tauri-plugin-mcp
Enables AI assistants to automate and test Tauri desktop applications through the Model Context Protocol. It provides tools for app management, UI interaction, and state inspection across multiple platforms without requiring CDP dependencies.
README
tauri-plugin-mcp
Cross-platform Tauri test automation plugin via MCP (Model Context Protocol).
Enables AI assistants like Claude to interact with your Tauri desktop app for testing and automation.
Features
- Cross-platform: Windows (Named Pipes) + macOS/Linux (Unix Sockets)
- No CDP dependency: Works on all WebView backends including macOS WKWebView
- MCP integration: Direct integration with Claude Code and other MCP clients
Prerequisites
- Node.js >= 18
- Tauri v2.x
- pnpm (recommended) or npm
- Rust with cargo
Quick Start
- [ ] Add Rust plugin to
src-tauri/Cargo.toml - [ ] Install npm package:
pnpm add github:DaveDev42/tauri-plugin-mcp#main - [ ] Register plugin in
src-tauri/src/lib.rs - [ ] Add
mcp:defaultpermission - [ ] Initialize bridge in
main.tsx - [ ] Create
.mcp.jsonfor Claude Code
Installation
1. Rust Plugin (src-tauri/Cargo.toml)
[dependencies]
tauri-plugin-mcp = { git = "https://github.com/DaveDev42/tauri-plugin-mcp" }
2. Frontend API (package.json)
pnpm add github:DaveDev42/tauri-plugin-mcp#main
3. MCP Server
The MCP server is included in the package at:
node_modules/tauri-plugin-mcp/packages/tauri-mcp/dist/index.js
Setup
1. Register the plugin (src-tauri/src/lib.rs)
pub fn run() {
tauri::Builder::default()
.plugin(tauri_plugin_mcp::init())
.run(tauri::generate_context!())
.expect("error while running tauri application");
}
2. Add permissions
Option A: In tauri.conf.json or config/*.json5 (recommended)
{
"security": {
"capabilities": [{
"identifier": "main-capability",
"windows": ["main"],
"permissions": ["core:default", "mcp:default"]
}]
}
}
Option B: Separate file (src-tauri/capabilities/default.json)
{
"$schema": "../gen/schemas/desktop-schema.json",
"identifier": "default",
"windows": ["main"],
"permissions": ["core:default", "mcp:default"]
}
3. Initialize the bridge (main.tsx)
// Initialize MCP bridge for E2E testing (dev mode only)
if (import.meta.env.DEV) {
import('tauri-plugin-mcp').then(({ initMcpBridge }) => {
initMcpBridge().catch(err => {
console.warn('[MCP] Bridge initialization failed:', err);
});
});
}
MCP Server Configuration
Add to .mcp.json in your project root:
{
"mcpServers": {
"tauri-mcp": {
"command": "node",
"args": ["node_modules/tauri-plugin-mcp/packages/tauri-mcp/dist/index.js"],
"env": {
"TAURI_PROJECT_ROOT": "."
}
}
}
}
Available Tools
| Tool | Parameters | Description |
|---|---|---|
app_status |
- | Check if app is running |
launch_app |
wait_for_ready?: boolean, timeout_secs?: number, features?: string[] |
Launch Tauri app via pnpm tauri dev |
stop_app |
- | Stop the app |
snapshot |
- | Get accessibility tree (returns ref numbers) |
click |
ref?: number, selector?: string |
Click element by ref or CSS selector |
fill |
ref?: number, selector?: string, value: string |
Fill input field |
press_key |
key: string |
Press keyboard key |
navigate |
url: string |
Navigate to URL |
screenshot |
- | Take screenshot (uses html2canvas) |
evaluate_script |
script: string |
Execute custom JavaScript |
get_console_logs |
- | Get console logs |
get_network_logs |
- | Get network logs |
Using features parameter
To launch with Cargo features:
launch_app({ features: ["my_feature"] })
This runs: pnpm tauri dev --features my_feature
Usage Example
Typical testing workflow:
1. launch_app({ timeout_secs: 120 })
2. snapshot() # Get element refs
3. click({ ref: 5 }) # Click button by ref
4. fill({ selector: "input[name='email']", value: "test@example.com" })
5. screenshot() # Verify result
6. stop_app()
How It Works
Claude Code <-> MCP Server <-> Socket <-> Tauri Plugin <-> JS Bridge <-> Your App
- Rust Plugin creates IPC server (Unix socket or Windows named pipe)
- MCP Server connects to IPC and exposes tools to Claude
- JS Bridge (
initMcpBridge()) enables DOM operations in WebView
Socket Paths
- Unix:
{project_root}/.tauri-mcp.sock - Windows:
\\.\pipe\tauri-mcp-{hash}(hash derived from project path)
Troubleshooting
"MCP bridge not initialized"
The JS bridge isn't running. Check:
initMcpBridge()is called in your frontend code- App is running in dev mode (
import.meta.env.DEV) - Check browser console for initialization errors
Socket connection failed
- Ensure the app is running (
launch_appfirst) - On Windows, check pipe path in logs:
[tauri-plugin-mcp] full_path: \\.\pipe\tauri-mcp-XXXXX - On Unix, check if
.tauri-mcp.sockexists in project root
App launch timeout
- Increase
timeout_secs(default: 60) - Check if
pnpm tauri devworks manually - Look for build errors in terminal output
snapshot returns empty
- Wait for app to fully load (use
wait_for_ready: true) - Check if bridge initialized (look for
[MCP]logs in console)
License
MIT OR Apache-2.0
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.