Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording but no hidden execution, data access, persistence, or destructive behavior.
Installers should be aware that this skill may auto-activate too broadly for ordinary PDF, editing, or workflow requests. Prefer explicit invocation or narrow its triggers if precise routing matters.
The description says to use the skill when a user asks for broad concepts like 'practical workflow, artifact, checklist, analysis, or implementation support,' which are common across many benign requests. This ambiguous invocation condition can make the skill match far outside its intended scope, leading to accidental activation and incorrect handling of user tasks.
The trigger list is overly broad and includes generic terms like 'nano', 'pdf', 'edit', 'natural', and 'language', which can cause the skill to activate in many unrelated conversations. This increases the chance of unintended routing, prompt-surface expansion, and interference with more appropriate skills, especially in environments where skills are auto-selected from user text.
The trigger list includes very broad generic terms such as 'nano', 'pdf', 'edit', 'natural', 'language', and 'instructions', which can cause the skill to activate for many unrelated user requests. Over-broad activation increases the chance that this skill intercepts prompts outside its intended scope, leading to incorrect routing, unintended disclosure of internal workflow behavior, or reduced reliability of higher-priority safety or task-specific skills.
Enabling implicit invocation without strict activation constraints allows the platform to auto-route ordinary user requests into this skill based on vague relevance. In combination with the broad productivity and PDF-related scope, this increases the risk of unintended invocation, privilege overreach, and exposure to adversarial prompt content from contexts where the skill was not deliberately selected.
The trigger sentence begins with a generic everyday phrase ('Help me') combined with broad topic terms, which can cause the skill to activate during normal conversation even when the user did not intend to invoke this specific skill. In an agent ecosystem, overbroad activation increases the chance of misrouting prompts, unintended tool usage, and bypass of more appropriate skills or safety checks.
The trigger sentence is written as a very broad natural-language request fragment ('Help me ...') that overlaps with ordinary user phrasing. In agent ecosystems that auto-select skills from trigger text, this can cause unintended invocation, misrouting user requests into this skill, and accidental execution of workflows the user did not explicitly choose.
This trigger is ambiguous about when the skill should activate and describes a general user need rather than a precise scope boundary. Because the skill targets common productivity and PDF-editing tasks, the broad activation language increases the chance of accidental matching across unrelated requests, leading to unwanted skill selection and potentially unsafe or confusing task handling.
The trigger phrases are overly broad and include generic terms like 'nano', 'pdf', 'edit', 'natural', and 'language', which can cause the skill to activate for ordinary user requests unrelated to this workflow. In an agent ecosystem, overbroad activation can misroute tasks, cause unintended handling of user data or documents, and increase the chance that this skill is invoked in contexts it was not designed to handle safely.
The example triggers repeat broad marketing-style language but do not define boundaries for when the skill should not be used. Without negative examples or precise thresholds, downstream routing systems or maintainers may overgeneralize the trigger logic and invoke the skill on loosely related prompts.
The description says to use the skill when users ask for broad categories like work-productivity, nano, pdf, edit, or any practical workflow, checklist, analysis, or implementation support related to the requirement, but it does not define clear activation boundaries. This ambiguity can cause accidental invocation in unrelated contexts, which is primarily a routing and reliability problem but can also weaken least-privilege behavior by exposing the skill in more conversations than necessary.
The default prompt uses generic terms like "help me" and broad workflow language that can overlap with normal user phrasing, increasing the chance this skill is invoked when the user did not specifically intend to use it. Because the skill is framed as broadly applicable to productivity, workflows, analysis, and implementation support, accidental routing is more likely and could cause prompt injection surface expansion or unintended behavior.
The second trigger sentence ('I need a practical workflow for ...') is also phrased broadly enough to match many benign requests about workflows, productivity, or PDFs without clear user intent to call this skill. While less generic than 'Help me,' it still risks accidental invocation and context hijacking in multi-skill environments.
This file is a Chinese-language README variant and does not indicate that users may choose another language or locale, nor does it justify a locale restriction as region-specific. Under the policy, forcing a specific language without opt-in can be a natural-language policy violation.
