任务守望者与弹窗杀手 (Task Watchdog)

挂机任务的守护者。定时检查屏幕进度条，自动关闭干扰性系统或广告弹窗，遇致命错误时发送报警。

This skill is coherent with its stated task but asks the agent to do sensitive things: take screenshots, simulate clicks, and optionally send alerts over the network. Before installing, consider: 1) Only enable this skill in a trusted environment and on machines where screen contents are not sensitive. 2) Prefer an explicit, vetted webhook URL (or require manual confirmation) and avoid sending screenshots to unknown endpoints. 3) Test in a controlled session so you can observe what the agent clicks — visual recognition is error-prone and could trigger 'restart' or 'install' actions despite the skill's prohibition on killing processes. 4) If you cannot accept any risk of data leakage from screenshots or automated clicks, do not enable this skill. Additional information that would raise confidence: explicit declarations of required GUI/input permissions, restrictions on allowed network endpoints, an explicit confirmation step before transmitting screenshots, and limits on click targets (e.g., click only within the locked task window).

Type: OpenClaw Skill Name: task-watchdog Version: 0.1.0 The OpenClaw skill 'task-watchdog' is designed to monitor long-running tasks, close pop-up windows, and send alerts if a task stalls. The `SKILL.md` explicitly disallows shell access (`shell: allowed: false`) and includes strong safety constraints, such as prohibiting process termination and preventing clicks on dangerous UI elements like 'Restart system' or 'Update and install'. Network access is allowed only for sending user-configured Webhook alerts, which is consistent with its stated purpose. There is no evidence of prompt injection, data exfiltration, malicious execution, or obfuscation. The instructions are clear, focused, and prioritize user safety.