Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed AI-led research workflow with broad autonomy language, but no hidden code, persistence, credential handling, or destructive behavior was found.
Before installing, be aware that this skill encourages the agent to take initiative on research tasks, including pulling sources and proposing analyses. Use it for research where you are comfortable supervising the scope, validating methodology, and checking citations before relying on or publishing results.
The skill description does not warn users that the agent may autonomously gather data, run analyses, and perform research actions that could affect privacy, external resources, or local system state. Given the skill's explicit design around AI-led execution and 'full-cycle ownership,' users may unknowingly invoke behaviors with meaningful security, integrity, or confidentiality implications.
The activation guidance is broad enough to trigger on many ordinary research or knowledge-gap requests, which can cause the agent to assume 'full-cycle ownership' and autonomously scan literature, gather data, and run analyses in contexts where the user did not explicitly consent to that level of autonomy. In this skill, that risk is amplified by language encouraging proactive execution and independent gap detection, increasing the chance of overreach into privacy-sensitive or system-impacting actions.
No suspicious patterns detected.
