godzillanwk-cmd/AiVmAA
Domain:
natural language processingRecord type:
software
Creator:
god
Host:
African American world 🌍🌎 history
# Cultural Vault — Scaffold
This repository contains a minimal scaffold for a Cultural Memory Engine:
- Qdrant vector DB (qdrant)
- Ingest worker (ingest/) that fetches page images, runs Tesseract OCR, chunks text, and upserts vectors
- API (api/) — FastAPI for semantic search and LLM-backed Q&A
- Minimal React UI (ui/) for search and ask
- Backup script and systemd unit to manage docker-compose on boot
Usage:
1. Copy `.env.example` to `.env` and fill in non-secret values. Do NOT commit secrets.
2. Build and run with Docker Compose:
```bash
docker compose build
docker compose up -d
```
3. Run a small ingest to seed data:
```bash
docker compose run --rm ingest --query "African American" --max-titles 5 --max-pages-per-title 2
```
Security:
- Do not expose Qdrant (port 6333) publicly.
- Put the API behind HTTPS and authentication (Nginx + certbot or VPN).
- Secure `.env` with chmod 600 and do not commit it.