Security audit
Security checks for vulnerabilities and agentic risk
This is a low-risk documentation helper for OpenAPI/Swagger work, with some overly broad activation wording but no hidden execution, persistence, credential use, or destructive behavior.
Install this if you want help with OpenAPI or Swagger documentation. Be aware that its broad trigger wording may make it appear for general API or developer-experience questions, so confirm it is being used only for API documentation tasks and review generated specs before publishing them.
The trigger sentences are generic and mechanically templated, making it easy for unrelated user requests to accidentally match and invoke this skill. Unintended invocation can route users into irrelevant or lower-quality workflows, and in agentic environments it can widen the attack surface by activating capabilities outside the user's actual intent.
The trigger phrases are broad and can activate on generic requests about APIs, Swagger, or documentation without sufficient scoping. In an agent environment, this can cause unintended skill invocation, leading the system to apply this skill in contexts the user did not explicitly request, which increases the chance of misrouting, over-collection of context, or inappropriate autonomous assistance.
The skill description and activation guidance are broad enough to match many ordinary software or API-related requests, which can cause the skill to activate outside its intended niche. Over-broad activation increases the chance that users are routed into an irrelevant workflow, producing misleading guidance or overshadowing a more appropriate skill.
The trigger keywords include generic terms such as 'software-and-data' and 'developer experience', which are too unspecific to safely constrain invocation. This can lead to accidental activation on unrelated prompts, reducing reliability and potentially causing inappropriate automation decisions.
The skill description and usage criteria are broad enough that the skill may be invoked for loosely related requests, especially because it mixes domain terms with generic workflow help. Over-broad activation can cause incorrect routing, unintended disclosure of contextual data to the skill, or reduced reliability if the agent applies this skill where a narrower skill should have been used.
The trigger keyword list includes terms such as 'software-and-data' and 'developer experience' that are common, high-level categories rather than precise indicators of this skill's purpose. Ambiguous keywords increase the chance of accidental activation, which can lead to misrouting and unnecessary exposure of user context to a skill that was not actually needed.
The default prompt contains a very broad natural-language trigger phrase tied to common topics like OpenAPI, Swagger, API documentation, and REST APIs. Combined with implicit invocation, this increases the chance the skill is auto-selected during ordinary conversation, causing unintended routing of user requests and expanding the attack surface for prompt injection or unintended data exposure through the skill.
The trigger sentence begins with a broad natural-language phrase ('Help me ...') that can plausibly appear in ordinary user conversation, increasing the chance the skill is invoked unintentionally. In an agent environment, accidental activation can route user requests into the wrong workflow, causing prompt/skill confusion, irrelevant actions, or expansion of the skill’s influence beyond intended boundaries.
This trigger sentence is still overly permissive because it uses generic request language ('I need a practical workflow for ...') rather than a precise activation condition. Ambiguous triggers increase the risk of unintended invocation, especially when the skill description already spans broad documentation and implementation support tasks, making routing mistakes more likely.
The file is a zh-CN README, but key user-facing content such as the demand description remains in English. This creates an implicit language constraint mismatch for Chinese-language users without indicating that the user can choose their preferred language.
The example triggers are malformed, truncated, and not framed as precise activation boundaries, so they do not provide a dependable pattern for safe invocation. Ambiguous examples increase the likelihood of mistaken activation and make it harder for maintainers to validate the intended scope.
L48-L50 的示例句基本复述需求验证文本，语句不自然且没有展示与非触发场景的区别，无法帮助界定技能何时应被调用。缺少负面示例会增加误触发风险。
