Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording, but it does not include executable code, credential handling, persistence, exfiltration, or destructive behavior.
Before installing, be aware that this skill may be invoked for broad web/search/API/productivity requests because of its generic keywords. It is best used when you specifically want Tavily-style web research workflow planning or related reliability guidance.
The skill description is broad enough to match many ordinary requests involving work, web, search, or implementation help, which increases the chance of accidental or unnecessary invocation. Over-broad activation can route unrelated user tasks into this skill, causing inappropriate behavior, user confusion, and unsafe automation choices when the skill's assumptions do not fit the request.
The keyword list includes very generic terms like 'web', 'search', 'via', 'api', and 'user', which are common across unrelated conversations and make unintended triggering highly likely. This can cause the skill to intercept broad categories of requests, leading to misrouting, reduced reliability, and potential execution of workflow guidance in contexts where it was not intended.
The trigger sentence is overly broad and can cause the skill to activate for generic requests that merely resemble the requirement text, rather than for clearly scoped Tavily/web-workflow tasks. In an agent environment, ambiguous activation increases the chance of unintended tool routing, irrelevant execution, and accidental exposure of user context to a skill that was not specifically requested.
Although phrased as a task request, the trigger remains ambiguous because it matches common language like 'I need a practical workflow' without sufficient domain constraints. This can cause over-triggering across unrelated productivity requests, making the skill context more dangerous because it is designed to produce workflows and implementation support that may steer agent behavior even when the fit is poor.
The trigger sentence is written as a generic help request rather than a narrowly scoped activation phrase, which increases the chance the skill is invoked during ordinary conversation. In an agent ecosystem, unintended invocation can cause the wrong workflow to run, leading to misrouting, irrelevant actions, or accidental use of external search/API functionality without clear user intent.
This trigger remains ambiguous because it embeds common conversational wording and lacks a precise condition for activation. In this skill context, which is tied to web/search workflows, accidental activation is more concerning because it could trigger external-search-oriented behavior or workflow generation when the user was only speaking generally.
The trigger phrases are broad and generic enough that this skill may activate for loosely related requests involving work productivity, web search, APIs, or bug fixing. In an agent environment, over-broad activation can cause the wrong skill to be invoked, leading to unintended external-search workflows, confusing task routing, or bypass of more appropriate specialized controls.
The example triggers are phrased in broad, natural language and largely repeat generic request forms like 'Help me' and 'I need a practical workflow,' without defining what specifically qualifies for this skill. That weak boundary makes it easier for trigger matching systems or authors to overgeneralize activation, increasing accidental invocation and inconsistent behavior.
技能描述说明了适用主题，但没有清晰给出“不该触发”的条件，导致路由器或代理在相似但不相关的请求上也可能调用该技能。边界不清会放大误触发问题，尤其在多技能环境中会降低结果可靠性并造成错误自动化。
触发关键词包含 web、search、user、api 等高度通用词，会与大量普通请求重叠，增加技能被非预期自动激活的概率。误触发后，代理可能套用与用户目标无关的工作流，带来错误建议、上下文污染或绕过更合适技能的路由。
The default prompt invokes the skill through a very broad phrase ('Use $work-productivity-tavily-web-workflow-helper to help me...') tied to loosely related concepts, which increases the chance of accidental or overly eager activation. In a web/search workflow helper, broad invocation can cause the agent to engage in unintended contexts, expanding access to external-search behaviors or workflow actions without sufficiently clear user intent.
The document is a Chinese-language README variant and explicitly labels itself as the Chinese user guide, but it does not indicate that users may choose another language or locale. This can be a natural-language policy concern when a skill experience is presented in a fixed locale without opt-in or documented justification.
示例触发句使用了 Help me、I need a practical workflow 这类常见表达，并与较长的需求描述混合，容易让作者或路由器形成过宽匹配模式。虽然单独危害较前两项低，但它会进一步强化误触发倾向，影响技能选择准确性。
The description and default prompt include Chinese text ('搜索-style') within otherwise English strings, but the file does not indicate that users can choose their language or that this locale mixing is required for a region-specific purpose. This can violate language/locale policy expectations when users are not given an explicit choice.
