Skip to main content

MCP server providing computer control tools for AI agents

Project description

realtimex-computer-use

A MCP (Model Context Protocol) server that provides computer control tools for AI agents, enabling browser automation and system interactions.

Features

  • Open URLs in web browsers
  • Support for multiple browsers (Chrome, Firefox, Safari, Edge)
  • Open URLs in new tabs or windows
  • Retrieve configured credentials for authentication
  • Graceful fallback to system default browser
  • Cross-platform support (Windows, macOS, Linux)

Tools

The server implements the following tools:

Browser Control

  • open_browser - Open a URL in the specified browser or system default
  • open_browser_new_tab - Open a URL in a new browser tab
  • open_browser_new_window - Open a URL in a new browser window

Each tool supports browser selection (chrome, firefox, safari, edge, default) and provides graceful fallback to the system default browser if the specified browser is unavailable.

Credential Management

  • get_credentials - Get available credentials for authentication

Returns credential names and types for the agent to ask which credential to use for login. Configured via CREDENTIAL_SERVER_URL environment variable.

Installation

Prerequisites

  • Python 3.10+
  • FastMCP framework
  • uv package manager

Install Steps

Install the package:

pip install realtimex-computer-use

Or using uvx for immediate use:

uvx realtimex-computer-use

MCP Client Configuration

To use this server with an MCP-compatible client, configure it to run the server via stdio transport.

Development Configuration (local installation):

{
  "mcpServers": {
    "realtimex-computer-use": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/realtimex-computer-use",
        "run",
        "realtimex-computer-use"
      ]
    }
  }
}

Production Configuration (published package):

{
  "mcpServers": {
    "realtimex-computer-use": {
      "command": "uvx",
      "args": [
        "realtimex-computer-use"
      ]
    }
  }
}

Development

Building and Publishing

  1. Sync dependencies and update lockfile:
uv sync
  1. Build package distributions:
uv build
  1. Publish to PyPI:
uv publish

Note: Set PyPI credentials via environment variables or command flags:

  • Token: --token or UV_PUBLISH_TOKEN
  • Username/password: --username/UV_PUBLISH_USERNAME and --password/UV_PUBLISH_PASSWORD

Debugging

For the best debugging experience, use the MCP Inspector.

Launch the MCP Inspector via npm:

npx @modelcontextprotocol/inspector uv --directory /path/to/realtimex-computer-use run realtimex-computer-use

The Inspector will display a URL that you can access in your browser to begin debugging.

Usage Examples

Open a URL in the default browser

{
  "tool": "open_browser",
  "arguments": {
    "url": "https://www.python.org"
  }
}

Open a URL in Chrome

{
  "tool": "open_browser",
  "arguments": {
    "url": "https://www.python.org",
    "browser": "chrome"
  }
}

Open a URL in a new tab

{
  "tool": "open_browser_new_tab",
  "arguments": {
    "url": "https://docs.python.org",
    "browser": "firefox"
  }
}

Get available credentials

{
  "tool": "get_credentials",
  "arguments": {}
}

Configuration: Set CREDENTIAL_SERVER_URL environment variable (defaults to http://localhost:3001)

Future Expansion

This package is designed to support additional computer control capabilities:

  • Desktop automation (PyAutoGUI integration)
  • File system operations
  • System information retrieval
  • Process management
  • Additional credential operations (select, validate)

Architecture

The codebase is organized for maintainability and extensibility:

realtimex-computer-use/
├── fastmcp.json          # FastMCP configuration (dependencies)
├── pyproject.toml        # Package configuration and metadata
├── smithery.yaml         # Smithery MCP registry configuration
└── src/
    └── realtimex_computer_use/
        ├── __init__.py   # Package entry point
        ├── __main__.py   # CLI entry point
        ├── server.py        # MCP server initialization and tool registration
        └── tools/           # Modular tool implementations
            ├── __init__.py
            ├── browser.py     # Browser control tools
            └── credentials.py # Credential management tools

Configuration Files:

  • fastmcp.json: Defines FastMCP dependencies and entrypoint (follows FastMCP 2.11.4+ standard)
  • pyproject.toml: Python package metadata, dependencies, and build configuration
  • smithery.yaml: Configuration for Smithery MCP server registry

Adding New Tools:

  1. Create a new module in src/realtimex_computer_use/tools/ (e.g., credentials.py)
  2. Implement tool functions with proper type hints and docstrings
  3. Register tools in server.py using mcp.tool()(module.function_name)

License

This project is proprietary software. All rights reserved.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

realtimex_computer_use-1.0.0.tar.gz (84.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

realtimex_computer_use-1.0.0-py3-none-any.whl (6.3 kB view details)

Uploaded Python 3

File details

Details for the file realtimex_computer_use-1.0.0.tar.gz.

File metadata

File hashes

Hashes for realtimex_computer_use-1.0.0.tar.gz
Algorithm Hash digest
SHA256 67f39d751098aa86caf50b63a589bd25eb05dda540d00d4814a3b9a53a8eac4f
MD5 7743e37af9a5a3cef347b868d8a6daa2
BLAKE2b-256 5bfe6ec5b5f7c7f83fc1ad68d83e4029708e98feabbcaebbceb166822c9420aa

See more details on using hashes here.

File details

Details for the file realtimex_computer_use-1.0.0-py3-none-any.whl.

File metadata

File hashes

Hashes for realtimex_computer_use-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 516d270ef6129665d9c73d58a27ec6ef1bc1158e07edbde9aeb8dba15c2647f2
MD5 6a0d3721d96267797c5c94a93911e33e
BLAKE2b-256 6c0ea9bff4a4795d4b16a371eeea05584a06bd2b54e99b98d6ad7269e4a05585

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page