Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording but no hidden code, credential handling, persistence, or destructive behavior.
Users should be aware that this skill may activate too broadly for generic Google Workspace, CLI, Gmail, Calendar, Drive, Contacts, or bug-fix requests. Consider narrowing triggers or disabling implicit invocation if precise routing matters.
The skill description is broad enough to match many ordinary productivity, Google, workspace, CLI, or implementation-help requests, which can cause the skill to activate outside its intended niche. Over-broad routing increases the chance of prompt/context hijacking at the orchestration layer, misapplication of the skill to unrelated tasks, and unintended access to user context or downstream tools.
The keyword list contains generic terms like 'google', 'workspace', 'cli', 'gmail', 'calendar', and 'bug fix' that appear in a wide range of benign user requests. Such unconstrained triggers make accidental invocation likely, which can let this skill intercept unrelated conversations and influence behavior or outputs beyond its intended domain.
The trigger sentences are broad, repetitive, and include common terms like 'help me', 'practical workflow', and generic productivity-related wording that can match ordinary user requests outside the intended scope. This can cause accidental invocation of the skill, leading to context hijacking, user confusion, or execution of an irrelevant workflow in place of the user’s actual intent.
The trigger phrases are generic enough to match common user requests about productivity, Google Workspace, CLI, or bug fixing, which can cause the skill to activate outside its intended scope. In an agent ecosystem, over-broad activation increases the chance of unintended tool use, context capture, or workflow execution when the user did not actually request this specific skill.
The trigger phrases are broad enough to match ordinary requests about productivity, Google Workspace, CLI, or bug fixing, which can cause the skill to activate outside its intended scope. In an agent ecosystem, overbroad activation can route unrelated user tasks into this skill, creating unintended prompt capture, workflow interference, and policy bypass opportunities if the skill injects its own process into normal conversations.
The example triggers repeat broad, ambiguous language rather than demonstrating precise activation criteria, which reinforces permissive matching behavior. This makes the skill more likely to be selected for loosely related requests and normalizes activation on vague prompts, increasing routing mistakes and downstream security/control issues.
The invocation guidance is written broadly enough to match generic requests for practical workflows, artifacts, checklists, analysis, or implementation support across common topics. That ambiguity increases the chance this skill is chosen over a more appropriate one, which can degrade task safety and reliability by applying mismatched instructions to unrelated user intents.
The trigger list includes very broad, high-frequency terms like "google", "workspace", and "cli", which can cause the skill to activate in many unrelated conversations. Unintended invocation can route users into the wrong workflow, causing confusing guidance, context leakage into an irrelevant skill path, or accidental execution of productivity-related steps when the user asked for something else.
The default prompt uses broad, everyday language such as 'help me' and general work-productivity phrasing, which can function as a vague trigger surface for invocation or over-selection. In combination with an implicitly invokable skill, this makes the skill easier to activate during ordinary user requests that were not meant for this specific workflow helper.
The skill enables implicit invocation but provides no narrowly defined trigger conditions, which increases the chance the agent will activate this skill in unrelated conversations. Because the skill is framed around broad productivity and workflow help, accidental routing could expose user context to the skill unnecessarily or cause it to influence tasks outside its intended scope.
Although this is labeled as a Chinese README, substantial instructional content and trigger text are written in English, and the file does not state that the user can choose their preferred language. This can amount to imposing a language/locale experience without explicit opt-in.
The file is a zh-CN skill document, but key operational description and examples embed substantial English text while not stating whether responses should be in Chinese, English, or based on user preference. This can create an implicit language policy that is unclear to users and may not respect locale choice.
