Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-style workflow helper with overly broad activation wording, but it does not contain code, hidden execution, credential handling, persistence, or destructive behavior.
Installers should be aware that this skill may activate for broad productivity or Google-related requests. It appears safe as a planning/documentation helper, but maintainers should narrow the triggers if they want to avoid accidental routing.
The trigger keywords are extremely broad and include generic terms like workspace, cli, gmail, calendar, and bug fix, which can cause the skill to activate for many unrelated requests. In an agent environment, overbroad routing can misapply this skill’s workflow guidance to tasks outside its intended scope, leading to incorrect handling, user confusion, or unsafe automation choices.
The trigger sentences are written as broad natural-language prompts that could match ordinary user requests about workflows, setup, or productivity without an explicit invocation boundary. In an agent skill-routing system, this can cause unintended activation or prompt shadowing, where this skill is selected over a more appropriate one and influences handling of unrelated requests.
The trigger sentences are broad, repetitive, and match generic productivity/help-seeking language, which increases the chance that the skill is invoked unintentionally for loosely related requests. In an agent environment, unintended invocation can route user tasks into the wrong workflow, causing inappropriate access patterns, confusing actions around Google Workspace-related operations, or unsafe automation being suggested when the user did not explicitly request this skill.
The trigger phrases are broad enough to match common requests like generic workflow help, bug fixing, or productivity assistance, which can cause the skill to activate outside its intended scope. In an agent ecosystem, this increases the chance of inappropriate routing, unintended tool usage, or the skill influencing tasks that were not meant to invoke Google/Gog-style workflow behavior.
The skill description is broad enough to match many ordinary productivity or Google Workspace requests, which can cause the skill to activate outside its intended niche. Over-broad routing increases the chance of inappropriate tool use, prompt interference, or lower-quality outputs because the agent may select this skill when a more specific or safer skill should handle the task.
The keyword list includes very generic terms like 'google', 'workspace', 'cli', 'gmail', 'calendar', and 'drive' without context constraints. In a skill-routing system, this can lead to unintended invocation across a wide range of unrelated requests, creating misrouting risk and potentially exposing the user to irrelevant or unsafe workflow guidance.
该文件整体以简体中文（zh-CN）呈现，路径名也指向特定 locale，但内容中没有说明这是用户可选语言，也没有声明仅面向特定地区或合规场景。根据规则，若技能强制特定语言而未提供选择或正当理由，属于自然语言政策风险。
The enablement condition in the description is vague, stating the skill should be used for broad categories like work-productivity, google, workspace, cli, or any practical workflow/checklist/analysis support. This ambiguity increases the chance that the skill is selected for tasks far beyond its intended purpose, which can degrade security controls, produce irrelevant actions, or interfere with safer specialized skills.
The default prompt trigger phrase is broad and tied to common productivity and workflow language, which increases the chance of unintended or implicit invocation during ordinary user requests. Because implicit invocation is enabled, this can cause the skill to activate when users did not specifically intend it, expanding the skill’s influence over conversations and increasing prompt-injection and misrouting risk.
The keyword list includes very common terms such as 'google,' 'workspace,' 'cli,' 'gmail,' 'calendar,' and 'drive,' which are likely to appear in many benign requests across unrelated domains. This broad matching surface increases the risk of accidental invocation, misrouting, and over-collection of user context by a skill that was not intentionally selected.
This file is presented as the Chinese user README, but key instructional content is written in English, which effectively forces a language choice without user opt-in. That can violate language/locale expectations for a locale-specific document.
The example trigger phrases model activation on vague, repeated requirement text rather than crisp boundaries, reinforcing ambiguous routing behavior. While less severe than the description and keyword list, these examples encourage broad matching and make it harder for maintainers or routers to distinguish intended use from ordinary requests.
