Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only helper for unit-test coverage work, with no evidence of hidden execution, credential access, persistence, or data exfiltration.
Installers should be aware that the skill may activate for broad testing or quality requests because implicit invocation is enabled, but its contents are documentation-only and should be safe for normal unit-test coverage assistance.
The invocation guidance lacks clear boundaries for when the skill should and should not be used, making it easy for a dispatcher or user to invoke it for loosely related software requests. In an agent ecosystem, ambiguous routing guidance can lead to unintended tool/skill selection, expanding the attack surface and enabling mis-scoped handling of requests.
The trigger sentence is excessively broad and matches common help-seeking phrasing, which can cause the skill to activate in situations not specifically intended for unit-test coverage assistance. Overbroad activation increases the chance of prompt/skill hijacking at the orchestration layer by routing unrelated requests into this skill and crowding out more appropriate safeguards or specialized skills.
The README lists activation phrases such as "Help me ..." and "I need a practical workflow for ...", which are natural-language patterns that overlap with common everyday requests. The file does not provide exclusion conditions, negative examples, or clear limits on when the skill should or should not activate, increasing the risk of unintended invocation.
The trigger keywords are broad enough to match many ordinary software-assistance requests, which can cause the skill to activate outside its narrowly intended scope. Over-broad activation is dangerous because it can overshadow more appropriate skills, route user requests through the wrong workflow, and increase the chance that unreviewed instructions in this skill influence unrelated tasks.
The example trigger phrases use generic everyday wording like 'Help me' and 'I need a practical workflow', which can train or encourage invocation on vague requests that are not specifically about unit-test coverage work. This broadens the skill's effective scope and increases the likelihood of accidental or inappropriate routing, especially in environments with multiple overlapping skills.
The skill description says it should be used for broad categories like 'software-and-data', 'testing', and 'regression', without clear limits on what falls outside scope. This is risky because ambiguous activation criteria can lead to accidental invocation on unrelated engineering tasks, increasing the chance of misrouting, confusing outputs, and unintended handling of requests better served by other specialized skills.
The trigger list includes broad generic terms such as 'testing' and 'quality', which can cause the skill to activate outside its intended unit-test and coverage scope. Over-broad activation is dangerous because it can route unrelated requests into this skill, producing irrelevant guidance, masking better-matched skills, or expanding the skill’s effective authority beyond what users expected.
The default prompt uses very broad, everyday terms such as 'help me' and general testing-related language, which can cause the skill to be invoked in situations beyond the user's specific intent. Because implicit invocation is enabled, this increases the risk of accidental routing, context capture, or unintended assistance being injected into unrelated conversations.
The trigger sentence is so broad and natural-language-like that it can match routine user phrasing unrelated to an explicit request for this skill, causing the agent to invoke the unit-test-coverage helper unexpectedly. In a skill-routing system, overly generic triggers increase the chance of misrouting, accidental activation, and inappropriate access to project context or code-analysis workflows.
The file identifies itself as the Chinese-language user guide via README.zh-CN.md, which encodes a specific locale variant (Simplified Chinese, China) but does not state that this is an optional localization or offer language/locale choice within the content. Under the policy, locale constraints should be opt-in or clearly justified.
