SKILLEMALL.ai
GitHub ↗ RU Sign in

CC windows-agent

全能 Windows 桌面控制：看屏幕/截图/控窗口(移动缩放激活关闭启动)/模拟鼠标键盘/点按钮填框/UI自动化/管理进程。当用户说"看屏幕/看窗口/截图/帮我点/输入文字/控制程序/打开应用/操作窗口/最小化/最大化/移动窗口"时使用。

ClawHub Agent Skills author: Sunset v0.0.1 MIT-0 14 files · 10 scripts body ≈ 2 648 tokens Open the sourceclawhub.ai↗ analyzed 2 d ago

全能 Windows 桌面控制：看屏幕/截图/控窗口(移动缩放激活关闭启动)/模拟鼠标键盘/点按钮填框/UI自动化/管理进程。当用户说"看屏幕/看窗口/截图/帮我点/输入文字/控制程序/打开应用/操作窗口/最小化/最大化/移动窗口"时使用。

As a process C 51/100 · Has gaps — weak spots: result and completion, when it triggers, failures and branches

ProcedureAI and agentsInfrastructuretype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
C
65/100
safety, quality, tests
Safety 60%
58
Quality 40%
75
Run on models
none yet
Process rating
C
51/100
Has gaps
Result and completion w 14
0
Failures and branches w 10
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Dangerous commands medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.

For the author

Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.

How to improve

  1. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 14

✓ No critical or high findings

Medium and low: 14
  • medium Dangerous commands cmd-execpolicy-bypass scripts/uiauto.ps1:42
    Runs PowerShell with execution policy bypassed
    if($Double){ & $script:pwshExe -NoProfile -ExecutionPolicy Bypass -File $scri…Ps1 -Action mouse-click -X $x -Y $y -Button left -DoubleClick 2>&1 | Out-Null }
  • medium Dangerous commands cmd-execpolicy-bypass scripts/uiauto.ps1:43
    Runs PowerShell with execution policy bypassed
    else { & $script:pwshExe -NoProfile -ExecutionPolicy Bypass -File $scri…Ps1 -Action mouse-click -X $x -Y $y -Button left 2>&1 | Out-Null }
  • medium Dangerous commands cmd-execpolicy-bypass scripts/uiauto.ps1:46
    Runs PowerShell with execution policy bypassed
    if($ClickFirst){ & $script:pwshExe -NoProfile -ExecutionPolicy Bypass -File $scri…Ps1 -Action mouse-click -X $cx -Y $cy -Button left 2>&1 | Out-Null; Start-Sleep -Milliseconds 80 }
  • medium Dangerous commands cmd-execpolicy-bypass scripts/uiauto.ps1:47
    Runs PowerShell with execution policy bypassed
    & $script:pwshExe -NoProfile -ExecutionPolicy Bypass -File $scri…Ps1 -Action type-text -Text $txt 2>&1 | Out-Null
  • medium Dangerous commands cmd-execpolicy-bypass scripts/vision.ps1:33
    Runs PowerShell with execution policy bypassed
    & "pwsh" -NoProfile -ExecutionPolicy Bypass -File "$PSScriptRoot\screen.ps1" -Action capture-region -X $RegionX -Y $RegionY -Width $RegionWidth -Height $RegionHeight -OutputPath $shot | Out-Null
  • medium Dangerous commands cmd-execpolicy-bypass scripts/vision.ps1:35
    Runs PowerShell with execution policy bypassed
    & "pwsh" -NoProfile -ExecutionPolicy Bypass -File "$PSScriptRoot\screen.ps1" -Action capture -OutputPath $shot | Out-Null
  • medium Dangerous commands cmd-execpolicy-bypass SKILL.md:24
    Runs PowerShell with execution policy bypassed
    pwsh -NoProfile -ExecutionPolicy Bypass -File "{baseDir}/scripts/<脚本>.ps1" -Action <动作> [参数]
  • low Dangerous commands cmd-execpolicy-bypass PREREQUISITES.md:22
    Runs PowerShell with execution policy bypassed (documentation table row)
    | **PowerShell 7 (pwsh)** | ✅ 必装 | 所有脚本一律用 `pwsh -NoProfile -ExecutionPolicy Bypass -File` 执行（**勿用 powershell.exe 5.1**，会 GBK 中文乱码）|
    table
  • low Dangerous commands cmd-execpolicy-bypass scripts/vision.ps1:44
    Runs PowerShell with execution policy bypassed (security demo / example)
    & "pwsh" -NoProfile -ExecutionPolicy Bypass -File "$PSScriptRoot\screen.ps1" -Action capture-window -Target $Target -OutputPath $shot | Out-Null
    demo
  • low Dangerous commands cmd-execpolicy-bypass scripts/window.ps1:305
    Runs PowerShell with execution policy bypassed (string literal in code, not executed)
    & $pwsh -NoProfile -ExecutionPolicy Bypass -File $input -Action mouse-move -X $cx -Y $cy 2>&1 | Out-Null
    code literal
  • low Dangerous commands cmd-execpolicy-bypass scripts/window.ps1:307
    Runs PowerShell with execution policy bypassed (string literal in code, not executed)
    & $pwsh -NoProfile -ExecutionPolicy Bypass -File $input -Action mouse-click -X $cx -Y $cy -Button left -DoubleClick 2>&1 | Out-Null
    code literal
  • low Dangerous commands cmd-execpolicy-bypass scripts/window.ps1:333
    Runs PowerShell with execution policy bypassed (string literal in code, not executed)
    & $pwsh -NoProfile -ExecutionPolicy Bypass -File $input -Action mouse-move -X $tx -Y $ty 2>&1 | Out-Null
    code literal
  • low Dangerous commands cmd-execpolicy-bypass scripts/window.ps1:335
    Runs PowerShell with execution policy bypassed (string literal in code, not executed)
    & $pwsh -NoProfile -ExecutionPolicy Bypass -File $input -Action mouse-click -X $tx -Y $ty -Button left 2>&1 | Out-Null
    code literal
  • low Dangerous commands cmd-execpolicy-bypass scripts/window.ps1:349
    Runs PowerShell with execution policy bypassed (string literal in code, not executed)
    & $pwsh -NoProfile -ExecutionPolicy Bypass -File $screen -Action capture -OutputPath $shot 2>&1 | Out-Null
    code literal

Files scanned: 14. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning description-no-when description does not say WHEN to use the skill (no "use when")

Process rating: all ten parameters 51/100

  • 0Result and completion. Does not say what the result is
  • 0Failures and branches. Linear process with no failure handling
  • 0Progress reporting. Says nothing while it works
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 2 mutating operations with no state check
  • 60Tools and files. Uses tools (bash) that frontmatter does not declare
  • 70Inputs and preconditions. Inputs and preconditions are listed
  • 100Steps. 15 steps
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 2648 tokens
  • low 17 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +3Description length 120: enough signal without eating the budget
  • +4Structure: 19 headings
  • +3Step-by-step instructions: 15 items
  • +4Has examples (14 code blocks)
  • +3All 10 scripts are documented

Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.

External checks

ClawHub: suspicious
This is a disclosed local Windows desktop-control skill, but it deserves Review because it can see, read, type, click, and manage processes across the desktop with limited built-in scoping.
LLM: suspicious (high) · 29 Aug 2026