QwenPaw is a personal AI assistant that runs on a local machine or in the cloud and connects to multiple chat applications. It provides memory, file workspaces, multiple agents, skills, plugins, and integrations with language-model providers and external tools. The catalogue entries are skills that extend its capabilities.
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 skills add agentscope-ai/QwenPaw --skill computer_usegit clone --depth 1 https://github.com/agentscope-ai/QwenPawWrote 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/agentscope-ai/qwenpaw/computer_use)<a href="https://agentmods.dev/skills/agentscope-ai/qwenpaw/computer_use"><img src="https://agentmods.dev/badge/skills/agentscope-ai/qwenpaw/computer_use.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 252 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00038 | $0.02498 |
| Opus 5 | $0.00019 | $0.01249 |
| Sonnet 5 | $0.00008 | $0.00500 |
| Haiku 4.5 | $0.00004 | $0.00250 |
Grade A, and why
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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 258 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Computer Use
Use Computer Use only for tasks that require a live desktop interface or visual verification. Prefer a purpose-built integration or command-line tool when it can complete and verify the task.
Use only the native desktop runtime. It operates on one approved application and one observed window at a time; it never accepts a free-form screen target.
Operating Loop
Follow this loop for every task:
- Discover the canonical application and the correct window.
- Observe the window and identify the requested state from current evidence.
- Define the next expected visible or accessible state change.
- Choose one action channel and perform the smallest useful action.
- Inspect the replacement observation before deciding the next action.
- Observe the final state and verify every requested outcome before reporting success.
Treat dispatched: true or an intermediate acknowledgement only as evidence
that input was sent, not that the application completed the operation. If the
final state is incomplete or uncertain, report that accurately.
Discover the Target
- Call
list_appsand select the canonical App ID. - Call
list_windows, optionally limited by that App ID. - Match the target by title, content, and observed state. When several windows are plausible, observe them read-only until one matches; never choose only because it is first or most recent.
- Keep using the matched
window_iduntil an action explicitly hands off to another window.
Use launch_app with a canonical App ID. If the application is not listed,
use an explicit absolute executable path on Windows or application-bundle path
on macOS. After launch, list its windows again because launch completion does
not prove that a usable window already exists.
When the runtime reports a missing system permission, stop and ask the user to grant it. Do not retry until the user confirms the permission was granted.
Read an Observation
observe_window returns a point-in-time window observation with screenshots
and accessibility text. Start with:
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.
- 8d ago First seen · 258 lines · 38 tokens per session scan A 4ee7391581a7
computer_use is a skill published in the GitHub repository agentscope-ai/QwenPaw (35,042 stars, last pushed today), licensed Apache-2.0. It adds 38 tokens to every session and 2,498 once invoked, about $0.0002 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
browser-setup
A setup guide for connecting Clacky's browser tool to Chrome or Edge. It supports macOS, Linux, and Windows installations used through WSL, and stores the connection settings in a configuration file.
channel-manager
Configure IM platform channels (Feishu, WeCom, Weixin, Discord, Telegram, DingTalk) for openclacky. Uses browser automation for navigation; guides the user to paste credentials and perform UI steps. Trigger on: "channel setup", "setup feishu", "setup wecom", "setup weixin", "setup wechat", "setup discord", "setup…
web-reader
Implement web page content extraction capabilities using the z-ai-web-dev-sdk. Use this skill when the user needs to scrape web pages, extract article content, retrieve page metadata, or build applications that process web content. Supports automatic content extraction with title, HTML, and publication time retrieval.
azure-playwright-workspaces
Expert knowledge for Playwright Workspaces development including troubleshooting, best practices, decision making, limits & quotas, security, and configuration. Use when managing Playwright Testing workspaces, tokens/RBAC, quotas, monitoring/metrics, or run/AADSTS7000112 issues, and other Playwright Workspaces related…
control-browser
Research and inspect pages through Lime's managed browser session.
openclaw-browser-auto
A setup guide for connecting OpenClaw, an AI-agent platform, to a remote Chrome browser through CDP, a browser control interface. It covers Docker-hosted browsers and Browserless.io, a hosted browser service.