MCP Serverfor Tauri
An MCP server that empowers AI assistants to build, test, and debug Tauri applications.
Skip to content
Visual Context
Capture and analyze screenshots to understand UI state and help with visual debugging
Live Debugging
Access console logs, window state, and system logs in real-time
IPC Monitoring
Capture and inspect Tauri IPC calls between frontend and backend in real-time
Device Management
List Android emulators and iOS simulators for mobile testing
WebView Automation
Click, type, scroll, find elements, and verify UI state in your app's webview
Plugin Bridge
Execute IPC commands and interact with the Tauri plugin system
Community Project
This is an unofficial community project, independently developed to enhance Tauri® development through AI assistance. Please note that this project is not affiliated with, endorsed by, or sponsored by The Tauri Programme within the Commons Conservancy.
What Is This?
MCP Server for Tauri® bridges AI assistants with your Tauri development environment via the Model Context Protocol. Control your entire dev workflow through natural language - test UI, debug issues, monitor IPC calls, and inspect your app's state.
Quick Start
1. Prerequisites
- Node.js 20+ and npm
- Rust and Cargo
- Tauri CLI:
npm install -g @tauri-apps/cli@next
2. Add the MCP Bridge Plugin to Your Tauri App
Add the Rust crate (from your src-tauri directory):
bash
cargo add tauri-plugin-mcp-bridgeOr manually add to Cargo.toml: tauri-plugin-mcp-bridge = "0.7"
Register in src-tauri/src/main.rs:
rust
fn main() {
let mut builder = tauri::Builder::default();
#[cfg(debug_assertions)]
{
builder = builder.plugin(tauri_plugin_mcp_bridge::init());
}
builder
.run(tauri::generate_context!())
.expect("error while running tauri application");
}Required Configuration
You must enable withGlobalTauri in your tauri.conf.json:
json
{
"app": {
"withGlobalTauri": true
}
}This exposes window.__TAURI__ which the MCP bridge plugin requires to communicate with your app.
Optional: TypeScript Bindings
The npm package @hypothesi/tauri-plugin-mcp-bridge is optional. It provides TypeScript bindings if you want to call the plugin from your app's frontend code. The MCP server communicates with the Rust plugin directly via WebSocket—no npm package needed.
3. Configure Your AI Assistant
Use install-mcp to add the server to your AI assistant:
bash
npx -y install-mcp @hypothesi/tauri-mcp-server --client claude-codeSupported clients: claude-code, cursor, windsurf, vscode, cline, roo-cline, claude, zed, goose, warp, codex
Manual Configuration
If you prefer to configure manually, add to your MCP config:
json
{
"mcpServers": {
"tauri": {
"command": "npx",
"args": ["-y", "@hypothesi/tauri-mcp-server"]
}
}
}Config file locations:
- Claude Code: Cmd/Ctrl+Shift+P → "MCP: Edit Config"
- Cursor:
Cursor Settings→MCP→New MCP Server - VS Code: Add to
settings.jsonundermcp.servers - Windsurf: Cascade pane → MCPs icon → settings icon
- Cline: See Cline MCP configuration guide
Architecture
The MCP server communicates with your Tauri application through:
- Plugin Client (WebSocket port 9223) - Native IPC for UI automation, DOM interaction, and direct commands via mcp-bridge plugin
Slash Commands (Prompts)
The server provides slash commands (prompts) for guided workflows:
/setup - Set up the MCP bridge plugin in your Tauri project
/fix-webview-errors - Find and fix JavaScript errors in your webview
17 Powerful Tools
The server exposes tools across 4 categories:
Setup & Configuration (1 tool) - Guided setup instructions
UI Automation & WebView (10 tools) - Gestures, screenshots, JS execution, element finding, logs
IPC Tools (5 tools) - IPC commands, monitoring, events
Mobile Development (1 tool) - Device listing