BrowserStack MCP Server Integration

What does “BrowserStack” MCP Server do?

BrowserStack MCP Server is a Model Context Protocol (MCP) server that connects AI assistants and developer tools directly to BrowserStack’s cloud-based testing infrastructure. It empowers developers and testers to automate, manage, and interact with BrowserStack’s real device and browser cloud from within their favorite AI-driven workflows. The server enables tasks such as launching manual or automated tests on real devices, debugging app crashes, testing websites across browsers, and integrating advanced app profiling—all via standardized MCP tools and prompts. By bridging the gap between AI assistants and BrowserStack’s robust test platform, it streamlines testing processes, enhances debugging, and accelerates development workflows for both manual testers and automation engineers.

List of Prompts

• “open my app on a iPhone 15 Pro Max”
  Launches your mobile application on a specific real device in BrowserStack’s cloud.
• “My app crashed on Android 14 device, can you help me debug?”
  Initiates debugging and diagnostics for app crashes on a specific device and OS version.
• “open my website hosted on localhost:3001 on Edge”
  Opens a local website on a real browser/device in the BrowserStack infrastructure.

List of Resources

• Device Grid Resource
  Provides access to the complete list of available real devices and browser/OS combinations in BrowserStack’s cloud.
• App Profiling Data
  Supplies app performance and profiling metrics from real-world testing sessions.
• Test Session Artifacts
  Exposes logs, screenshots, and videos generated during testing sessions for further analysis.
• Device & Browser Documentation
  Offers documentation links for available devices, OS versions, and supported browsers.

List of Tools

No direct list of tools (such as from server.py) is provided in the available documentation or code tree.

Use Cases of this MCP Server

• Manual App Testing
  Instantly open and interact with mobile apps on a wide range of real devices, allowing testers to replicate and debug real-world scenarios without emulators.
• Debug App Crashes
  Utilize profiling and diagnostic tools to analyze and resolve app crashes on specific device/OS combinations in the BrowserStack cloud.
• Manual Web Testing
  Test websites on real browsers and devices, including those hosted locally, ensuring cross-browser compatibility without local browser installations.
• Automated Test Integration
  Integrate BrowserStack’s device and browser grid into automated QA flows, enabling continuous testing as part of CI/CD pipelines.
• Comprehensive Device Coverage
  Leverage the full spectrum of up-to-date devices and browsers, including newly released hardware on launch day, to broaden and modernize test coverage.

How to set it up

Windsurf

1. Ensure Node.js is installed on your system.
2. Locate the Windsurf configuration file.
3. Add the BrowserStack MCP Server to the mcpServers object using the following JSON snippet:

   {
     "mcpServers": {
       "browserstack-mcp": {
         "command": "npx",
         "args": ["@browserstack/mcp-server@latest"]
       }
     }
   }
   

4. Save the configuration and restart Windsurf.
5. Verify the setup by checking if the server is running and accessible.

Claude

1. Install Node.js if not already present.
2. Open Claude’s integration/configuration settings.
3. Add BrowserStack MCP Server as follows:

   {
     "mcpServers": {
       "browserstack-mcp": {
         "command": "npx",
         "args": ["@browserstack/mcp-server@latest"]
       }
     }
   }
   

4. Save changes and restart Claude.
5. Confirm that the MCP server is reachable from Claude’s tools panel.

Cursor

1. Confirm Node.js is installed.
2. Edit Cursor’s mcpServers configuration.
3. Insert:

   {
     "mcpServers": {
       "browserstack-mcp": {
         "command": "npx",
         "args": ["@browserstack/mcp-server@latest"]
       }
     }
   }
   

4. Restart Cursor to apply the new server configuration.
5. Test the integration by issuing a prompt and confirming BrowserStack options appear.

Cline

1. Prerequisite: Node.js should be installed.
2. Navigate to Cline’s configuration files.
3. Add the MCP server:

   {
     "mcpServers": {
       "browserstack-mcp": {
         "command": "npx",
         "args": ["@browserstack/mcp-server@latest"]
       }
     }
   }
   

4. Save and restart Cline.
5. Check Cline’s UI for BrowserStack testing capabilities.

Securing API Keys

To keep your BrowserStack API keys secure, use environment variables:

{
  "mcpServers": {
    "browserstack-mcp": {
      "command": "npx",
      "args": ["@browserstack/mcp-server@latest"],
      "env": {
        "BROWSERSTACK_API_KEY": "${BROWSERSTACK_API_KEY}"
      },
      "inputs": {
        "apiKey": "${BROWSERSTACK_API_KEY}"
      }
    }
  }
}

Replace "${BROWSERSTACK_API_KEY}" with your actual environment variable or secret manager reference.

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "browserstack-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “browserstack-mcp” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Our opinion

BrowserStack MCP Server offers robust integration for real-device and browser testing through AI and MCP, with clear setup guidance and valuable prompts/resources. However, the lack of explicit tool listings and information on advanced MCP features (roots, sampling) slightly limits transparency for power users.

MCP Score

Score: 7/10
Well-documented, with valuable resources and prompt examples. Would benefit from more explicit tool documentation and details on advanced MCP features.