95%, a genesis block is automatically merged into the connected tree
• Source Links — Every relationship links back to its Wikipedia/Wikidata source
• Infinite Growth — The agent never stops. Every person discovered queues their parents and children for research

genesis-tree/
├── schema.sql              ← PostgreSQL schema + seed data
├── render.yaml             ← Render deployment config
├── .env.example            ← Environment template
├── backend/
│   ├── main.py             ← FastAPI app + all API endpoints
│   ├── agent.py            ← Autonomous agent (runs forever)
│   ├── init_db.py          ← One-time DB setup script
│   └── requirements.txt
└── frontend/
    └── index.html          ← Full UI (single file, no build step)

git init
git add .
git commit -m "Initial Genesis Tree"
git remote add origin https://github.com/YOUR_USERNAME/genesis-tree.git
git push -u origin main

1. Go to render.com and sign up/login
2. Click New → Blueprint
3. Connect your GitHub repo
4. Render will read render.yaml and automatically create:
   • A web service (FastAPI app + agent)
   • A PostgreSQL database (free tier)
5. Click Apply

Once deployed, open the Render shell for your web service and run:

cd backend && python init_db.py

This creates all tables and seeds categories + the G0 genesis root.

Your app is live. The agent starts automatically and begins growing the tree from 12 seed persons (Zeus, Abraham, Alexander the Great, Genghis Khan, etc.)

# 1. Create a PostgreSQL database
createdb genesis

# 2. Set up environment
cp .env.example .env
# Edit .env with your database URL

# 3. Install dependencies
cd backend
pip install -r requirements.txt

# 4. Initialize DB
python init_db.py

# 5. Start the server
uvicorn main:app --reload --port 8000

Open http://localhost:8000

1. Seeds 12 highly-connected historical figures on first run
2. Pulls the highest-priority job from the queue
3. Looks up the person on Wikidata (structured) → fallback to Wikipedia (text parsing)
4. Extracts father, mother, children with confidence scores
5. Writes relationships + source URLs to DB
6. Queues every discovered person for their own research
7. Checks if any genesis block can be merged (confidence ≥ 95%)
8. Repeats — forever, with a 2-second polite delay between requests

The graph grows exponentially: 12 seeds → ~60 parents found → ~300 grandparents → ...

Every genesis block in the system eventually connects to a single root — or is identified as its own named branch of humanity.

Progress metric: The number of open genesis blocks, going toward zero over time.

The system never fully reaches zero — but every merge is a step toward the most complete picture of human lineage ever assembled.

Built with FastAPI · PostgreSQL · Wikipedia/Wikidata APIs · No paid services required