Model Context Protocol moderate risk

Omniparser Autogui

This MCP server uses OmniParser to analyze the screen and automate GUI interactions on Windows, enabling actions based on visual context.

Connections & Capabilities

Connects To

GitHub

Capabilities

readwriteexec

Quickstart

Config

{
  "mcpServers": {
    "omniparser_autogui_mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "D:\\CLONED_PATH\\omniparser-autogui-mcp",
        "run",
        "omniparser-autogui-mcp"
      ],
      "env": {
        "PYTHONIOENCODING": "utf-8",
        "OCR_LANG": "en"
      }
    }
  }
}

Exposed MCP Tools (1)

moderate

omniparser_autogui

Analyzes the screen using OmniParser and automates GUI interactions based on the analysis.

Automates GUI actions, which can have unintended consequences if the screen analysis is inaccurate.

Safety Assessment

This server's safety depends on the accuracy of OmniParser and the scope of GUI automation. It's relatively safe when targeting specific windows and used for simple tasks. However, it becomes risky when automating complex workflows or operating on the entire screen without careful testing.

Requires explicit configuration in `claude_desktop_config.json`.
Can target specific windows, limiting scope of operations.
Environment variables allow some control over OmniParser behavior.

Automates GUI interactions, potentially leading to unintended actions.
Relies on OmniParser for screen analysis, which may have inaccuracies.
No built-in sandboxing; actions are performed with the user's privileges.
Potential for denial-of-service if OmniParser consumes excessive resources.