Automates Windows desktop and browser tasks through accessibility APIs, UI controls, workflow recording, and structured inspection.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent ā or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag ā we're steadily working through the catalog.
š” Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Terminator MCP Agent.
The louis030195/terminator-mcp-agent MCP server connects an MCP-compatible AI assistant to Windows desktop automation. It can locate interface elements using metadata such as name, role, window, control type, or automation ID, then perform actions including clicking and typing. It also supports launching and managing applications, listing windows, identifying the active window, and handling multiple displays.
Browser automation is available through a Chrome extension. The project describes support across pixels, browser DOM data, and the Windows Accessibility tree, allowing an agent to work with both desktop applications and browser interfaces. Screenshots can be captured for displays or individual elements, but the automation model does not require a vision model or manual cursor control.
The MCP process runs locally and exposes Terminator capabilities to an AI client. Selectors are based on accessibility metadata rather than coordinates alone. To build reliable selectors, inspect the target application's Windows Accessibility Tree with Accessibility Insights for Windows or Inspect.exe. These tools reveal fields such as the element name, role, control type, and automation ID, which can then be used in selectors like name:Seven, role:Button, or window:Calculator.
Workflow recording can capture human interactions and produce deterministic automation. The project states that AI is used for recovery when needed, while recorded workflows provide the primary execution path. Browser sessions can be reused, including existing cookies and authentication state, according to the project documentation.
The louis030195/terminator-mcp-agent MCP server can be launched with npm through npx:
For MCP clients that use JSON configuration, add a server entry with npx as the command and -y, terminator-mcp-agent@latest as its arguments. The README also shows optional LOG_LEVEL and RUST_BACKTRACE environment settings with values such as info and 1. A separate Claude Code command is provided using claude mcp add.
The louis030195/terminator-mcp-agent MCP server currently supports Windows only. The README marks macOS and Linux automation, CLI support, and MCP-agent support as unavailable on those platforms. Python support is listed as partial, while the TypeScript, workflow, CLI, KV, MCP, and Rust libraries are listed as Windows-supported.
Browser automation requires the Chrome extension described in the feature table. Reliable selector creation may require Accessibility Insights for Windows or Inspect.exe. The material does not document remote hosting, OAuth, or API-key configuration for the MCP process. The project is MIT-licensed, and its stated MCP installation uses the npm package terminator-mcp-agent.
Factual signals from GitHub, npm, and our automated checks ā not a rating.
No reviews yet ā be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/louis030195-terminator-mcp-agent)<a href="https://allmcps.com/mcp/louis030195-terminator-mcp-agent"><img src="https://allmcps.com/api/badge/louis030195-terminator-mcp-agent?style=directory" alt="Terminator MCP Agent on AllMCPs" /></a>