Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only unit-test helper skill with no hidden code execution, persistence, credential handling, or data exfiltration behavior found.
Installers should know this skill may be invoked by broad testing-related phrases because implicit invocation is enabled. Review or narrow the trigger language if precise routing matters, especially in environments with multiple software-engineering skills.
The invocation guidance lists keywords and trigger sentences but does not define concrete boundaries, exclusions, or disambiguation rules. This makes routing permissive and can lead to unintended invocation whenever users mention testing-related concepts, which increases the chance of misfires and unreliable agent behavior.
The trigger sentence is phrased as a broad natural-language request ('Help me ...') that could plausibly appear in ordinary conversation, making accidental or overbroad skill activation more likely. In an agent setting, ambiguous invocation boundaries can cause the wrong skill to run, leading to irrelevant actions, prompt-context pollution, or bypass of a more appropriate specialized workflow.
The trigger phrases are extremely broad and include common software terms like 'unit tests', 'testing', and 'regression', making accidental invocation or over-selection of this skill more likely during ordinary user requests. In an agent routing context, this can cause the skill to activate when the user did not explicitly ask for it, potentially biasing responses or crowding out more appropriate skills.
The manifest description contains very broad activation terms such as 'software-and-data', 'testing', and 'analysis' that can match many ordinary developer requests outside the specific unit-test-coverage scope. This can cause the skill to be invoked in unintended contexts, increasing the chance of irrelevant guidance, prompt-surface expansion, or accidental routing over more appropriate skills.
The trigger examples use highly generic sentence forms such as 'Help me...' and 'I need a practical workflow for...' wrapped around long natural-language text, which are collision-prone with normal user requests. Such vague triggers make accidental activation more likely and can cause this skill to intercept unrelated requests that merely resemble common help-seeking phrasing.
The skill description defines usage conditions with a wide scope, covering multiple generic software topics without clear boundaries or exclusion rules. This can lead to unintended activation and misapplication of the skill, especially in agent systems where routing logic may rely heavily on description text.
The trigger keywords are broad and underspecified, including generic terms like testing, regression, and quality, which can cause the skill to be invoked in situations outside its intended scope. Over-broad activation increases the chance of inappropriate routing, irrelevant guidance, or accidental interception of requests better handled by a different skill.
The default prompt is overly broad and grammatically vague, using generic phrases like 'help me' and broad testing-related terms without clear trigger boundaries. In combination with allow_implicit_invocation: true, this can cause the skill to be invoked in contexts where the user did not clearly intend to use it, creating prompt-routing confusion and unnecessary exposure of the skill's instructions.
The trigger sentence is broad enough to match ordinary user phrasing like 'Help me' or 'I need a practical workflow,' which can cause the skill to activate outside its intended scope. Overbroad activation increases the chance that unrelated requests are steered into this testing workflow, leading to prompt-routing errors, confusing outputs, and potential misuse if downstream behavior assumes testing-related context that is not actually present.
In a Chinese-language README, all listed trigger sentences are written in English and presented as the activation method, with no indication that Chinese alternatives are supported. This can amount to a language-policy concern because it implicitly forces English for invocation without offering the user a language choice.
