Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent shift-left QA guidance skill with no hidden execution, persistence, credential use, or destructive behavior; the main caution is an optional unpinned companion-skill install command.
Before installing the companion workflow, confirm you trust the Kokxi/qa-test-skills source and prefer a pinned version or reviewed commit. Expect this skill to provide Chinese-language QA planning checklists and early testing review guidance; it does not itself show hidden code execution or persistence.
The description includes broad trigger phrases such as '左移', '提前测试', and general references to early project involvement. Overly broad activation language can cause unintended invocation in unrelated conversations, leading the agent to apply the wrong workflow, over-collect context, or override a more appropriate skill.
The primary user-facing description and operational guidance are written entirely in Chinese, while also embedding an English usage sentence, but there is no statement that users may choose their preferred language. This can amount to an implicit locale policy violation because the skill appears to assume a fixed language without opt-in or justification.
The when-to-use field repeats ambiguous activation conditions and lacks clear scope boundaries, which increases the chance of misrouting the agent into this skill. In multi-skill environments, such ambiguity can degrade reliability and produce inappropriate guidance when user intent only partially overlaps.
The skill recommends installing a package via npx skills add Kokxi/qa-test-skills without pinning a specific version or immutable source. This creates a supply-chain risk: a later malicious or compromised package release could be fetched and executed unexpectedly by users following the instructions.
The document title and core instructions are entirely in Chinese, and nothing in the file offers an alternate language or states that the skill is intentionally limited to a Chinese-speaking audience. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.
