Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording, but it does not request sensitive access, persistence, code execution, or hidden data handling.
Installers should be aware that this skill may be selected too broadly because of generic keywords and implicit invocation. It is best used only for explicit Nano Banana Pro-style workflow or skill-authoring requests; maintainers should narrow the triggers and add out-of-scope examples before broad deployment.
The trigger examples are broad and generic enough to match ordinary user requests, which can cause the skill to be invoked when the user did not explicitly intend it. In an agent ecosystem, unintended invocation can lead to incorrect routing, surprise behavior, and accidental application of the skill in contexts where its assumptions do not fit.
The manifest description says to use the skill when a user asks for broad terms like "practical workflow," "artifact," "checklist," "analysis," or "implementation support," which are common across many unrelated tasks. Ambiguous activation guidance at the manifest level increases the chance that orchestration systems or maintainers invoke this skill in contexts far beyond the intended requirement.
The trigger list includes highly generic terms such as "nano," "banana," "pro," "generate," "edit," and "bug fix," which can match many unrelated user requests. This can cause the skill to activate outside its intended scope, leading to inappropriate routing, confusing outputs, or unintended influence over unrelated tasks.
The trigger keywords include extremely broad everyday terms such as “nano”, “banana”, “pro”, “generate”, “edit”, and “images”, which can cause the skill to activate in many unrelated contexts. This creates routing ambiguity and can unintentionally inject this skill’s instructions into conversations where it does not belong, reducing predictability and potentially interfering with safer or more appropriate skills.
The trigger examples include very broad, ordinary language such as requests for 'a practical workflow' combined with common words like 'nano', 'banana', or 'pro', which can cause the skill to activate for unrelated user requests. Unintended invocation can route conversations into an irrelevant workflow, causing incorrect assistance, user confusion, and unsafe overreach if the skill starts making assumptions outside its intended scope.
The invocation guidance does not clearly define the activation boundary, relying on broad example sentences rather than precise criteria. This ambiguity increases the chance that orchestrators or users will invoke the skill in unrelated contexts, reducing reliability and potentially exposing downstream workflows to unintended inputs.
The trigger phrases are broad enough to activate on generic terms like 'nano', 'pro', 'generate', or 'images', which can cause the skill to run in unintended contexts. In an agent environment, ambiguous routing can lead to the wrong workflow handling user data or requests, reducing reliability and potentially bypassing safer or more appropriate skills.
The example triggers are vague, repetitive, and do not define boundaries for when the skill should or should not run. Without negative examples or clear scope constraints, implementers may overfit activation to broad language, increasing accidental invocation and reducing trust in skill selection.
文件名与内容表明该技能为中文（zh-CN）版本，但文中未说明这是可选语言、也未向用户提供语言切换或偏好选择。按规则，未经过用户选择而固定语言/locale 可能构成自然语言政策问题。
The description says to use the skill when users mention broad terms like work-productivity, nano, banana, or pro, or when they need any practical workflow, checklist, analysis, or implementation support for the requirement. These activation conditions are overly expansive and poorly bounded, making the skill eligible for a wide range of unrelated requests and increasing the chance of inappropriate invocation.
The default prompt and description use very broad activation language such as 'use when a user asks for work-productivity' and generic terms like 'practical workflow, artifact, checklist, analysis, or implementation support.' This can cause the skill to be invoked in many unrelated contexts, increasing the chance of prompt hijacking, unintended delegation, or disclosure of context to a skill that was not specifically requested.
Implicit invocation is enabled while the skill metadata does not define precise trigger constraints. In combination with the broad description, this creates a realistic risk that the platform auto-invokes the skill for loosely related prompts, exposing user context or allowing the skill to influence responses without clear user intent.
The activation examples are ambiguous and do not clearly define the boundaries for invocation, making it easy for the skill router or an agent to select this skill for loosely related productivity, image, or bug-fix requests. In this context, the skill is a general workflow helper, so overbroad activation expands its reach and increases the chance of misrouting rather than direct code execution, but that still creates reliability and safety problems.
The Chinese README contains a full English paragraph in the requirements section, which can impose an unintended language experience on users of the zh-CN documentation. Because the file does not indicate that the user can choose language or that English is required for a justified reason, this creates a language/locale policy concern.
Key user-facing sections in this zh-CN README, including the workflow description and all listed trigger phrases, are written in English. This may force a specific language on users reading the Chinese version of the skill documentation without any explicit opt-in or justification.
