Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation wording but no code, persistence, credential handling, or hidden data access.
This skill is reasonable to install if you want a general workflow helper, but expect broad automatic activation around Google, Workspace, CLI, Gmail, Calendar, Drive, Contacts, or bug-fix wording. Review outputs before applying any generated workflow or code, especially when working with business or account data.
The manifest description uses very broad activation language such as generic domains and common task types ('workflow', 'checklist', 'analysis', 'implementation support'). This can cause the skill to be invoked for many loosely related requests, creating skill-routing confusion and increasing the chance that a user is steered into unintended guidance or lower-quality handling.
The trigger list includes highly common terms like 'google', 'workspace', 'cli', 'gmail', 'calendar', and 'bug fix' without qualifiers. These broad keywords can match unrelated everyday requests, leading to over-triggering and possible interception of prompts better handled by other skills or default behavior.
The trigger section uses broad, common terms like "google," "workspace," "cli," and generic help phrases that can cause the skill to activate for many unrelated user requests. Over-broad activation boundaries are dangerous because they increase unintended invocation, causing the agent to apply this workflow in the wrong context and potentially override a more appropriate or safer skill.
The trigger phrases are generic enough to match ordinary productivity or help-seeking requests, which can cause the skill to activate outside the user's clear intent. In an agent ecosystem, over-broad invocation increases the chance of unintended access to workflows involving Google Workspace-related actions or guidance, creating confusion, mis-execution, or unsafe automation paths.
The trigger phrases are broad, generic, and closely resemble normal user productivity requests, which can cause the skill to activate in situations far beyond its intended scope. In an agent ecosystem, this increases the chance of unintended invocation, prompt hijacking of ordinary tasks, or accidental routing of sensitive workflow requests into a skill the user did not explicitly choose.
The example trigger phrases are written in generic, natural language and repeat broad requirement text rather than showing clear activation boundaries. This makes the skill easier to invoke accidentally and normalizes ambiguous routing behavior, which can reduce reliability and cause users to receive irrelevant workflow guidance.
技能描述将适用范围表述为广泛的“work-productivity, gog, google, workspace, cli”及“practical workflow/artifact/checklist/analysis/implementation support”，边界不清，容易在任务编排时被解释为通用生产力技能。这样的模糊性会放大误路由风险，使代理在缺乏最小适配校验的情况下调用该技能，降低结果可靠性并可能把敏感工作上下文暴露给不必要的处理路径。
触发关键词包含非常通用的词，如“google”“workspace”“cli”“bug fix”，会让技能在大量无关场景中被误触发。若代理依赖这些触发条件自动选用技能，可能把本技能错误地插入与其目标不匹配的任务流中，导致错误建议、越权处理工作流上下文，或掩盖更合适的专用技能。
The default prompt uses broad, everyday terms like work-productivity, google, workspace, cli, workflow, checklist, and analysis without meaningful scoping, which increases the chance of unintended or implicit activation outside the user's actual intent. Because implicit invocation is enabled, this can cause the skill to trigger in many normal conversations and inject its instructions or workflow behavior into unrelated tasks, creating prompt-surface expansion and possible prompt hijacking opportunities.
The README explicitly labels separate English and Chinese instruction files, but does not state how the language is selected or that the user may choose their preferred language. Because locale/language behavior is documented without an opt-in mechanism or justification, this may conflict with language-choice policy expectations.
The file explicitly presents itself as a Chinese-language user guide ("README.zh-CN.md") but does not indicate that language selection is optional or that users can choose another locale. Under the policy, language or locale constraints should either be user-selectable or clearly justified.
该技能文档整体以中文呈现，且文件名为 SKILL.zh-CN.md，但正文未说明这是可选语言版本，也未向用户提供语言/地区选择。根据规则，若技能对语言/locale 有固定约束而无用户 opt-in，可能构成自然语言政策问题。
