Web-QA
An AI-powered MCP server that automates web testing workflows by enabling recording, execution, and discovery of tests through natural language prompts.
What is Web-QA?
Web-QA is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ai-powered mcp server that automates web testing workflows by enabling recording, execution, and discovery of tests through natural language prompts.
An AI-powered MCP server that automates web testing workflows by enabling recording, execution, and discovery of tests through natural language prompts.
This server falls under the Browser Automation category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- An AI-powered MCP server that automates web testing workflow
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx web-qaConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Web-QA
Web-QA is an AI-powered MCP server built on the VibeShift platform that automates web testing workflows through natural language, enabling developers to record Playwright-based test scripts, execute them deterministically, discover test scenarios on live sites, and perform security scans — all without writing test code by hand. It is particularly suited for vibe-coding workflows where AI is generating application code and needs to simultaneously generate and run regression tests to validate that changes do not break existing functionality.
Prerequisites
- Python 3.10+ with pip and virtualenv support
- Node.js and npm (for Playwright/Patchright browser automation)
- An LLM API key (e.g., OpenAI) set as LLM_API_KEY in a .env file
- An MCP-compatible client such as Claude Desktop or a VS Code extension with MCP support
- Git to clone the VibeShift repository (Web-QA runs from the same codebase)
Clone the VibeShift repository
Web-QA is part of the VibeShift project. Clone the repository and navigate into it.
git clone https://github.com/GroundNG/VibeShift
cd VibeShiftSet up the Python environment and install dependencies
Create a virtual environment, activate it, install Python dependencies, and then install Patchright browser binaries used for Playwright automation.
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
pip install -r requirements.txt
patchright install --with-depsConfigure your LLM API key
Copy the environment example file and set your LLM API key. This is used for AI-assisted test discovery and analysis.
cp .env.example .env
# Edit .env:
# LLM_API_KEY="your-openai-or-other-llm-api-key"Add the Web-QA MCP server to your client configuration
Register the server in your MCP client configuration. It runs via uv from the cloned repository directory.
{
"mcpServers": {
"web-qa": {
"command": "uv",
"args": ["--directory", "/path/to/VibeShift", "run", "mcp_server.py"]
}
}
}Restart your MCP client and start testing
Restart Claude Desktop or your MCP client. Ask it to list available recorded tests, then try recording your first test scenario by describing user actions in plain language.
Web-QA Examples
Client configuration
Claude Desktop config launching the Web-QA / VibeShift MCP server via uv.
{
"mcpServers": {
"web-qa": {
"command": "uv",
"args": ["--directory", "/Users/yourname/VibeShift", "run", "mcp_server.py"]
}
}
}Prompts to try
Example prompts for recording tests, running regressions, discovering test scenarios, and scanning for vulnerabilities.
- "Record a test: go to https://myapp.local/login, type 'testuser' in the username field, type 'Pass123!' in the password field, click the Login button, verify 'Dashboard' appears"
- "Run the regression test output/test_login_20240601_120000.json"
- "Discover potential test steps starting from https://myapp.local/"
- "List all available recorded web tests"
- "Analyze the login form at https://myapp.local/login for XSS vulnerabilities"Troubleshooting Web-QA
Browser automation fails or Patchright cannot find browsers
Run 'patchright install --with-deps' again inside the activated virtual environment to reinstall browser binaries. On Linux, this also installs required system libraries. Ensure you are running from the venv where patchright is installed.
Test recording produces empty or incorrect JSON test files
Describe actions precisely in your prompt, including exact field names, button labels, and expected visible text to verify. Vague descriptions like 'fill in the form' can result in incomplete recordings. Reference the exact HTML element labels where possible.
LLM API key errors during discovery or analysis
Confirm the .env file is in the root of the VibeShift repository and contains LLM_API_KEY with a valid key. The uv runner loads .env automatically from the working directory specified in --directory.
Frequently Asked Questions about Web-QA
What is Web-QA?
Web-QA is a Model Context Protocol (MCP) server that ai-powered mcp server that automates web testing workflows by enabling recording, execution, and discovery of tests through natural language prompts. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Web-QA?
Follow the installation instructions on the Web-QA GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Web-QA?
Web-QA works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Web-QA free to use?
Yes, Web-QA is open source and available under the Apache 2.0 license. You can use it freely in both personal and commercial projects.
Web-QA Alternatives — Similar Browser Automation Servers
Looking for alternatives to Web-QA? Here are other popular browser automation servers you can use with Claude, Cursor, and VS Code.
Chrome DevTools MCP
★ 40.6kAI-powered Chrome automation server with natural language element detection. Control Chrome browser through MCP protocol for testing, debugging, and performance analysis. Features 91% accuracy in element location, works with free AI models, and suppo
UI TARS Desktop
★ 34.9k📇 🏠 - Browser automation capabilities using Puppeteer, both support local and remote browser connection.
Playwright
★ 32.8kA production-ready browser automation server that enables AI assistants to interact with web pages using tools for navigation, element interaction, and data extraction. It features a built-in Inspector UI and robust crash recovery for reliable automa
Page Agent
★ 18.0kJavaScript in-page GUI agent. Control web interfaces with natural language.
Chrome
★ 11.7kAn extension-based MCP server that enables AI assistants to control your browser, leveraging existing sessions and login states for automation and content analysis. It provides over 20 tools for semantic tab search, interactive element manipulation,
LAMDA
★ 7.8kThe most powerful Android RPA agent framework, next generation mobile automation.
Browse More Browser Automation MCP Servers
Explore all browser automation servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Web-QA in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Web-QA?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.