Security audit
Security checks for vulnerabilities and agentic risk
This is a simple unit-test coverage helper skill with broad activation wording, but no evidence of hidden execution, data theft, persistence, or destructive behavior.
Install this if you want a lightweight workflow helper for unit tests and coverage. Be aware it may activate for broad testing or quality-related requests because implicit invocation is enabled and the trigger terms are loose; explicit invocation with $unit-test-coverage-helper is clearer.
The trigger phrase is so generic that it can activate on ordinary user language unrelated to explicit intent to use this skill, causing the agent to route requests unexpectedly. In a skill-selection system, broad triggers can be abused for prompt hijacking of workflow selection or simply degrade reliability by invoking this skill for many unrelated software questions.
The invocation guidance does not define clear boundaries for when this skill should or should not be used, which makes it prone to accidental selection for generic software requests. In an agent system, ambiguous routing can degrade reliability, expose unrelated contexts to unnecessary instructions, and create opportunities for skill misuse through crafted prompts.
The trigger phrase is extremely broad and can match ordinary requests about testing or software help, causing the skill to activate outside its intended scope. Over-broad activation increases the chance of prompt-routing errors, where this skill overrides more appropriate skills or injects irrelevant instructions into unrelated workflows.
The trigger phrases are very broad and map to common requests about testing, coverage, and practical workflows. In an agent-routing system, this can cause the skill to activate in situations where the user did not explicitly request it, leading to misrouting, unnecessary capability exposure, or unintended execution paths. The skill context makes this somewhat more concerning because testing-related requests are common across many software tasks, so overlap is likely.
The manifest description is broadly worded and includes many generic phrases such as software-and-data, testing, analysis, and implementation support, which can cause the skill to activate for requests outside its intended scope. Over-broad routing can expose users to irrelevant or unintended instructions, increase prompt-surface area, and make it easier for a loosely related request to be hijacked into this skill’s workflow.
The trigger examples use very generic phrasing like 'Help me' and 'I need a practical workflow' paired with repeated requirement text, without strong scope constraints. This increases the chance of accidental invocation from ordinary user requests, causing misrouting and unnecessary application of the skill in contexts where it may not fit.
The skill description defines activation in very broad terms across multiple common software topics, which can cause the agent to invoke this skill outside its intended niche. Over-broad routing is dangerous because it can override a more appropriate skill, produce irrelevant test-coverage workflows for unrelated requests, and increase the chance that untrusted repository content is handled under the wrong operational assumptions.
The keyword list includes generic terms like testing, regression, and quality without qualifiers, making accidental triggering likely in many unrelated engineering conversations. In an agentic system, this kind of ambiguous activation broadens the skill's authority surface and can lead to misrouting, confusing outputs, or unsafe assumptions about the user's real intent.
The default prompt is phrased so broadly that it can match many ordinary requests about software, tests, regression, or implementation help, increasing the chance of unintended or implicit invocation. Because implicit invocation is enabled, this creates a prompt-scope and consent issue where the skill may activate when the user did not clearly ask for it, which can inject unexpected instructions or workflow changes into unrelated conversations.
The file identifies itself as the Chinese-language user guide without indicating that language is optional or that users may choose another locale. This can create a language/locale policy issue when presented as the operative skill documentation without explicit opt-in.
The example trigger sentence starts with a highly generic phrase, 'Help me', which reinforces loose matching behavior and teaches the routing layer to associate vague requests with this skill. While less severe than the broad description and keyword scope, it still contributes to unintended invocation and reduced precision in skill selection.
No suspicious patterns detected.
