agentmods.dev

win-computer-use

A guide for controlling a live Windows computer through an automation service that can operate apps, files, the mouse, keyboard, screen, and related system functions.

In plain words
What is it for?
Opening and managing applications, clicking and typing, taking screenshots, reading or changing files, using OCR, running processes, recording the screen, and playing macros.
Why use it?
It gives an AI assistant a defined way to perform and verify tasks on a Windows desktop instead of relying only on text instructions.

Skill for Claude CodeCodex

1 16d ago A scan 123 tokens original MIT
View the source file JNX03/WindownComputerUse
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/jnx03/windowncomputeruse/win-computer-use
Any agent
npx skills add JNX03/WindownComputerUse --skill win-computer-use
Clone the repo
git clone --depth 1 https://github.com/JNX03/WindownComputerUse

Made for: Claude Code, Codex.

Per session 123 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,582 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00123 $0.02582
Opus 5 $0.00062 $0.01291
Sonnet 5 $0.00025 $0.00516
Haiku 4.5 $0.00012 $0.00258

Measured 2d ago against content hash 952c5e77bf9b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

win-computer-use scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/startup_check.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/win-computer-use/SKILL.md · 177 lines

How it starts

The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.

win-computer-use — operational guide for AI agents

You have access to the win-computer-use MCP server. It exposes 64 tools that let you drive a real Windows desktop: mouse, keyboard, screen capture, OCR, window/app management, files, HTTP, processes, audio, screen recording, macros, and a labeled on-screen virtual cursor. Tool names in this guide use the mcp__win-computer-use__ prefix that Claude clients see; other clients may strip it - apply the same prefix convention your client uses.

This file is the entry point. Detailed references are split into smaller files in references/ and loaded only when relevant.

TL;DR

  1. You have a virtual cursor rendered as an always-on-top labeled overlay. Clicks default to Win32 PostMessage so the user's real cursor is not touched. Pass use_real_cursor: true when PostMessage won't work (Paint canvas, Chromium page bodies, games).
  2. Tools return ok: true for "I dispatched the call", not "the effect happened". Always screenshot and verify for UI-changing actions whose effect you cannot predict.
  3. The emergency hotkey may not be active in this install. Check emergency_stop_status().started on session start and warn the user if it is false.
  4. OCR may be broken on the host (missing winrt-Windows.Globalization transitive dep). Have a hotkey/coord fallback ready.
  5. Prefer direct tools over UI when you can: use write_text_file instead of typing into Notepad, http_get instead of driving a browser, clipboard_set instead of typing long text.
  6. Don't toggle permission_set_bypass silently and don't close dirty docs without saving - close_window discards changes without prompting.

First-turn checklist

Run these in parallel at session start, then act on the results:

get_state()                  -> overlay alive? heartbeat fresh?
permission_status()          -> bypass on/off, what apps are allowed
emergency_stop_status()      -> if .started=false, WARN the user
emergency_resume()           -> clear stale stop flag from a prior session
list_monitors()              -> coord system and virtual desktop bounds
list_windows()               -> what is open; what to avoid touching
screenshot(monitor_index=1)  -> ground yourself visually

Read the full file on GitHub · 177 lines

Files

What ships with it

9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 177 lines · 123 tokens per session scan A 952c5e77bf9b

Subscribe to this mod's changes

win-computer-use is a skill published in the GitHub repository JNX03/WindownComputerUse (1 stars, last pushed 16d ago), licensed MIT. It adds 123 tokens to every session and 2,582 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Browser automationagentai-agentsanthropicautomationclaudeclaude-code
Related

Other skills, from other repositories

dingtalk_channel_connect

Use a headed browser to automatically complete DingTalk channel integration for QwenPaw. Applicable when the user mentions DingTalk, developer console, Client ID, Client Secret, bot, Stream mode, binding or configuring a channel. Supports pausing when a login page is detected and resuming after the user logs in.

agentscope-ai/QwenPaw · 69 tokens

dingtalk_channel_connect

使用可视浏览器自动完成 QwenPaw 的钉钉频道接入。适用于用户提到钉钉、DingTalk、开发者后台、Client ID、Client Secret、机器人、Stream 模式、绑定或配置 channel 的场景；支持遇到登录页时暂停，等待用户登录后继续。.

agentscope-ai/QwenPaw · 78 tokens

browser_cdp

Use this skill when the user explicitly wants to connect to a running Chrome browser, scan local CDP ports, specify a cdpport, or share a single browser across multiple agents/tools. By default browser opens no debugging port; pass an explicit cdpport only when the user wants another local tool to attach.

agentscope-ai/QwenPaw · 70 tokens

browser_cdp

当用户明确希望连接到已运行的 Chrome 浏览器、扫描本地 CDP 端口、显式指定 cdpport，或让多个 agent / 工具共享同一个浏览器时，使用本 skill。browser 默认不开放调试端口；仅当用户明确希望其他本地工具附加时才显式传入 cdpport。.

agentscope-ai/QwenPaw · 85 tokens

browser_visible

当用户需要控制 browser 的浏览器启动方式时，使用本 skill。browser 默认由 Playwright 直接管理、不开放调试端口（需让其他本地工具附加时显式传 cdpport）；headed 控制是否显示窗口，privatemode 保留用于兼容、不再改变默认行为，browserargs 传入额外的 Chromium 启动参数，executablepath 指定自定义浏览器可执行文件路径。.

agentscope-ai/QwenPaw · 108 tokens

browser_visible

Use this skill when the user needs to control the browser launch mode for browser. By default browser is managed by Playwright and opens no debugging port (pass an explicit cdpport to let another local tool attach); headed controls whether the window is visible, and privatemode is kept for backward compatibility and…

agentscope-ai/QwenPaw · 75 tokens