Security audit
Security checks for vulnerabilities and agentic risk
This is a plain Markdown planning skill that guides task breakdown and does not request sensitive access, persistence, or code execution.
Install this if you want Chinese-oriented task planning and decomposition help. Users who prefer another language or only want explicit activation may want the publisher to narrow the trigger and state language-matching behavior, but no malicious or high-impact behavior was found.
The skill content is written to produce Chinese-language output by default without indicating that the agent should respect the user's language preference. While this is not a direct security exploit, it can degrade usability, cause misunderstanding of safety-critical instructions, and interfere with downstream systems expecting another language.
The activation condition is broad enough to trigger on many ordinary user requests involving multi-step work, which can cause the skill to intervene when not explicitly requested. In an agent system, overly broad automatic activation can change behavior unexpectedly, increase prompt-surface exposure, and crowd out more appropriate or safer skills.
No suspicious patterns detected.
