Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable payload, data access, persistence, or exfiltration behavior, though its trigger metadata is overly broad and may cause unintended activation.
Install only if you want a general SkillScan-style workflow helper. Be aware that its broad trigger words and implicit invocation setting may cause it to appear in loosely related security or workflow conversations, so users or maintainers should narrow the triggers for more predictable routing.
The description says to use the skill when a user asks for broad categories like work-productivity, security, or any practical workflow or checklist, which is too ambiguous to safely scope invocation. In an agent environment, such vague criteria can cause the skill to activate across many unrelated requests and unexpectedly steer outputs in contexts where it was not intended.
The trigger keyword list is excessively broad and includes generic terms such as "security," "every," "must," and "before," which are likely to appear in many unrelated prompts. This creates an overbroad activation surface where the skill may be invoked in unintended contexts, increasing the chance it influences security-sensitive workflows without user intent.
触发关键词包含“security”“every”“must”“pass”“before”等高频日常词汇，范围过宽，容易在与该技能无关的普通对话中被误触发。误触发会把用户请求导向该技能的工作流，造成错误的安全建议、上下文污染或绕过更合适技能的路由。
The skill enables implicit invocation while using a very broad, generic description and default prompt terms such as work-productivity, security, workflow, checklist, and analysis. This can cause the agent to auto-select the skill in many unrelated contexts, expanding the skill’s reach beyond clear user intent and increasing the chance of prompt-scope abuse, unintended data exposure, or unsafe execution paths triggered through normal conversation.
The trigger guidance is broad enough to match common phrases like 'security', 'workflow', 'help me', and 'I need a practical workflow', which can cause the skill to activate outside the author's intended context. In an agent ecosystem, unintended activation can redirect tasks, override more appropriate skills, or introduce unwanted workflow steps into unrelated requests.
The trigger set includes very generic terms such as "security", "gate", "every", "must", "pass", and "before", which are common in ordinary conversation and likely to cause unintended skill activation. In an agent environment, over-broad activation can route unrelated user requests into this skill, causing incorrect workflow execution, context leakage across tasks, or accidental application of security-oriented actions where they were not requested.
The example trigger sentences largely repeat the same broad language from the description and do not add meaningful constraints or disambiguation. Because examples often guide matching behavior and author expectations, these examples reinforce over-activation and make accidental invocation more likely.
技能描述中的启用条件覆盖面很大，如“需要实用流程、产物、检查清单、分析或实现支持时使用”，但未给出明确边界，导致路由器难以区分该技能与其他通用生产力或安全技能的适用范围。这会增加误选技能和返回不相关输出的概率，削弱系统可预测性。
示例触发句只展示了正例，且内容高度概括，没有说明哪些相似表达不应触发该技能。这会让实现方或路由组件过度泛化匹配规则，进一步放大误触发问题。
The trigger phrases are extremely broad and include generic terms like "security," "gate," "every," and "bug fix," which can cause the skill to activate for unrelated requests. In an agent-routing context, this increases the chance of unintended invocation, misrouting, and accidental exposure of workflow behavior in contexts the user did not intend.
The README lists separate English and Chinese instruction files, but does not state how the user's preferred language is selected or whether the user can opt in. This can create ambiguity around locale behavior and may conflict with language-choice policy expectations.
The file is labeled as a Simplified Chinese README, but substantial user-facing content and all trigger phrases remain in English. This can constitute a language/locale policy issue because the skill appears to force or assume English phrasing for activation without explicit user opt-in or justification.
The file mixes English content with Chinese titles and evidence references, but does not explain whether the skill operates in multiple languages or whether users can choose their preferred language. This can create an implicit language/locale assumption without documented opt-in or justification.
