Security audit
Security checks for vulnerabilities and agentic risk
This is a simple prompt-improvement skill with broad automatic activation wording, but it does not request tools, credentials, persistence, network access, or file access.
Install this if you want the agent to proactively help rewrite vague prompts. Be aware it may trigger more often than desired and may use Chinese template labels unless you instruct your agent to keep prompt suggestions in your preferred language.
The activation criteria are broad enough that the skill may trigger on many normal user messages that are merely somewhat underspecified. In an agent system, unintended automatic invocation can override normal handling, cause prompt rewriting without consent, and create reliability or instruction-following issues across unrelated tasks.
Several core principles and the prompt template are presented in Chinese, while the surrounding document is in English, and the skill does not state that users can choose their preferred language. This creates an implicit language constraint that may conflict with policy requiring user opt-in or documented justification for locale-specific behavior.
The statement that the skill 'activates automatically when detecting vague input' lacks boundaries, thresholds, and disallowed contexts. This can lead to pervasive interception of user requests, producing unwanted reformulation behavior and increasing the chance that the agent misroutes or transforms tasks that should be handled directly.
