Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation/workflow helper with overly broad auto-invocation wording, but it does not contain code, credential handling, persistence, or hidden high-impact behavior.
Installing this skill should be low risk, but users should be aware it may activate too readily for generic PDF, editing, or productivity requests. Prefer explicit invocation when you want this workflow helper, and maintainers should narrow the trigger terms and implicit invocation policy before broad deployment.
The description says to use the skill when a user asks for broad categories like "work-productivity," "pdf," "edit," or any "practical workflow, artifact, checklist, analysis, or implementation support." That invocation condition is so expansive that it can match many normal conversations, increasing the chance of accidental invocation and reducing operator control over when the skill should run.
The trigger keyword list includes extremely broad terms such as "nano," "pdf," "edit," "natural," and "language," which are common in unrelated user requests. This can cause the skill to activate outside its intended scope, leading to misrouting, unpredictable behavior, or unintended disclosure of irrelevant workflow guidance into unrelated 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 for many unrelated user requests. In an agent environment, this increases the chance of incorrect routing, unintended instruction injection into unrelated tasks, and user confusion about why this workflow is being applied.
The default prompt contains very broad activation language tied to common productivity terms, making it likely the skill could be invoked in ordinary user conversations unrelated to the intended workflow. This can cause unintentional routing, prompt injection surface expansion, and unexpected access to the skill's instructions or behaviors when users did not explicitly request it.
Enabling implicit invocation without tight trigger constraints allows the platform to auto-select this skill based on vague user wording, increasing the chance of accidental activation. In a broadly described productivity skill, this creates a larger unintended execution surface and raises the risk of misrouting, unauthorized tool behavior, or abuse through crafted prompts that opportunistically trigger the skill.
The trigger sentences are broad natural-language phrases such as 'Help me...' and 'I need a practical workflow...', which can match ordinary user requests unrelated to this specific skill. That increases the chance of unintended activation, causing the agent to route tasks into this skill when the user did not explicitly request it, potentially leading to incorrect behavior or unsafe workflow execution.
The trigger phrases are broad enough to match ordinary requests containing generic terms like 'work-productivity', 'nano', 'pdf', 'edit', and 'natural language', which can cause the skill to activate outside its intended scope. In an agent ecosystem, this can lead to inappropriate routing, accidental execution of irrelevant instructions, and increased exposure to prompt-surface abuse from unrelated user inputs.
The example trigger sentences are generic and self-referential, reinforcing that the skill may be invoked on vague language rather than clear, bounded intent. In agent ecosystems, such examples can effectively become operational patterns, increasing false activations and confusing routing logic.
The description defines invocation using a mix of broad topic tags and a vague condition ('when the user needs practical workflow/artifact/checklist/analysis/implementation support around this demand'), leaving too much room for subjective matching. This ambiguity can cause the skill to be selected in contexts it was not designed for, which is especially risky because the skill is framed as a general helper for workflows and implementation support.
The activation guidance does not clearly define scope boundaries or non-matching cases, so an orchestrator or upstream agent may invoke the skill for loosely related 'work-productivity' or 'pdf' requests. In a multi-skill environment this ambiguity increases the risk of misrouting, unintended prompt capture, and the skill influencing tasks it was not designed to handle.
The trigger sentences are written in broad, natural language that can match many ordinary user requests, increasing the chance the skill activates outside its intended scope. Overbroad activation can route unrelated tasks into this skill, causing prompt/skill confusion, inappropriate handling of user data, or unsafe delegation based on a weak keyword match.
The file descriptions specify English and Chinese variants as fixed artifacts, but the README does not indicate whether users can choose their preferred language or how language selection occurs. This can create a locale-policy concern if the skill defaults to a language without explicit user opt-in.
A Chinese-language README contains core descriptive content in English, and the trigger examples are also English-led, which can impose a language preference without explicit user opt-in. The file does not indicate that users may choose their preferred language or that English is required for a justified regional reason.
