WebEvalAgent
Unleashes LLM-powered agents to autonomously execute and debug web apps directly in your code editor, with features like webapp navigation, network traffic capture, and console error collection.
What is WebEvalAgent?
WebEvalAgent is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to unleashes llm-powered agents to autonomously execute and debug web apps directly in your code editor, with features like webapp navigation, network traffic capture, and console error collection.
Unleashes LLM-powered agents to autonomously execute and debug web apps directly in your code editor, with features like webapp navigation, network traffic capture, and console error collection.
This server falls under the Browser Automation category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Unleashes LLM-powered agents to autonomously execute and deb
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx webevalagent-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use WebEvalAgent
WebEvalAgent MCP Server connects your AI coding assistant to an autonomous web application testing agent that drives a real browser, captures screenshots, collects console errors, and records network traffic — all from within your code editor. Powered by BrowserUse and Playwright, it lets you describe a test scenario in natural language (e.g. "complete the checkout flow and report any errors") and receive a structured report covering UX issues, JavaScript exceptions, and failed network requests. It is especially useful during "vibe coding" sessions where you want to validate that a newly generated web feature actually works end-to-end without switching to a separate test runner.
Prerequisites
- An Operative.sh account and API key — obtain free at operative.sh/mcp
- The operative.sh CLI installed on your machine
- An MCP-compatible client such as Claude Desktop, Claude Code, or Cursor
- A locally running web application to evaluate (or a public URL)
Install the operative.sh CLI
Run the official install script, which sets up the CLI and the WebEvalAgent MCP server in one step.
curl -LSf https://operative.sh/install.sh -o install.sh && bash install.sh && rm install.shObtain your API key
Sign up or log in at operative.sh/mcp to get a free OPERATIVE_API_KEY. This key authenticates the browser automation agent.
Add the server to your MCP client config
Add the WebEvalAgent server to your claude_desktop_config.json with your API key as an environment variable.
{
"mcpServers": {
"web-eval-agent": {
"command": "npx",
"args": ["webevalagent-mcp-server"],
"env": {
"OPERATIVE_API_KEY": "your_operative_api_key"
}
}
}
}Optional: set up persistent browser authentication
Use the setup_browser_state tool to open an interactive browser session where you can log in to your app. Cookies and local storage are saved and reused in subsequent evaluations.
Restart your MCP client and start your web app
Restart the MCP client. Start your local web application on a known port. The server exposes two tools: web_eval_agent (autonomous evaluation) and setup_browser_state (authentication setup).
WebEvalAgent Examples
Client configuration
claude_desktop_config.json for the WebEvalAgent MCP server with the required API key.
{
"mcpServers": {
"web-eval-agent": {
"command": "npx",
"args": ["webevalagent-mcp-server"],
"env": {
"OPERATIVE_API_KEY": "your_operative_api_key"
}
}
}
}Prompts to try
Prompts that invoke the web_eval_agent tool to autonomously test web application flows.
- "Evaluate my app at http://localhost:3000 — run through the full signup flow and report any UX issues or console errors"
- "Test the checkout process on http://localhost:3000/shop — add an item to the cart, proceed to checkout, and report any network failures"
- "Evaluate the login page at http://localhost:3000/login with invalid credentials and check that the error message is displayed correctly"
- "Run a full smoke test of http://localhost:5173 — navigate to every page linked from the homepage and report any 404s or JS errors"Troubleshooting WebEvalAgent
Authentication error: OPERATIVE_API_KEY is invalid or missing
Retrieve your API key from operative.sh/mcp and ensure it is set in the env block of your MCP config. Restart the MCP client after updating the config.
web_eval_agent cannot reach localhost URLs
Ensure your web application is running and listening on the specified port before invoking the tool. Use `curl http://localhost:3000` from the terminal to verify the app is reachable. The browser launched by the agent runs on the same machine as the MCP server.
Evaluation hangs or times out on complex pages
Break complex evaluations into smaller tasks with specific objectives. Instead of 'test everything', use targeted prompts like 'test only the login flow'. The agent works best with clear, scoped task descriptions.
Frequently Asked Questions about WebEvalAgent
What is WebEvalAgent?
WebEvalAgent is a Model Context Protocol (MCP) server that unleashes llm-powered agents to autonomously execute and debug web apps directly in your code editor, with features like webapp navigation, network traffic capture, and console error collection. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install WebEvalAgent?
Follow the installation instructions on the WebEvalAgent GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with WebEvalAgent?
WebEvalAgent works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is WebEvalAgent free to use?
Yes, WebEvalAgent is open source and available under the Apache 2.0 license. You can use it freely in both personal and commercial projects.
WebEvalAgent Alternatives — Similar Browser Automation Servers
Looking for alternatives to WebEvalAgent? 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 WebEvalAgent 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 WebEvalAgent?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.