Skip to main content
Glama
Sign In
Autopilot Browser MCP Server

Autopilot Browser MCP Server

Official
by autopilotbrowser
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Remote

Autopilot Browser MCP Server

MCP server for executing Autopilot Browser workflows via Claude Desktop and other AI agents.

Installation

Add this to your Claude Desktop configuration file:

MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%/Claude/claude_desktop_config.json

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

Related MCP server: Playwright Vision MCP Server

Usage

Once installed, Claude can:

  1. Search for available workflows by keyword

  2. List all available workflows

  3. Execute workflows with the appropriate inputs

Example Commands

  • "Show me all available Autopilot workflows"

  • "Search for web scraping workflows"

  • "Run the data extraction workflow with URL https://example.com"

Getting an API Key

Visit Autopilot Browser to sign up and generate your API key.

Features

  • Workflow Search: Find workflows by keyword

  • Workflow Execution: Run automated browser workflows

  • Real-time Results: Get instant feedback from workflow execution

Support

For issues or questions:

Available Tools

2 tools
run_workflowA

Execute a specific Autopilot Browser workflow with the provided inputs. Use search_workflows first to find available workflows and their required inputs.

Inputs: workflowName (required), workflowInputs (required).

search_workflowsB

Search for available Autopilot Browser workflows by keyword. Returns a list of workflows that match the search term. If no search term is provided, returns all workflows.

Inputs: searchTerm.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    B3.4/5.0

    Scored across 2 tools

    Disambiguation5/5

    The two tools have clearly distinct purposes: search_workflows is for discovery and information retrieval, while run_workflow is for execution. There is no overlap in functionality, and the descriptions explicitly differentiate them with one being preparatory and the other operational.

    Naming Consistency5/5

    Both tools follow a consistent verb_noun naming pattern (search_workflows, run_workflow) with snake_case throughout. The naming is predictable and aligns well with their functions, making them easily distinguishable and readable.

    Tool Count2/5

    With only 2 tools, the server feels thin for a browser automation domain that typically involves more operations like navigation, clicking, or data extraction. While the tools cover basic workflow execution, the limited count suggests significant gaps in functionality for comprehensive browser control.

    Completeness2/5

    The tool set is severely incomplete for a browser automation server. It lacks core operations such as page navigation, element interaction, data scraping, or session management. The existing tools only handle workflow search and execution, leaving most browser automation tasks unsupported and likely causing agent failures.

    Resources

    Related MCP Connectors

    Related MCP Servers

    Appeared in Searches