Security audit
Security checks for vulnerabilities and agentic risk
This is a small helper skill that suggests common OS commands for opening files, folders, URLs, or repositories, with no executable code or hidden behavior found.
Before installing, be aware that the skill name is very short, so it may activate more often than a more specific opener skill. Use it when you want quick commands for opening a clearly identified file, folder, URL, or repository, and verify the suggested OS command matches your actual system.
The skill name 'o' is extremely short and the description positions it as a broadly applicable opener skill, making accidental invocation and collisions with ordinary user language more likely. In an agent setting, ambiguous triggering can route unrelated requests into this skill and cause the assistant to emit commands the user did not intend to receive.
The invocation examples rely on very common phrases like 'open this' or 'how do I open this from the terminal' without strong guardrails on what counts as an eligible target. This broad matching increases the chance of misfires in unrelated conversations, which is risky because the skill is designed to produce executable OS commands.
The skill content is presented in Chinese and directs a specific response style without indicating that the user can choose another language. Because language/locale policy violations apply to all file types, this can be considered a forced language preference unless the user has explicitly opted into Chinese.
No suspicious patterns detected.
