Security audit
Security checks for vulnerabilities and agentic risk
This is a prompt-only educational skill about the letter B, with no code execution or sensitive access, though its activation wording is broader than ideal.
This skill is safe to install as a static knowledge prompt. Be aware that it may activate on broad references to the letter B, so users may want more precise routing language if they use many domain-specific skills.
The README describes the skill as activating whenever a user 'mentions or implies topics related to the letter B,' which is unusually broad for a prompt-based skill. This can cause unintended invocation in many unrelated conversations where 'B' appears incidentally, leading to overreach, response hijacking, or reduced reliability of skill routing rather than direct code execution risk.
The listed triggers include highly ambiguous contexts such as science, music, grading, ratings, and codes involving 'B,' all of which are common across many unrelated user requests. In a prompt-based agent system, this broad matching increases the chance the skill will activate outside its intended scope and interfere with more relevant skills, creating prompt-routing and context-confusion risk.
The activation criteria are extremely broad, including any mention of the letter "B" and many loosely related domains such as science, music, grading, and computing. This can cause the skill to trigger unintentionally in unrelated conversations, leading to prompt hijacking of the assistant’s response flow, reduced reliability, and possible interference with higher-priority or safer domain-specific skills.
The header 'Usage / 使用方法' introduces Chinese alongside English, but the document does not explain whether multilingual output is optional or user-selected. This can be read as an implicit language/locale choice without explicit opt-in.
No suspicious patterns detected.
