Graft
Graph visualization for Craft documents
Craft Extensions Hackathon Winner
Graft turns your Craft document space into an interactive graph. Explore connections between documents, tags, and folders in 2D or 3D — all processed locally in your browser with no data ever leaving your device.
Sponsor
If Graft is useful to you, consider leaving a tip to support development.
$
For self-sustainability reasons we accept $5+ donations
Features
- Interactive graph: Navigate your document space as a force-directed graph with zoom, pan, and click-to-open
- 3D visualization: Switch between 2D and 3D views for complex spaces
- Tag and folder linking: See how documents connect through shared tags and folder hierarchy
- AI summaries: Generate document summaries using your own API key (OpenAI-compatible)
- Bulk tag rename: Rename tags across your entire space in one operation
- Progressive fetching: Load large spaces incrementally without timeouts
- Deeplinks: Click any node to open the document directly in Craft
- Mobile-friendly: Responsive layout works on tablets and phones
- Privacy-first: Everything runs in your browser — no server-side storage
Getting started
Prerequisites
- A Craft account with documents
- Craft API credentials (API URL and Key) from the Imagine extension
How it works
- Open graft.to
- Enter your Craft API URL and API Key
- Graft fetches your document structure via the Craft API
- Your space appears as an interactive graph — explore connections, click nodes to open docs
Credentials are stored in your browser's localStorage only. Graft never sends them to any server.
Privacy & security
- Local-first: All data processing happens in your browser
- No data retention: Graft does not store or log any document content
- Proxy pass-through: API calls route through a minimal proxy to handle CORS — no data is stored
- Credentials in localStorage only: Your API key never leaves your device except to call the Craft API
Open source
Graft is MIT-licensed and fully open source. The graph visualization library is designed to be reusable for other knowledge-graph applications.