Security audit
Security checks for vulnerabilities and agentic risk
This skill is a documentation helper with some overly broad activation wording, but it does not include hidden execution, persistence, credential handling, or destructive behavior.
Installers should know this skill may activate on broad API/documentation phrasing, so explicit invocation is preferable when using it. Its inspected contents are otherwise limited to OpenAPI/Swagger documentation assistance.
The trigger sentence begins with a generic conversational phrase ('Help me ...'), which is likely to appear in ordinary user requests unrelated to explicit skill invocation. This can cause unintended activation of the skill, leading the agent to route prompts into this capability when the user did not mean to invoke it.
The trigger sentences are generic and awkwardly templated, making them likely to match user requests that only loosely relate to OpenAPI documentation. In an agent-routing context, this can cause unintended invocation of the skill, leading to irrelevant actions, confusion, or unsafe context capture when a more appropriate skill should have been selected.
The trigger phrases are so generic that the skill may activate for loosely related requests, causing unintended invocation and increasing the chance that unrelated or sensitive user tasks are routed through this skill. In an agent ecosystem, broad activation can create prompt-routing confusion, expose extra context to the skill, and make downstream misuse easier even if the file contains no direct code execution.
The skill description and activation criteria are broad enough to match generic software, API, or documentation requests that may not specifically need this skill. Over-broad routing can cause the agent to invoke the skill in unintended contexts, leading to irrelevant guidance, reduced reliability, and possible overshadowing of more appropriate skills.
The example trigger phrases use common conversational language such as 'Help me' and 'I need a practical workflow' with insufficient specificity, which increases the chance of accidental invocation from unrelated user requests. In an agent-routing system, this can degrade prompt selection accuracy and create confused-deputy behavior where the wrong skill handles a task outside its intended scope.
The skill description and activation guidance use broad topical phrases like OpenAPI, Swagger, API documentation, and REST API without enough contextual constraints or exclusion criteria. This can cause unintended invocation in loosely related conversations, increasing the chance that the skill overrides a better-suited workflow or processes sensitive API-related content when not necessary.
The keyword list consists of common technical terms that appear in many normal engineering conversations, but it does not define scope boundaries, disambiguation rules, or negative triggers. In a routing system, this can lead to over-broad skill activation, causing irrelevant guidance, prompt interference, or accidental handling of requests outside the skill's intended documentation-focused purpose.
The default prompt is framed as a broad natural-language trigger tied to a wide set of user intents, and the policy also allows implicit invocation. This can cause the skill to activate in contexts where the user did not clearly consent to using this specific capability, increasing the chance of prompt-routing mistakes, unintended data exposure to the skill, or overreach into unrelated conversations.
The sentence 'I need a practical workflow for ...' is too ambiguous because it describes a common request form without sufficiently constraining the subject or requiring explicit skill selection. In a multi-skill environment, this increases the chance of accidental invocation and incorrect routing based on vague overlap with other planning or workflow-related skills.
The file lists English and Chinese instruction and guide files, but does not state how the user's preferred language is selected or whether language choice is optional. This can create a locale-policy concern if the skill defaults users into a language without explicit opt-in.
No suspicious patterns detected.
