qa-use
Verified Safeby desplega-ai
Overview
Provides browser automation and QA testing capabilities for AI agents, including automated tests, session monitoring, and batch execution.
Installation
npx @desplega.ai/qa-use-mcpEnvironment Variables
- QA_USE_API_KEY
- QA_USE_API_URL
- QA_USE_APP_URL
- QA_USE_REGION
- TUNNEL_HOST
Security Notes
The server uses a broad CORS policy ('Access-Control-Allow-Origin', '*') for its HTTP mode, though it's protected by Bearer token authentication. Tunnel mode (`localtunnel`) exposes a local browser WebSocket endpoint to the public internet, which inherently increases attack surface. While it uses deterministic subdomains and secure defaults (HTTPS for tunnel host), users should be aware of the implications of exposing local services. Installation of Playwright browsers involves forking a child process (`playwright install chromium`), which is a privileged operation. No direct 'eval' or obvious malicious patterns are observed, and good practices like `--no-sandbox` for Playwright are used. The overall security relies heavily on the security of the `desplega.ai` platform and careful user configuration.
Similar Servers
playwright-mcp
A Model Context Protocol (MCP) server enabling LLMs to automate browser interactions through structured accessibility snapshots, bypassing traditional vision models.
playwright-mcp-server
Provides a minimal yet robust server for AI agents to automate web browser interactions including navigation, DOM manipulation, and network monitoring, with an emphasis on token-aware outputs.
mcp-browser-use-server
An MCP server that enables AI agents to control web browsers using the browser-use library for automation and content extraction.
playwright-mcp
This project provides a step-by-step guide and boilerplate for integrating an MCP (Multi-Modal Code Generation Platform) Playwright server to generate and execute UI and API tests using LLM agents.