Security audit
Security checks for vulnerabilities and agentic risk
This is a static park-operations assistant skill with disclosed Chinese and English content and no executable behavior or hidden access.
This appears safe to install as a static informational skill. Before production use, replace the sample park data, confirm policy and rent figures are accurate, and consider tightening activation examples so the skill is only used when the user clearly asks for park operations help.
The quick-start phrases are short, generic imperatives like '查询园区概况' and '获取物业服务' that could be triggered during ordinary conversation rather than as clearly scoped commands. In an agent skill, overly broad trigger phrases can cause unintended skill activation or routing, leading the assistant to answer from this skill in contexts where the user did not explicitly intend to invoke it.
The scenario examples use natural, everyday requests in both Chinese and English, such as asking about rent, incentives, or statistics, without distinguishing whether these are activation triggers or only in-skill prompts. This increases the risk of accidental invocation, prompt collision with other skills, or unintended handling of general user queries by this skill.
The file is predominantly structured around Chinese interactions, and several sections provide only Chinese examples while no instruction indicates that users may choose their preferred language. Under the policy rule, locale-specific behavior should either be optional or explicitly justified as region-specific.
This YAML file defines all visible prompts, buttons, and messages exclusively in Simplified Chinese, which can amount to a language/locale policy violation if the skill forces that locale without user opt-in. Because the file contains no natural-language indication that Chinese is optional or that the skill is explicitly region-specific, it may constrain user interaction to a single language by default.
This manifest-style YAML contains most user-facing values, labels, and process descriptions in Chinese, with only limited English duplication for a few fields. For a generally scoped skill/config file, forcing a single language/locale without user opt-in can violate language/locale policy unless the regional constraint is explicitly documented and justified.
The example JSON explicitly sets the language to "zh-CN", which can imply a fixed locale behavior. Under the policy criteria, forcing a specific language or locale without user opt-in or a documented regional justification is a natural-language policy concern.
The second JSON example fixes the output language to "en-US" without any accompanying statement that the user can choose the locale. This may conflict with the language/locale policy if interpreted as mandated behavior rather than a configurable example.
This file is explicitly an English (en-US) language pack, and the user-facing strings are fixed to that locale. Under the policy, locale-specific behavior should either offer user choice or clearly document why the restriction is justified; no such opt-in or justification appears in this file.
No suspicious patterns detected.
