Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording but no evidence of hidden code, data access, persistence, or destructive behavior.
Before installing, be aware that this helper may be invoked for generic PDF, edit, or workflow requests because its triggers are broad and implicit invocation is enabled. Prefer explicit use for Nano PDF-style workflow planning, and consider narrowing the trigger keywords if you manage the skill.
The skill description includes very broad activation terms such as 'work-productivity', 'pdf', 'edit', 'analysis', and 'implementation support', which can match many unrelated user requests. Over-broad routing can cause the wrong skill to activate, leading to confused task handling, unsafe context mixing, or unreviewed workflow guidance being applied outside its intended scope.
The trigger list contains generic everyday words including 'nano', 'pdf', 'edit', 'natural', 'language', and 'instructions' without scope constraints. These terms are common across many benign requests, so the skill may be invoked unexpectedly and intercept prompts far outside its intended domain, increasing the chance of incorrect or unsafe task execution.
The trigger sentence begins with a very generic phrase ('Help me ...'), which can match ordinary user requests that are unrelated to this skill. In agent-routing systems, overly broad activation patterns can cause unintended invocation, leading the skill to run in the wrong context and potentially produce irrelevant or unsafe workflow actions.
The trigger examples are generic enough to match ordinary PDF/help-workflow requests, which can cause the skill to activate outside its intended scope. Over-broad activation increases the chance of prompt hijacking at the routing layer, user confusion, or unintended substitution of a more appropriate skill with one that may apply the wrong workflow.
The trigger phrases are broad and include generic terms like 'nano', 'pdf', 'edit', and 'natural language instructions', which can cause the skill to activate for many ordinary user requests unrelated to the intended workflow. Over-broad activation increases the chance of unintended routing, prompt/context hijacking from irrelevant tasks, and user confusion about why this skill was invoked.
The file describes separate English and Chinese documentation artifacts, and this README itself is entirely in Chinese, but it does not state that language selection is optional or user-driven. This can create a locale policy issue because the skill presentation implicitly fixes language based on file variant rather than explicit user opt-in.
The example trigger sentences are vague and largely restate the requirement text rather than showing clear positive and negative activation boundaries. Ambiguous examples make it harder for routing systems and maintainers to understand intended usage, which increases the likelihood of misfires and inappropriate skill selection.
技能描述中的启用条件写法较宽泛，使用“当用户提出...或需要...支持时使用”这类模糊表述，但没有清晰界定什么请求不应启用该技能。边界不清会扩大攻击面和误用概率，使代理更容易在不相关任务中调用该技能，影响结果可靠性和安全控制的一致性。
触发关键词包含非常通用的词，如“nano”“pdf”“edit”“natural”“language”“instructions”，容易在大量无关对话中误触发该技能。误触发会让代理在错误上下文中套用该工作流，导致任务路由偏差、输出不相关内容，甚至把本应由更合适技能处理的请求导向本技能。
The default prompt uses a very broad invocation phrase tied to common terms like work-productivity, nano, and pdf, which can cause the skill to activate in contexts far beyond its intended scope. Combined with allow_implicit_invocation: true, this increases the chance of unintended routing, prompt collisions with other skills, and the skill being invoked on unrelated user requests where it may influence outputs unexpectedly.
The second trigger remains activation-ambiguous because 'I need a practical workflow for ...' is still broad enough to match many unrelated planning requests. This increases the chance of skill misrouting, which is especially risky for a broadly described productivity skill because it may intercept requests outside its intended PDF-specific scope.
该技能文件主体为中文，但触发关键词和示例触发句主要以英文给出，未说明用户是否可使用其他语言或是否必须采用这些英文短语。按照规则，若技能对语言或 locale 施加约束而无用户 opt-in，可能构成自然语言策略问题。
