Grafyn

A desktop knowledge graph with full-text search, multi-LLM canvas, and MCP integration.
Windows · macOS · Linux

Download · Features · Quick Start · MCP for Claude

Early development — expect rough edges. Bug reports welcome via Issues.

---

Features

Knowledge Management

• Markdown notes with [[wikilinks]] and YAML frontmatter (Obsidian-compatible)
• Full-text search powered by Tantivy with graph-aware ranking
• Backlink graph with force-directed D3.js visualization
• Note workflows — Draft, Evidence, Canonical status progression

Multi-LLM Canvas

• Compare models side-by-side — send one prompt to GPT-4, Claude, Gemini, and 100+ others simultaneously
• Real-time streaming — parallel responses via OpenRouter
• Semantic note context — automatically retrieves relevant notes as LLM context
• Debate mode — models critique and build on each other's responses
• Infinite canvas — drag, zoom, and pan with D3.js
• Smart web search — auto-detects queries that benefit from live web results

AI-Powered Tools

• Distillation — split large notes into atomic notes and topic hubs using LLM or rules-based extraction
• Link Discovery — AI suggests connections between notes (Zettelkasten-style)
• Conversation Import — bring in chat history from ChatGPT, Claude, Grok, and Gemini

MCP Integration

• Native Rust MCP server bundled with the app for Claude Desktop
• 10 tools: note CRUD, search, backlinks, outgoing links, recall, and conversation import
• Zero-config — copy the config snippet from Grafyn Settings into Claude Desktop

Quick Start

Download

Grab the latest installer from Releases:

Platform	File
Windows (64-bit)	Grafyn_*_x64-setup.exe
Windows (ARM)	Grafyn_*_arm64-setup.exe
macOS (Apple Silicon)	Grafyn_*_aarch64.dmg
Linux (Debian/Ubuntu)	grafyn_*_amd64.deb
Linux (Universal)	grafyn_*_amd64.AppImage

Grafyn auto-updates after installation.

Build from Source

Prerequisites: Node.js 20+, Rust via rustup, and Tauri v1 dependencies.

cd frontend
npm install
node scripts/generate-icons.cjs    # first build only

npm run tauri:dev                   # dev mode with hot reload
npm run tauri:build                 # production build

Configuration

On first launch, Grafyn walks you through setup:

1. Vault path — where your markdown notes are stored (default: ~/Documents/Grafyn/vault/)
2. OpenRouter API key — required for Canvas, distillation, and link discovery (get one here)

All data stays on your machine. No account needed.

MCP Integration

Connect Claude Desktop to your knowledge base:

1. Open Grafyn Settings and copy the MCP config snippet
2. Add it to your claude_desktop_config.json:

{
  "mcpServers": {
    "grafyn": {
      "command": "path/to/grafyn-mcp",
      "args": ["--vault", "path/to/vault", "--data", "path/to/data"]
    }
  }
}

The MCP server is a standalone Rust binary (grafyn-mcp) bundled with the installer. It shares the same vault and search index as the desktop app.

Architecture

Single binary, no server — Tauri wraps a Vue 3 frontend with a Rust backend.

Tauri Desktop App
├── Vue 3 Frontend (WebView)
│   └── Tauri IPC (invoke)
├── Rust Backend
│   ├── 65 IPC commands across 13 modules
│   ├── Tantivy full-text search
│   ├── petgraph link graph
│   └── OpenRouter LLM integration
├── grafyn-mcp (bundled MCP server)
└── ~/Documents/Grafyn/
    ├── vault/  (markdown notes)
    └── data/   (search index, canvas sessions, settings)

Tech Stack

Layer	Technology
Frontend	Vue 3, Vite, Pinia, D3.js
Desktop	Tauri 1.8
Search	Tantivy 0.22
Graph	petgraph 0.6
LLM	OpenRouter via reqwest
MCP	rmcp (stdio transport)
Updates	Cloudflare R2 + Workers

Contributing

1. Fork and create a feature branch
2. Make changes and add tests
3. Run cargo test and npm run test:run
4. Submit a pull request

See CLAUDE.md for detailed architecture docs, IPC command reference, and CI pitfalls.