UniAuto MCP Server logo

UniAuto MCP Server

by samuelvinay91

UniAuto MCP Server is a universal test automation server for web and desktop applications, featuring self-healing capabilities and AI integration. It aims to revolutionize test automation by reducing maintenance costs and democratizing test creation.

View on GitHub

Last updated: N/A

What is UniAuto MCP Server?

UniAuto MCP Server is a universal test automation server that combines web and desktop automation with AI-driven features like self-healing selectors and test generation. It integrates with AI assistants via the Model Context Protocol (MCP) to enable natural language test creation and execution.

How to use UniAuto MCP Server?

To use UniAuto MCP Server, clone the repository, install dependencies, configure the environment variables (including the Claude API key), and start the server. You can then use the provided API endpoints to execute automation commands, manage test cases, generate tests with AI, and perform advanced testing like visual comparison and accessibility checks. It can also be integrated with Smithery.ai for seamless connection with AI assistants like Claude.

Key features of UniAuto MCP Server

  • Multi-Layered Self-Healing Automation

  • AI-Driven Test Generation & Maintenance

  • MCP (Model Context Protocol) Integration

  • Unified Web + Desktop Automation

  • Visual Intelligence

  • Automated test code generation in multiple frameworks and styles

  • Test case management and execution history

Use cases of UniAuto MCP Server

  • Reducing test maintenance costs by automatically adapting to UI changes.

  • Enabling non-technical users to create and execute tests using natural language.

  • Automating cross-domain testing workflows that span web and desktop applications.

  • Integrating AI assistants into the test automation process for intelligent test generation and execution.

FAQ from UniAuto MCP Server

What is MCP (Model Context Protocol)?

MCP is a standardized API that allows AI assistants to directly control testing infrastructure, enabling seamless integration between LLMs and web/desktop automation.

Which AI models are supported for test generation?

The server primarily supports Claude but is designed to be compatible with other LLMs as well.

What testing frameworks are supported for test code generation?

The server supports generating test code for Playwright, Cypress, Jest, Selenium, WebDriverIO, and more.

How does the self-healing mechanism work?

The server uses a 5-tier self-healing mechanism, including repository-based alternative selectors, role-based accessibility selectors, dynamic looser selector generation, visual element matching, and contextual text-proximity detection.

Is desktop automation supported?

Yes, the server provides desktop automation capabilities alongside web automation, allowing for cross-domain testing.