uniauto-mcp-server

UniAuto MCP Server

UniAuto Logo

smithery badge

Universal Test Automation MCP Server for web and desktop automation with self-healing capabilities and AI integration.

Cutting-Edge Features

UniAuto MCP Server incorporates several cutting-edge and futuristic features that position it at the forefront of test automation technology:

1. Multi-Layered Self-Healing Automation

The server’s 5-tier self-healing mechanism is revolutionary, using a sophisticated cascade of recovery strategies:

This multi-strategy approach creates extraordinarily resilient tests that can survive even major UI overhauls.

2. AI-Driven Test Generation & Maintenance

Integration with Claude and other LLMs enables:

3. MCP (Model Context Protocol) Integration

The MCP implementation is particularly forward-thinking:

4. Unified Web + Desktop Automation

The integration of both web testing (via Playwright) and desktop automation within a single framework provides:

5. Visual Intelligence

The implementation of visual matching provides:

Core Features

Impact on Test Automation Ecosystem

UniAuto MCP Server has the potential to transform the test automation landscape:

1. Maintenance Cost Reduction

The self-healing capabilities dramatically reduce the #1 pain point in test automation - maintenance costs. Studies suggest that organizations spend 30-40% of testing time on test maintenance. UniAuto could reduce this by 70-80%, freeing QA resources for more strategic work.

2. Democratization of Test Automation

The AI integration allows non-technical stakeholders to create tests using natural language, potentially expanding test automation adoption by:

3. Resilience Revolution

By solving the brittleness problem in modern automation, UniAuto shifts the entire ecosystem toward resilience-focused testing rather than selector-focused testing. This paradigm shift fundamentally changes how tests are written and evaluated.

4. AI Amplification in Testing

As an MCP-compatible system that works with Claude and other models, UniAuto positions automation as an “AI-native” technology, creating a foundation for:

5. Cross-Domain Testing Standards

By unifying web and desktop testing under one framework, UniAuto helps establish new standards for end-to-end testing that transcend the current siloed approach to different application types.

Installation

# Clone the repository
git clone https://github.com/yourusername/uniauto-mcp-server.git
cd uniauto-mcp-server

# Install dependencies
npm install

# Install Playwright browsers
npx playwright install

# Create environment file
cp .env.example .env

Configuration

Edit the .env file to configure the server:

# Server Configuration
PORT=3000
NODE_ENV=development

# Database (optional)
MONGODB_URI=mongodb://localhost:27017/uniauto

# Claude Configuration (required for AI processing)
CLAUDE_API_KEY=your_claude_api_key
CLAUDE_MODEL=claude-3-7-sonnet-20240229

Claude API Integration

UniAuto MCP Server uses the official Anthropic SDK to integrate with Claude:

const Anthropic = require('@anthropic-ai/sdk');

const anthropic = new Anthropic({
  apiKey: process.env.CLAUDE_API_KEY,
});

const response = await anthropic.messages.create({
  model: "claude-3-7-sonnet-20240229",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello, Claude" }],
});

Usage

# Start the server in development mode
npm run dev

# Start the server in production mode
npm start

API Endpoints

Health Check

Automation Commands

Test Case Management

AI Integration

Test Generation

Advanced Testing

MCP Integration

MCP Integration

This server is compatible with the Model Context Protocol (MCP) used by AI assistants like Claude. To use it with Claude or other AI models:

  1. Start the server
  2. Register the MCP server with the AI assistant
  3. The AI can now use the automation capabilities via the MCP protocol

Smithery.ai Integration

smithery badge

UniAuto MCP Server integrates seamlessly with Smithery.ai, making it easy to connect with Claude 3.7 and other AI assistants:

For seamless integration with AI assistants, you can use Smithery.ai:

# Install Smithery CLI
npm install -g @smithery/cli

# Connect UniAuto to Smithery
smithery connect uniauto-mcp-server

# Connect to an AI assistant
smithery connect --assistant claude

Once connected, you can use Claude Code, Claude Web, VSCode with Claude Extension, or Cursor to control UniAuto for automated testing.

See the Smithery Setup Guide for detailed instructions.

AI Assistant Integration

UniAuto works with various AI assistants:

Test Generation Capabilities

UniAuto MCP Server can automatically generate test code in various frameworks and styles:

Advanced Testing Capabilities

UniAuto MCP Server provides comprehensive testing capabilities beyond just generating test code:

See the Test Generation Guide for detailed API documentation and examples.

Self-Healing Capabilities

The server uses several strategies for self-healing selectors:

  1. Alternative selector repository
  2. Role-based selectors (Playwright specific)
  3. Looser CSS selector generation
  4. Visual element matching using Playwright’s built-in capabilities
  5. Nearest text-based element finding

Playwright Benefits

Playwright offers several advantages for test automation:

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

This site is open source. Improve this page.