AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompare
Home/
Workflow Automation/
desktop-touch-mcp
desktop-touch-mcp logo

desktop-touch-mcp

Active·★ 22·MIT·Updated 2026-10-22

desktop-touch-mcp is a high-performance Windows MCP server that enables AI clients like Claude and Cursor to safely and semantically interact with the desktop. It uses a Rust native engine for speed, semantic discover-then-act targeting, and per-action perception guards to prevent errors.

#Windows Automation#AI Agent#Desktop Control#UI Automation#Chromium CDP#Rust#Semantic Targeting
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.

Guides
#Credential Management
$ Install
$ npx -y @harusame64/desktop-touch-mcp
↗ Visit site★ GitHub
01

Features

01High-performance Rust Native Core for fast UI Automation (UIA) and image processing with SSE2 SIMD.
02Set-of-Marks (SoM) visual fallback using Windows OCR for interacting with UIA-blind applications like games or RDP sessions.
03Secure Key Locker for automatic terminal credential autofill (SSH, sudo passwords) with Windows DPAPI encryption.
04LLM-native design with reactive perception and token reduction for efficient AI interaction, including macro batching and diff mode.
05Emergency stop (Failsafe) mechanism triggered by parking the mouse in the top-left corner to halt runaway automation.
02

Compatibility

Windows
OS
Verified via docs
Node.js
Runtime
Verified via docs
PowerShell
Fallback
Verified via docs
Claude CLI
Client
Verified via docs
Chromium Browsers
Browser
Verified via docs
03

Quick start

1
$ npx -y @harusame64/desktop-touch-mcp
04

Use cases

↳Enabling AI agents to control and automate various Windows desktop applications, including Notepad, Excel, and Chrome.
↳Securely automating terminal commands requiring SSH or sudo passwords using the integrated Key Locker.
↳Performing advanced web browser automation on Chrome/Edge/Brave using Chromium Debugging Protocol without traditional tools.
↳Efficiently observing and interacting with desktop UI elements using semantic targeting (discover-then-act) for reliable operations.
05

Alternatives

⚡Top VS:
VSiphone-use★333→VSwindows-computer-use-mcp★48→VStmux-mcp★57→

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Compatibility03Quick start04Use cases05Alternatives
VS
Aetox
★55
→
iphone-use logo
iphone-use★ 333
响应快 省Token, Let AI agents (Claude Code, Codex, any MCP client) drive a real iPhone · 让 AI agent 操作真实 iPhone:读屏、点击、输入、滑动,USB 或 Wi-Fi,自有 XCTest runner,可回放的 App flow。Open source, self-hosted on macOS.
VSCompare→
windows-computer-use-mcp logo
windows-computer-use-mcp★ 48
Windows Computer Use for AI Agents. Both a tool (22 MCP tools for click, type, screenshot, OCR, UI inspection) and an agent (autonomous mission engine, macro recorder, intent-based discovery, event watchers). Built with opencode (DeepSeek V4). Ships as MCP server, web UI, and Tauri desktop app.
VSCompare→
tmux-mcp logo
tmux-mcp★ 57
Lets agents create sessions, split panes, run commands, and capture output with TMUX.
VSCompare→
Aetox logo
Aetox★ 55
Desktop AI agent for Windows — an architecture that expands what any model can do. Image/video OCR, offline speech transcription, an agent-driven browser, real .xlsx/.pptx/.docx output, MCP, and a delegated team of agents. Runs on your machine, with any provider.
VSCompare→
Aetox logo
Aetox★ 55
Desktop AI agent for Windows — an architecture that expands what any model can do. Image/video OCR, offline speech transcription, an agent-driven browser, real .xlsx/.pptx/.docx output, MCP, and a delegated team of agents. Runs on your machine, with any provider.
VSCompare→
scrcpy-mcp logo
scrcpy-mcp★ 124
MCP server for Android device control via ADB and scrcpy — gives AI agents vision and control over Android devices
VSCompare→
godot-agent logo
godot-agent★ 53
Godot AI agent CLI, Skill, and MCP server with structured JSON/schema output — headless scene/script/export automation plus live runtime control.
VSCompare→
dcc-mcp-core logo
dcc-mcp-core★ 70
Core runtime and shared tooling for DCC-MCP: safe, skill-first MCP integration across Maya, Blender, Houdini, Photoshop, and custom DCC hosts.
VSCompare→
See all alternatives →
Harusame64 /
desktop-touch-mcp
↗

Windows computer-use MCP server: drive any desktop app via semantic discover-then-act targeting (entities + leases, not pixel coordinates), with per-action perception guards, a native Rust UIA engine, Chrome CDP, and Key Locker credential autofill for ssh/sudo password prompts. Works with Claude, Cursor, and any MCP client.

22MIT
Last commit1d ago
StatusActive
CategoryWorkflow Automation
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit