Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only OpenAPI documentation helper with some overly broad activation wording but no hidden execution, data access, persistence, or destructive behavior.
This skill is reasonable to install for OpenAPI or Swagger documentation tasks. Be aware that its broad trigger terms and implicit invocation could make it appear during generic REST API or developer-experience conversations, so confirm it is the right helper when a request is not clearly about API documentation.
The trigger sentences are broad and map to common user requests such as asking for help or a workflow, which increases the chance of accidental or inappropriate skill activation. In an agent system, ambiguous invocation conditions can cause the wrong skill to engage, leading to irrelevant handling, prompt-scope confusion, or unintended access to downstream capabilities.
The individual trigger phrases overlap with everyday speech and are not sufficiently distinct from normal conversation. This makes accidental activation more likely, especially in multi-skill environments where broad phrases can route unrelated requests into this skill and degrade safety and correctness.
The trigger phrases are broad and generic enough that the skill could activate for loosely related requests without strong scoping or user intent verification. In an agent setting, this can cause inappropriate invocation, expanding the skill's influence over conversations and increasing the chance of unintended actions or disclosure through over-application of its workflow.
The skill description and activation guidance are broad enough to match many ordinary software requests beyond narrowly scoped OpenAPI documentation work. Over-broad routing can cause the agent to invoke this skill in unintended contexts, leading to irrelevant guidance, reduced security review quality, or accidental handling of tasks the skill was not designed to perform.
The trigger keywords include generic phrases like software-and-data and developer experience, which are likely to overlap with many unrelated user requests. In agent routing systems, such generic triggers increase false activations, which can misdirect workflows and potentially bypass more appropriate specialized skills or safeguards.
The skill description activates on very broad categories such as software-and-data, openapi, swagger, api documentation, and rest api, without clear boundaries or disambiguation rules. This can cause the skill to trigger during many ordinary engineering conversations, increasing the chance of unintended invocation and inappropriate steering of the agent's behavior.
The trigger keywords include highly generic terms like developer experience and broad technical phrases that overlap with routine discussions, making accidental activation likely. In an agent environment, overly general triggers can misroute user requests, inject irrelevant workflow instructions, and reduce reliability or safety of task selection.
The default prompt trigger phrase is broad and maps to common user language about APIs, OpenAPI, Swagger, software, and documentation, which increases the chance of unintended or automatic invocation. Because implicit invocation is also enabled, ordinary requests may activate this skill when the user did not clearly intend to call it, expanding the attack surface for prompt injection, data leakage, or unexpected agent behavior.
The trigger description spans multiple lines and lacks clear scope boundaries, making it ambiguous when the skill should be invoked. Ambiguous routing criteria can cause over-triggering or misclassification of user intent, especially because the skill description includes broad software-and-data and API-related terms that appear in many benign conversations.
The trigger sentence on line 80 is overly broad and can match ordinary user phrasing, causing the skill to activate outside its intended OpenAPI documentation scope. This creates a prompt-routing vulnerability where unrelated requests may be captured and handled by this skill, increasing the chance of irrelevant, misleading, or unsafe outputs in downstream workflows.
The malformed example trigger sentences do not clearly communicate when the skill should be invoked, which can lead to ambiguous or unstable routing behavior. While not directly exploitable like code execution, unclear activation examples increase misclassification risk and make it harder to validate that the skill will only run in intended situations.
