home / mcp / autopilot browser mcp server

Autopilot Browser MCP Server

Autopilot Browser Workflows MCP Server

Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "autopilotbrowser-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@autopilotbrowser/mcp-server"
      ],
      "env": {
        "AUTOPILOT_API_KEY": "your_api_key_here"
      }
    }
  }
}

You can run Autopilot Browser workflows through an MCP server that exposes the Autopilot capabilities to Claude Desktop and other AI agents. This server lets you search, list, and execute browser workflows from your automation client with a simple, secure API surface.

How to use

Use this MCP server with your MCP client to search for available workflows, list them, and execute the workflows you need. You specify inputs, and the server orchestrates the Autopilot Browser workflows, delivering real-time feedback as tasks progress.

How to install

Prerequisites: Ensure you have Node.js installed on your machine.

Install the MCP server as shown in the code example by configuring your client to run the server process via npx and supplying your API key.

{
  "mcpServers": {
    "autopilot": {
      "command": "npx",
      "args": ["-y", "@autopilotbrowser/mcp-server"],
      "env": {
        "AUTOPILOT_API_KEY": "your_api_key_here"
      }
    }
  }
}

Additional notes

This MCP server provides a focused workflow experience for Autopilot Browser through the MCP interface. The key environment variable to supply is AUTOPILOT_API_KEY, which authorizes your instance to access Autopilot Browser features.

Available tools

search_workflows

Find workflows by keyword to quickly locate relevant automation tasks within Autopilot Browser.

execute_workflow

Run a selected browser workflow with the required inputs and capture results in real time.

real_time_results

Receive instant feedback and live updates as a workflow executes, including status and outputs.