home / mcp / vibetest use mcp server
Launches multiple Browser-Use agents to test UI quality, links, and accessibility across live and localhost sites.
Configuration
View docs{
"mcpServers": {
"browser-use-vibetest-use": {
"command": "/full/path/to/vibetest-use/.venv/bin/vibetest-mcp",
"args": [],
"env": {
"GOOGLE_API_KEY": "YOUR_API_KEY"
}
}
}
}Vibetest Use is an MCP server that orchestrates multiple Browser-Use agents to test a vibe-coded website for UI bugs, broken links, accessibility issues, and other technical problems. It’s ideal for validating both live sites and local development environments, helping you ensure your site works across real-world usage scenarios.
You will interact with Vibetest Use through an MCP client to launch a vibetest server instance that runs multiple browser-based testing agents. Start by ensuring your testing environment has a Python 3.11+ runtime and a valid Google API key. Use your MCP client to connect to Vibetest Use, then configure the test run by specifying the target URL, the number of agents, and whether tests run in headless mode. The server will automatically manage a pool of Browser-Use agents to perform UI checks, link validation, and accessibility assessments on the specified site.
Prerequisites you need before installing Vibetest Use: a Python 3.11+ runtime and a Google API key. You also need support in your MCP client for MCP with Python-based servers.
Steps to install and set up Vibetest Use locally:
uv venv
source .venv/bin/activate
uv pip install -e .
# Install the browser driver dependencies
playwright install chromium --with-deps --no-shellTo run Vibetest Use, you connect via an MCP client and configure the vibetest MCP server entry with the appropriate command and environment variables. The recommended environment variable is GOOGLE_API_KEY which you use to authenticate API calls from the server.
# Example local server start via Cursor manual setup
{
"mcpServers": {
"vibetest": {
"command": "/full/path/to/vibetest-use/.venv/bin/vibetest-mcp",
"env": {
"GOOGLE_API_KEY": "your_api_key"
}
}
}
}Launches and manages a pool of Browser-Use agents to perform automated UI testing across the target site.
Runs UI tests to detect layout issues, rendering bugs, and visual regressions.
Checks for broken or redirected links within the tested pages.
Performs accessibility checks to identify common WCAG issues.