Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation terms, but it does not request sensitive access, persistence, credentials, or automatic execution.
Before installing, be aware that this skill may be invoked for ordinary PDF or editing prompts because its triggers are broad. It appears safe from a security standpoint, but users should confirm it is the intended workflow helper when working on unrelated PDF tasks.
The skill description is written broadly enough that common user requests about 'work-productivity', 'pdf', 'edit', or general workflow help could invoke this skill outside its intended niche. Over-broad activation increases the chance the agent applies the wrong workflow, causing misrouting, confusing outputs, or unsafe handling of unrelated documents and tasks.
The trigger list includes generic everyday terms such as 'nano', 'pdf', 'edit', 'natural', 'language', and 'instructions', which are likely to appear in many unrelated prompts. In an agentic environment, such broad keywords can cause frequent unintended invocation, tool misselection, and context confusion, especially when the skill can influence document-processing workflows.
The trigger keywords include very broad terms such as "nano", "pdf", "edit", "natural", and "language", which can match many unrelated user requests. This can cause accidental skill activation, leading the agent to apply the wrong workflow, override more appropriate skills, or expose users to irrelevant automation paths.
The trigger language is broad enough to match ordinary user phrasing such as generic requests about PDFs, editing, or practical workflows, which can cause the skill to activate outside its intended niche. Over-broad activation increases the chance of prompt routing errors, irrelevant instruction injection into unrelated tasks, and unintended execution paths driven by adversarial or ambiguous phrasing.
The trigger phrases are generic help-seeking sentences that can match ordinary user requests unrelated to this skill, which increases the chance of accidental invocation or over-broad routing. In an agent ecosystem, this can cause the wrong skill to activate, leading to unintended handling of user data or actions under an unrelated workflow.
The trigger phrases are broad enough to activate on generic terms like 'nano', 'pdf', 'edit', or 'workflow', which can cause the skill to run in contexts the user did not intend. In an agent ecosystem, overbroad activation can route unrelated requests into this skill, increasing the chance of incorrect actions, unsafe workflow advice, or prompt-surface abuse through unintended invocation.
文件名和正文均表明这是一个 zh-CN 中文版本技能，但内容未说明是否允许根据用户偏好切换语言，也没有提供语言选择或显式 opt-in。根据规则，若技能强制特定语言而无用户选择或明确正当理由，属于自然语言政策风险。
The description says to use the skill when users ask for broad topics like work-productivity, nano, pdf, or edit, but it does not clearly define where this skill should not apply. Ambiguous activation boundaries increase the chance of overbroad routing and incorrect assistance in contexts unrelated to the intended Nano Pdf-style workflow use case.
The default prompt contains broad, everyday activation terms and automatically routes users to this skill without meaningful constraints. This can cause unintended invocation during unrelated conversations, exposing users to incorrect workflow actions, prompt hijacking opportunities, or unexpected behavior from a skill that was not explicitly requested.
The invocation guidance provides weak scope boundaries and encourages matching on common terms and templated sentences without defining exclusion conditions. In an agent environment, that makes accidental or attacker-induced invocation easier, which can misroute tasks, override more appropriate skills, or expose downstream workflow behavior to untrusted inputs in contexts the author did not intend.
The README states that SKILL.md and README.md are English, while separate zh-CN files are Chinese, but it does not explain how the user's language preference is determined or that language selection is optional. This can imply a fixed locale behavior without explicit user opt-in, which may conflict with language-choice policy expectations.
The file is presented as a Chinese README, but key descriptive content is written in English, which can impose a language/locale experience inconsistent with the file's declared audience. There is no indication that users can choose language preference or that the mixed-language format is intentional and justified.
