Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent Chinese recipe helper that only uses local recipe data and low-sensitivity local JSON files for favorites and shopping lists.
Install only if you are comfortable with a Chinese-language recipe assistant saving recipe favorites and shopping-list items locally. Avoid storing sensitive dietary or personal notes in its lists on shared machines.
The README title, feature description, and usage examples are entirely in Chinese, with no indication that the skill supports other languages or that Chinese-only behavior is intentional and justified. This can indicate a language policy violation if the skill implicitly constrains interactions to a specific language without user opt-in.
The usage section lists example inputs like "今天吃什么" and "我有土豆和牛肉" without defining whether these are actual activation triggers, examples only, or constrained to a specific context. These phrases are common everyday speech and may cause unintended invocation if used as trigger conditions.
The skill activates on very broad phrases such as "今天吃什么" and "推荐菜", plus scenario descriptions like "用户不知道吃什么", without clear boundaries on when the skill should or should not trigger. These are common conversational phrases and the file provides no exclusion conditions or negative examples to prevent unintended invocation.
All user-facing natural-language strings and examples are in Chinese, and the file header labels the tool as a Chinese recipe assistant without indicating language choice or opt-in. This can violate language/locale policy when a skill imposes a specific language rather than letting the user choose or documenting a justified region-specific constraint.
All headings, triggers, examples, and instructions are written exclusively in Chinese, which effectively forces a single language experience. The file does not state that the skill is intentionally region-specific or offer the user an opt-in or alternative language.
The skill declares file read/write access for storing favorites but does not clearly warn users that recipe preferences and usage metadata may be persisted locally. While the stored sample data is low sensitivity, undisclosed local storage can create privacy and transparency issues, especially on shared systems or managed workspaces.
The function overwrites shopping_list.json with an empty list, which is a destructive data operation. Although the function has an internal docstring, there is no user-facing confirmation prompt or warning near the operation itself, and invoking it from the query handler immediately deletes stored list data.
