home / mcp / debugg ai mcp server

Debugg AI MCP Server

Zero-Config, Fully AI-Managed End-to-End Testing for all code gen platforms.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "debugg-ai-debugg-ai-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@debugg-ai/debugg-ai-mcp"
      ],
      "env": {
        "DEBUGGAI_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

You can run Debugg AI’s MCP server to test your web application with an AI browser agent. Point the server at a URL or localhost, describe what you want tested, and the agent returns pass/fail results with screenshots. This enables automated, AI-assisted browser testing integrated into your MCP workflow.

How to use

To use this MCP server, configure your MCP client to connect to the Debugg AI MCP endpoint. You will provide an API key and, optionally, a target URL or local port for your app. Once configured, you can describe tests in natural language and watch the AI agent navigate your app, perform interactions, and report results with screenshots.

Practical usage patterns include testing a public URL, a development server running on localhost, or any URL you can reach from the testing environment. You can specify credentials or login details if your app requires authentication, and you can request screenshots at key steps to verify visual outcomes.

How to install

Prerequisites: you need Node.js and npm or a container runtime if you plan to use Docker.

Install the MCP client integration and start using Debugg AI’s MCP server with the following steps.

# Using npm you can install and test the MCP locally if you have a package ready
npm install
npm test
npm run build

Configuration and usage examples

Configure your MCP client to include the server connection details and the API key you obtain from Debugg AI.

Example configuration snippet for the MCP client (stdio configuration using npx):

{
  "mcpServers": {
    "debugg-ai": {
      "command": "npx",
      "args": ["-y", "@debugg-ai/debugg-ai-mcp"],
      "env": {
        "DEBUGGAI_API_KEY": "your_api_key_here"
      }
    }
  }
}

Additional content

Alternative containerized setup is available if you prefer running the MCP server in a Docker container. Provide your API key as an environment variable to enable authentication.

docker run -i --rm --init -e DEBUGGAI_API_KEY=your_api_key quinnosha/debugg-ai-mcp

Notes on usage

The MCP server supports describing the test in natural language via the check_app_in_browser endpoint. The agent will navigate, interact, and provide pass/fail results along with screenshots for each test run.

Available tools

check_app_in_browser

Runs an AI browser agent against your app. The agent navigates, interacts, and reports back with screenshots.