Security audit
Security checks for vulnerabilities and agentic risk
This skill provides OpenAPI/Swagger documentation workflow guidance and contains no executable code, persistence, credential handling, or hidden data access; its main caveat is overly broad activation wording.
Installers should know this skill may activate for broad API or software-documentation prompts. It is otherwise a low-authority documentation helper; users who want tighter behavior should narrow triggers to explicit OpenAPI, Swagger, or REST API documentation requests.
The invocation section lists generic trigger patterns such as "Help me" and "I need a practical workflow," which overlap heavily with ordinary user speech and do not clearly bound when this skill should activate. The description also lacks exclusion conditions or negative examples, increasing the risk of unintended invocation.
The trigger sentence is so broad and unnatural that it could activate on loosely related user text rather than an explicit request to use this skill. Overbroad activation increases the chance of the wrong skill being selected, which can misroute user intent, cause irrelevant output, or crowd out safer or more appropriate skills.
The trigger sentences are broad and partially malformed, which increases the chance that the skill is invoked in contexts the user did not clearly intend. Unintended invocation can route unrelated prompts into this skill, causing confusing behavior, incorrect task handling, or disclosure of user context to an unnecessary workflow layer.
This is a Chinese README, but the usage section presents trigger phrases only in English and does not offer a Chinese-language alternative or state that language choice is optional. That can amount to a language policy issue because it implicitly forces a specific language for activation without documented user opt-in.
The skill description is broadly framed around common API-documentation tasks and can match a wide range of ordinary software requests without clear boundaries or exclusions. Overly broad activation increases the chance the skill is invoked in unintended contexts, which can cause incorrect routing, user confusion, or inappropriate application of workflow instructions.
The trigger keywords include generic phrases like 'software-and-data' and broadly applicable API/documentation terms without disambiguation rules. This makes accidental activation more likely, potentially causing the skill to intercept unrelated requests and apply unsuitable guidance or artifacts.
L04 将触发范围表述为“当用户提出 software-and-data, openapi, swagger, api documentation, rest api，或需要围绕该需求获得实用流程、产物、检查清单、分析或实现支持时使用”，其中既包含非常宽泛的标签（如 software-and-data），也没有清楚说明哪些请求不应触发该技能。这种描述容易与一般性开发求助场景重叠，缺少边界和排除条件。
The trigger keywords are broad and lack exclusion criteria, so the skill may be invoked for loosely related prompts rather than clear OpenAPI-documentation tasks. In an agentic environment, misrouting can cause the system to apply the wrong workflow, generate irrelevant artifacts, or process sensitive API material under incorrect assumptions.
The skill enables implicit invocation while providing a broad, underspecified description of when it should be used. This can cause the agent to auto-select the skill in contexts the user did not clearly intend, increasing the chance of prompt injection exposure, incorrect tool use, or unintended handling of sensitive API/specification content.
The activation guidance is ambiguous because it mixes broad keywords and generic trigger templates without clearly defining when the skill should or should not run. In a multi-skill agent, this ambiguity can lead to unintended invocation, reducing routing reliability and potentially exposing users to incorrect workflow steps or outputs not suited to their request.
The file lists language-specific skill artifacts for English and Chinese, but does not state how the user's preferred language is selected or whether language choice is optional. This can create an implicit locale policy without documented user opt-in.
该文件整体为 zh-CN 本地化版本，但文案中未说明语言选择机制，也没有告知这是可选语言版本或允许根据用户偏好切换语言。按规则，若技能在自然语言层面默认强制特定语言而无用户选择，可能构成语言/区域策略问题。
