Vibetest Use

Vibetest Use is an MCP server that runs automated QA testing on websites to detect UI bugs, broken links, accessibility issues, and technical problems. It connects MCP-compatible development tools, such as Claude Code and Cursor, directly to multiple autonomous Browser-Use agents powered by Playwright and Google's Gemini models. Web developers, software engineers, and QA testers use this server to audit both production domains and local development environments without manually scripting repetitive end-to-end tests. By executing concurrent browser sessions, Vibetest Use explores sites autonomously to surface front-end regressions, layout discrepancies, and navigational failures. Developers can control the depth of each test session by adjusting the agent count and toggling headless or visible browser modes. This enables continuous inspection of web projects during code generation or regular debugging, giving teams quick feedback on whether new interface updates work as intended across real browser interactions.

Category: Other & General Purpose

Tags: demo, template, testing

Visit Vibetest Use

How to install and configure Vibetest Use

1. Create and activate a Python virtual environment, then install dependencies: bash uv venv source .venv/bin/activate uv pip install -e . 2. Install the necessary Chromium browser binaries: bash playwright install chromium --with-deps --no-shell 3. Register with Claude Code using the CLI: bash claude mcp add vibetest /full/path/to/vibetest-use/.venv/bin/vibetest-mcp -e GOOGLE_API_KEY="your_api_key" 4. Alternatively, configure Cursor manually in your MCP settings config: json { "mcpServers": { "vibetest": { "command": "/full/path/to/vibetest-use/.venv/bin/vibetest-mcp", "env": { "GOOGLE_API_KEY": "your_api_key" } } } }

What you can do with Vibetest Use

• Launching multiple Browser-Use agents concurrently to discover UI bugs and rendering defects across responsive layouts. - Crawling complex web applications to detect dead buttons, non-functioning forms, and broken navigational links. - Testing web applications hosted on localhost development servers before deploying updates to production staging environments. - Inspecting web pages to identify accessibility issues and markup errors using multi-agent automated browsing sessions. - Performing automated headless smoke tests on staging URLs to verify technical stability after recent code changes.

Key facts

• https://github.com/browser-use/vibetest-use
• Other & General Purpose
• demo, template, testing

Part of MCP Servers

How do I install Vibetest Use?

Set up a virtual environment using uv, install the local package with uv pip install -e ., and install Chromium via Playwright using playwright install chromium --with-deps --no-shell. Finally, link the executable vibetest-mcp to your MCP client like Cursor or Claude Code.

What can Vibetest Use do?

Vibetest Use runs automated quality assurance tests using autonomous Browser-Use agents. It scans local development servers or live websites for broken links, UI defects, accessibility barriers, and general technical problems across concurrent browser sessions.

Which MCP clients work with Vibetest Use?

Vibetest Use works with AI coding assistants and MCP clients that support the Model Context Protocol, including Claude Code and Cursor, provided you configure the server command and your Google API key.

What requirements are needed to run Vibetest Use?

You need Python 3.11 or higher, an MCP-compatible client like Claude Code or Cursor, and a valid Google API key supporting gemini-2.0-flash to power the Browser-Use agents during automated testing.

Is Vibetest Use open source?

Yes, Vibetest Use is open source software released under the MIT license on GitHub by the browser-use organization.