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.
npx agentmods add skills/24kchengye/desktop-controller-skill/desktop-controller-skillnpx skills add 24kchengYe/desktop-controller-skill --skill desktop-controller-skillgit clone --depth 1 https://github.com/24kchengYe/desktop-controller-skillWrote 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/24kchengye/desktop-controller-skill/desktop-controller-skill)<a href="https://agentmods.dev/skills/24kchengye/desktop-controller-skill/desktop-controller-skill"><img src="https://agentmods.dev/badge/skills/24kchengye/desktop-controller-skill/desktop-controller-skill.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00304 | $0.03574 |
| Opus 5 | $0.00152 | $0.01787 |
| Sonnet 5 | $0.00061 | $0.00715 |
| Haiku 4.5 | $0.00030 | $0.00357 |
Grade A, and why
desktop-controller scanned grade A with 1 finding 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 5d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl --noproxy localhost -s http://localhost:9222/json/version How it starts
The opening of the file, as written. The whole thing — 331 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Universal Desktop Controller
Automate any Windows desktop application using a combination of Win32 API, Playwright (for web/Electron apps), and screenshot-based visual feedback. Inspired by OpenAI's playwright-interactive Codex skill, adapted for Claude Code.
Smart Routing: Choose the Right Engine
CRITICAL DECISION: Before any action, determine which engine to use:
User Request → Is it a web/browser/Electron app?
├── YES → Playwright (CSS selectors, instant, precise)
│ Examples: email webmail, Slack, browser tasks, web forms
│ ✅ page.click('#compose') — instant, 100% accurate
│
└── NO → Win32 API (native desktop automation)
Examples: WeChat, QQ, DingTalk, Notepad, File Explorer
✅ SetCursorPos + mouse_event — works for any native app
Why this matters: For the Tsinghua email (a web app), using Playwright CSS selectors is 10-50x faster than screenshot→analyze→click. OpenAI's playwright-interactive proved this: DOM selectors beat vision-based coordinate guessing every time for web content.
| Engine | Speed | Precision | Use For |
|---|---|---|---|
| Playwright | <1s per action | 100% (DOM selector) | Web apps, Electron apps, browser tasks |
| Win32 API | 2-3s per action | 95% (coordinate-based) | Native desktop apps (WeChat, QQ, etc.) |
| Screenshot+Vision | 10-15s per action | 80% (AI guessing) | Last resort / unknown UI |
Architecture
┌──────────────────────────────────────────────────┐
│ Desktop Controller Skill │
├──────────────┬──────────────┬────────────────────┤
│ Win32 Layer │ Playwright │ Visual Feedback │
│ (Native) │ (Web/Electron)│ (Screenshot+AI) │
├──────────────┼──────────────┼────────────────────┤
│ FindWindow │ Browser │ CaptureScreen │
│ SendKeys │ Page.click │ CaptureWindow │
│ SetCursorPos │ Page.fill │ → Claude Vision │
│ mouse_event │ Page.goto │ → Verify State │
│ Clipboard │ Locator │ → Decide Next Step │
└──────────────┴──────────────┴────────────────────┘
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.
- 5d ago First seen · 331 lines · 304 tokens per session scan A cba34b20c0bb
desktop-controller is a skill published in the GitHub repository 24kchengYe/desktop-controller-skill (10 stars, last pushed 5mo ago), licensed MIT. It adds 304 tokens to every session and 3,574 once invoked, about $0.0015 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work.
opencli-autofix
Automatically fix broken OpenCLI adapters when commands fail. Load this skill when an opencli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
opencli-browser-sitemap
Use when driving a website with opencli browser and sitemap context is available, requested, or needed to avoid blind navigation. Guides agents to consume site sitemap files lazily, choose adapter/browser fallback paths, resume from state signatures, and mark stale sitemap entries without trusting them over live…
pinchtab
Use this skill when a task needs browser automation through PinchTab: open a website, inspect interactive elements, click through flows, fill out forms, scrape page text, reuse a dedicated automation profile with user approval, export screenshots or PDFs, manage multiple browser instances, or fall back to the HTTP API…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.