Know where everything is. A personal inventory system with AI-powered image recognition.
CrowNest is a lightweight, self-hosted web app for cataloging your belongings with photos, descriptions, and storage locations. Runs on Mac, Linux, and Raspberry Pi — accessible from any device in your network.
- Smart Cataloging — Add items with photos, descriptions, categories, and tags
- AI Image Recognition — Snap a photo, let AI suggest name & description (Claude, OpenAI, or Ollama)
- Nested Locations — Organize hierarchically: House → Basement → Shelf → Box
- Visual Gallery — Browse all your items as a photo grid, filterable by location/category
- QR Codes — Generate & print QR codes for boxes, scan to see contents
- Quick Capture — Just snap a photo, AI does the rest
- Lending Tracker — Track who borrowed what and when
- Insurance Overview — Total value per location for home insurance
- Image Cropping — Crop photos right in the browser before saving
- Dark Mode — Easy on the eyes
- PDF & CSV Export — Generate inventory reports and backups
- PIN Protected — Simple PIN-based access control
- Fully Responsive — Works great on phone and desktop
git clone https://github.com/yourusername/crownest.git
cd crownest
pip install -r requirements.txt
python app.pygit clone https://github.com/yourusername/crownest.git
cd crownest
pip3 install -r requirements.txt
python3 app.pyThen open http://<your-ip>:5050 on any device in your network.
Default PIN: 1234 — Log in with this PIN on first start. A guided setup wizard will walk you through configuring locations, categories, and AI settings. You can change the PIN during setup or later in Settings.
Copy .env.example to .env and configure:
| Variable | Default | Description |
|---|---|---|
SECRET_KEY |
change-me |
Flask session secret (change this!) |
PIN |
1234 |
Access PIN |
AI_PROVIDER |
none |
claude, openai, ollama, or none |
ANTHROPIC_API_KEY |
— | For Claude Vision |
OPENAI_API_KEY |
— | For GPT-4o Vision |
OLLAMA_URL |
http://localhost:11434 |
Ollama server URL |
OLLAMA_MODEL |
llava |
Ollama vision model |
Settings can also be changed in the app under Settings.
| Provider | Cost | Speed | Quality | Offline |
|---|---|---|---|---|
| Claude (Sonnet) | ~$0.01/image | Fast | Excellent | No |
| OpenAI (GPT-4o) | ~$0.01/image | Fast | Excellent | No |
| Ollama (LLaVA) | Free | Slower | Good | Yes |
- Backend: Python + Flask
- Database: SQLite
- Frontend: Tailwind CSS + Vanilla JS
- Image Processing: Pillow + Cropper.js
MIT — see LICENSE