Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only helper for unit test coverage work, with broad trigger wording but no hidden execution, persistence, credential use, or destructive behavior.
Installers should be aware that this skill may trigger for broad software-quality wording, so it is best used when the user specifically wants help adding unit tests or improving coverage in an existing codebase.
The trigger sentence is broad enough to match ordinary user requests about testing or practical workflows, which can cause the skill to activate outside its intended scope. Over-broad activation increases the chance of prompt routing mistakes, irrelevant instruction injection into unrelated tasks, and user confusion about why this skill was selected.
The invocation guidance lists generic keywords and trigger phrases without meaningful scope boundaries, making the skill eligible for a wide range of common software-help queries. In an agentic environment, this can lead to inappropriate skill selection, unnecessary instruction precedence, and accidental suppression of more suitable or safer specialized skills.
The trigger sentence is broad enough to match many ordinary requests about testing or practical workflows, which can cause the skill to activate outside its intended scope. Over-broad invocation increases the chance of prompt hijacking-by-routing, where this skill is selected for general software requests and influences behavior in contexts not specifically about unit-test coverage improvement.
The trigger phrases are broad, natural-language requests that can easily match ordinary user speech unrelated to explicitly invoking this skill. That can cause unintended auto-activation, leading the agent to route requests into this skill when the user did not mean to, increasing the chance of incorrect tool use or context leakage across workflows.
The manifest description uses broad activation language such as 'software-and-data' and generic requests for workflows, analysis, or implementation support. This can cause the skill to be invoked for common engineering requests outside its intended scope, creating prompt-routing confusion and increasing the chance that unrelated tasks receive testing-focused instructions or artifacts.
The trigger keywords include generic terms like 'testing', 'regression', and 'quality' without qualifiers. These high-frequency words are likely to match many unrelated user requests, leading to overbroad activation and unintended delegation to this skill.
The example trigger sentences are phrased so generally that they provide weak gating signals for invocation. In systems that learn or pattern-match from examples, vague examples broaden the activation surface and can make the skill fire on ordinary requests that only loosely resemble the intended unit-testing use case.
The skill description enables use for a wide set of common topics like software-and-data, testing, and regression without clear boundaries. In an agentic environment, this ambiguity increases the chance of misrouting general requests into this skill, which can override more appropriate skills or produce off-target outputs.
The trigger keywords are broad and overlap with ordinary software-development conversations, making accidental activation likely. This can cause the agent to invoke the skill in contexts where it is not the best fit, leading to irrelevant guidance, prompt-scope confusion, or unintended handling of user requests.
The default prompt uses broad, everyday-language terms like 'help me' plus common software topics, which can cause the skill to be invoked in situations beyond the author's intended scope. Because implicit invocation is enabled, this increases the chance of over-triggering on general development requests and unintentionally inserting the skill into unrelated workflows.
The invocation examples are generic and do not define clear boundaries for when the skill should or should not be used. Ambiguous examples make accidental triggering more likely, which can misroute user requests and apply this skill's workflow in contexts where it is not appropriate.
No suspicious patterns detected.
