Security audit
Security checks for vulnerabilities and agentic risk
The skill is markdown-only and shows no code execution or persistence, but its JSON-transformer purpose is mixed with unrelated GNN/Transformer learning content and triggers.
Review this skill carefully before installing. It does not appear to execute code or access private data, but it is mislabeled and cluttered with unrelated learning content, so it may activate for the wrong requests or give irrelevant guidance for JSON work.
The trigger phrases are broad and map to very common JSON-related requests, increasing the chance this skill will activate in situations where a more specific or safer skill should handle the task. Because the skill content is already semantically inconsistent, overbroad activation materially raises the risk of inappropriate behavior or prompt/skill hijacking through routine user language.
The embedded sub-skill content clearly conflicts with the declared purpose of a JSON transformation skill by introducing unrelated GNN/trajectory-prediction material. This kind of semantic drift can cause the agent to invoke the wrong instructions for common JSON tasks, leading to misrouting, degraded reliability, and potentially unsafe behavior if unrelated capabilities are executed under a trusted skill name.
The secondary trigger text uses vague help-seeking language tied to the unrelated embedded sub-skill, which can cause accidental invocation from ambiguous user requests. In the context of a mislabeled skill, ambiguous activation broadens the attack surface for confusion, unintended routing, and execution of instructions outside the user's actual task.
The file presents user-facing descriptions in both English and Chinese, and the body content is primarily Chinese, but it does not state any user-selectable language preference or opt-in behavior. This can create a locale/language policy issue if the skill defaults to a language the user did not request.
No suspicious patterns detected.
