Windows-MCP is an MCP server that lets AI agents operate Windows computers through actions such as navigating files, opening applications, controlling windows, and interacting with user interfaces. It is used for desktop automation, software testing, and other tasks that require an agent to use Windows.
Borrowing it
Nothing to install: this file belongs to CursorTouch/Windows-MCP. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/CursorTouch/Windows-MCP/main/.claude/skills/windows-mcp-tool-tester/SKILL.mdgit clone --depth 1 https://github.com/CursorTouch/Windows-MCPWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cursortouch/windows-mcp/windows-mcp-tool-tester)<a href="https://agentmods.dev/skills/cursortouch/windows-mcp/windows-mcp-tool-tester"><img src="https://agentmods.dev/badge/skills/cursortouch/windows-mcp/windows-mcp-tool-tester/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cursortouch/windows-mcp/windows-mcp-tool-tester"><img src="https://agentmods.dev/badge/skills/cursortouch/windows-mcp/windows-mcp-tool-tester.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 87 Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.Fix: Set explicit rate limits, timeouts, and resource quotas for API calls, file operations, and compute. Implement circuit breakers for runaway loops.
- medium Prompt Injection · line 250 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00139 | $0.06451 |
| Opus 5 | $0.00069 | $0.03225 |
| Sonnet 5 | $0.00028 | $0.01290 |
| Haiku 4.5 | $0.00014 | $0.00645 |
Grade A, and why
windows-mcp-tool-tester 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 11d ago.
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.
Copies of this mod
2 near-identical copies found in the catalogue:
- windows-mcp-tool-tester — 100% identical, 0 lines differ
- windows-mcp-tool-tester — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 586 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Windows-MCP Tool Tester
An automated testing skill that generates comprehensive test cases for a single Windows-MCP tool, executes them, and produces a structured test report with pass/fail results, performance metrics, and actionable recommendations.
Core Principles
- One tool per invocation. If the user doesn't specify which tool to test, ask them before proceeding.
- Black-box testing only. Derive test cases exclusively from the MCP tool description and parameter schema — never read source code. Silence in the schema is a documentation gap, not a testing hint.
- Auto-generate test cases from the tool's MCP description and parameter schema. Cover common scenarios, edge cases, parameter combinations, and error handling paths.
- Measure two dimensions: correctness (return value matches expectations) and response time (end-to-end, including MCP overhead).
- Mandatory side-effect verification: every tool call that may modify system state MUST be independently verified — no exceptions, no sampling.
- Safe cleanup: track process PIDs spawned during testing; only kill those specific PIDs during teardown, never kill by process name alone.
- Safety first: Windows-MCP has full system access with no sandboxing. Tests involving
destructive tools (FileSystem delete, Registry set/delete, Process kill, PowerShell) can
modify or destroy data. Running in a VM or Windows Sandbox is strongly recommended.
Before executing destructive test cases, confirm the user accepts the risk. See
SECURITY.md. - Produce a structured report at the end (see Step 4).
Step 0: Identify the Target Tool
If the user hasn't specified a tool, present the full list and ask them to pick one:
App, PowerShell, Screenshot, Snapshot, Click, Type, Scroll, Move, Shortcut, Wait, MultiSelect, MultiEdit, Clipboard, Process, Notification, FileSystem, Registry, Scrape
Once a tool is confirmed, proceed to Step 1. Do NOT test multiple tools in one session.
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.
- 11d ago First seen · 586 lines · 139 tokens per session scan A df6dfb97f7b3
windows-mcp-tool-tester is a skill published in the GitHub repository CursorTouch/Windows-MCP (6,964 stars, last pushed 4d ago), licensed MIT. It adds 139 tokens to every session and 6,451 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
verification-protocol
Mandatory end-to-end verification for .NET MAUI and Comet app tasks. Enforces dependency ownership — broken tools and missing config are YOUR problem to fix, not blockers to report. USE FOR: any task touching UI, pages, navigation, services, or app behavior. Triggers: after code changes, before marking todos done…
maui-unit-testing
Add MAUI app tests around ViewModels, services, platform abstractions, and unit/integration/device boundaries. USE FOR: xUnit, avoiding MauiProgram.CreateMauiApp in unit tests, fakes/mocks, injectable Essentials interfaces, PlatformNotSupportedException, navigation/data services, AppProjectReference, device-test…
host-computer-use
Beta desktop control through the connected A0 CLI host. Use for the user's host/local computer screenshots, screen inspection, menus, native app UI, OS-level clicking, scrolling, typing, or checking computeruseremote status. Use instead of linux-desktop for host/local machine control. Do not use for ordinary browser…
robotgo-flow
Use when building YAML-driven Windows RPA workflows in Go — step-by-step desktop automation with image template matching, interactive recording mode, hotkey triggers. RobotGo-Flow: YAML-based Windows RPA framework built on RobotGo.
testing-expert
Expert-level software testing with unit tests, integration tests, E2E tests, TDD/BDD, and testing best practices. Use when the user mentions TDD, BDD, unit tests, integration tests, or end-to-end tests, or when the task involves Testing Fundamentals, Unit Testing, Integration Testing, or End-to-End Testing.
electron-expert
Expert in Electron framework, desktop app development, IPC, and cross-platform packaging. Use when the user mentions desktop, Node.js, cross platform, Windows, macOS, or Linux, or when the task involves Electron Architecture, Process Types, IPC Communication, or App Lifecycle.