# louis030195/terminator-mcp-agent [Health: Active]

**Category:** 🖥️ OS Automation  
**Repository:** https://github.com/mediar-ai/terminator/tree/main/terminator-mcp-agent  
**GitHub Stars:** 1663  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/louis030195-terminator-mcp-agent

## Description
Desktop GUI automation using accessibility APIs. Control Windows, macOS, and Linux applications without vision models or screenshots. Supports workflow recording, structured data extraction, and browser DOM inspection.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What louis030195/terminator-mcp-agent does

louis030195/terminator-mcp-agent connects MCP-based agents to desktop applications through operating-system accessibility APIs. It is intended for GUI automation across Windows, macOS, and Linux, allowing an agent to control supported applications through their accessible interface rather than interpreting screenshots.

The server also covers three related tasks: recording workflows, extracting structured data, and inspecting browser DOM content. Together, these capabilities support both action-oriented automation and the retrieval of structured information from desktop or browser interfaces.

## How it works

Automation is based on accessibility information exposed by the operating system and applications. This approach gives the agent an interface-oriented way to interact with desktop software, rather than requiring a vision model to identify controls in an image. It also avoids a screenshot-driven workflow.

Browser DOM inspection provides a separate structured view of web pages inside a browser context. Workflow recording can capture an interaction sequence for later use, while structured data extraction focuses on returning application or interface information in a machine-readable form.

The cross-platform scope covers Windows, macOS, and Linux applications. The relevant application must expose usable accessibility information for interface automation; visual content alone is not the operating model described for this server.

## Tools and capabilities

- Control desktop applications through accessibility APIs.
- Automate GUI interactions on Windows, macOS, and Linux.
- Record workflows for repeatable interaction sequences.
- Extract structured data from application interfaces.
- Inspect browser DOM structures.
- Run automation without vision models or screenshots.

## Limitations and notes

This server is designed around accessibility APIs, so it is distinct from automation systems that act by analyzing screenshots or using vision models. Its supported operating-system scope is Windows, macOS, and Linux. The strongest fit is an application or browser workflow where controls and content are available through accessibility or DOM structures. Interfaces that cannot expose the required structure may not be suitable for this approach.

_Full upstream README: https://allmcps.com/mcp/louis030195-terminator-mcp-agent/readme_