Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation wording, but it does not contain code, credential access, persistence, exfiltration, or destructive behavior.
This skill appears safe to install as a low-privilege workflow helper. Users should be aware that it may activate more often than intended because its trigger words are broad; tightening triggers or disabling implicit invocation would improve predictability.
The skill description and usage guidance use very broad, common terms such as 'work-productivity', 'task', 'analysis', and 'implementation support' as activation cues. In an agentic environment, this can cause the skill to trigger for many unrelated requests, unnecessarily injecting instructions and expanding the skill's influence beyond its intended scope.
The keyword list includes ambiguous high-frequency words like 'proactive', 'transform', 'task', 'needs', and 'partners', which are likely to collide with ordinary user prompts. This makes accidental invocation highly probable, and in multi-skill systems can lead to prompt hijacking, instruction interference, or suppression of more appropriate skills.
The trigger list includes broad everyday terms such as “task”, “needs”, and “partners”, which can cause this skill to activate in many unrelated conversations. In an agent environment, accidental activation can misroute user requests, override more appropriate skills, and lead to unintended actions or confusing guidance.
The trigger examples are broad, natural-language phrases that overlap with ordinary user requests, making accidental activation likely. In an agent skill system, overbroad invocation can route unrelated requests into this workflow, causing unintended behavior, prompt/context hijacking opportunities, or interference with more appropriate skills.
The trigger sentences are broad, generic, and include common workflow terms like "help me," "practical workflow," and "handle," which can cause the skill to activate in situations the user did not clearly intend. In an agent ecosystem, ambiguous activation increases the risk of prompt hijacking, accidental delegation, or inappropriate use of this skill's instructions in unrelated contexts.
The trigger phrases are broad enough to match ordinary productivity requests such as asking for help, workflows, tasks, or bug fixes, which can cause the skill to activate outside its intended scope. In an agent ecosystem, overbroad activation can misroute user requests, invoke the wrong workflow, and increase the chance that a higher-privilege or more autonomous skill handles unrelated tasks.
该技能文件整体以简体中文（zh-CN）呈现，但内容中未说明这是可选语言版本、未提供用户语言选择，也没有给出必须限定为该地区/语言的业务理由。按规则，强制特定语言而无用户 opt-in 可能构成语言/区域政策问题。
The usage description is very broad and lacks clear inclusion and exclusion boundaries, so the skill may be selected for loosely related workflow, productivity, or task-oriented requests. This increases the chance of inappropriate routing and weakens predictability and safety in multi-skill agent systems.
The sample trigger phrases are generic and do not show contrasting non-trigger cases, so they fail to teach a safe activation boundary. This can reinforce over-triggering behavior and make accidental invocation more likely during natural-language matching.
The manifest description and default prompt are broad, vague, and framed to match many generic productivity requests, which can cause the skill to be invoked outside a narrowly intended use case. In combination with proactive-agent wording, this increases the chance of unintended routing, overreach into unrelated tasks, or user surprise about why the skill activated.
Enabling implicit invocation without activation criteria allows the platform to auto-select this skill for a wide range of loosely related requests. Because the skill is positioned as proactive workflow help, it may intercept user tasks unexpectedly, leading to unintended actions, confusing behavior, or expansion of the skill's effective authority beyond what the user explicitly requested.
The activation section does not clearly define when the skill should and should not trigger, leaving scope ambiguous. This increases the chance that the skill activates on loosely related productivity requests, which can misroute execution and expand the attack surface for prompt-trigger abuse in multi-skill environments.
The file describes separate English and Chinese documentation variants, but it does not state how the user's language preference is selected or whether language choice is optional. This can create a locale-handling policy concern if the skill defaults users into a language without explicit opt-in.
This Chinese README presents core descriptive content and user-facing usage text in English, which can impose a language expectation inconsistent with the file's locale labeling. There is no note that English is intentional, optional, or user-selectable.
