Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation helper for OpenAPI/Swagger work, with some broad trigger wording but no code execution, persistence, credential handling, or hidden data flow.
This skill is reasonable to install if you want help with OpenAPI or Swagger documentation. Be aware that its activation phrases are broad, so it may trigger on some general API or developer-experience requests; users should invoke it explicitly for OpenAPI/Swagger tasks or narrow the trigger wording if precise routing matters.
The trigger sentence begins with an extremely broad conversational phrase ('Help me ...'), which can cause the skill to activate for many unrelated user requests. Overbroad activation increases the chance of unintended routing, causing the agent to apply this skill in contexts where OpenAPI documentation support was not actually requested.
The trigger sentences are generic and template-like, making it easy for loosely related prompts to activate this skill unintentionally. In an agent environment, over-broad activation can cause the wrong workflow to run, leading to prompt-routing mistakes, irrelevant document generation, or inappropriate access to contextual API material.
The documented trigger phrases are broad and template-like, matching generic requests such as 'help me' or 'I need a practical workflow' with only loosely specific domain text. This can cause unintended invocation of the skill in unrelated conversations, increasing the chance that the agent routes user input into the wrong workflow or exposes unnecessary processing paths.
The skill description is scoped to a broad class of software-and-data requests and includes generic activation language like OpenAPI, Swagger, and API documentation without clear exclusion boundaries. This can cause over-invocation, where the agent applies the skill in unrelated contexts, increasing the chance of irrelevant guidance, prompt-surface expansion, or interference with more appropriate skills.
The trigger keywords include generic terms such as software-and-data, openapi, swagger, api documentation, and rest api, several of which are common across many benign engineering requests. Overly generic triggers make accidental invocation likely, which can route user requests into an unnecessary or incorrect workflow and broaden the skill's operational reach beyond its intended domain.
The skill description and trigger scope are broad and include generic terms like software-and-data, openapi, swagger, api documentation, and rest api, which can cause the skill to activate in conversations that only partially match the intended task. This increases the chance of unintended invocation, leading to irrelevant guidance, context hijacking, or interference with a more appropriate skill, though it does not directly enable code execution or data exfiltration.
The default prompt contains a very broad natural-language trigger covering common terms like software, data, API documentation, OpenAPI, Swagger, and REST API workflows. Combined with allow_implicit_invocation=true, this increases the chance the skill is invoked in ordinary conversations without clear user intent, which can cause prompt routing hijack, unexpected tool use, or context exposure to the skill.
These trigger sentences are ambiguous and do not define clear activation boundaries beyond repeating a long requirement statement. Ambiguous triggers can misclassify user intent and invoke the skill for loosely related workflow requests, reducing routing integrity and potentially bypassing more appropriate skills or safeguards.
The README lists separate English and Chinese instruction and guide files, but does not state how the user's preferred language is selected or that language choice is optional. This can create a language/locale policy concern if the skill defaults users into a language without explicit opt-in.
This is a zh-CN README, but several key instructional fields, including the demand description, workflow description, and trigger phrases, are written in English. That creates an implicit language constraint and inconsistent locale behavior without stating whether users can choose their preferred language.
The example trigger sentences are malformed, truncated, and not representative of realistic user requests, which reduces their value as safe invocation guidance. Poor examples make routing behavior less predictable and can lead to incorrect activation or failure to activate when needed.
The example trigger phrases are unnatural and overly templated, and they do not meaningfully restrict when the skill should be used. Weak trigger examples can cause imprecise routing behavior and accidental invocation because they reinforce broad matching without clarifying boundaries or required intent.
No suspicious patterns detected.
