Security audit
Security checks for vulnerabilities and agentic risk
This is a low-impact guidance skill with overly broad activation wording, but it contains no code, credential access, persistence, or destructive behavior.
This appears safe to install as a lightweight workflow helper. Be aware that its current triggers are broad, so users or maintainers should prefer explicit invocation and consider narrowing the routing language if they want fewer accidental activations.
The skill description is broad enough to match many ordinary productivity, Google, CLI, and workflow-related requests, which can cause the skill to activate outside its intended scope. Overbroad routing increases the chance of prompt hijacking of unrelated tasks, unintended instruction precedence, and unsafe tool usage in contexts the skill was not designed to handle.
The keyword list contains vague, high-frequency terms such as "google," "workspace," "cli," "gmail," and "bug fix" that are likely to collide with many unrelated user prompts. In an agent environment, such high-collision triggers can misroute sessions to this skill, causing unintended behavior, policy bypass opportunities, or interference with more appropriate specialized skills.
The trigger keywords include very generic terms such as "google", "workspace", "cli", and common productivity app names, which can cause this skill to activate for many unrelated user requests. Over-broad activation increases the chance of incorrect skill selection, unintended context capture, and response interference from a skill whose scope is only loosely related to the user’s intent.
The trigger sentence is overly broad and resembles ordinary user language, which can cause the skill to activate in situations far outside its intended scope. In an agent environment, this increases the chance of unintended invocation, context capture, and inappropriate workflow execution for unrelated Google/work-productivity requests.
The trigger phrases are broad and generic enough that the skill may activate for loosely related requests involving productivity, Google, workspace, or bug-fixing. In an agent ecosystem, ambiguous activation can cause unintended routing, leading the skill to handle requests outside its intended scope and potentially influence workflows or outputs when a more appropriate skill should have been selected.
The documented trigger phrases are broad enough to match common productivity or Google Workspace requests, which can cause unintended skill activation. In an agent ecosystem, this creates routing confusion and may pull users into this skill when they intended a different workflow, reducing predictability and potentially exposing unrelated task context to the wrong skill logic.
The example trigger sentences model activation on ambiguous, catch-all requests instead of demonstrating clear boundaries for intended use. This reinforces loose matching behavior and makes it easier for normal user phrasing to invoke the skill unintentionally, increasing confusion and the risk of inappropriate instruction application.
The description defines invocation using both broad category labels and an expansive natural-language scope, making it unclear when the skill should actually be used. This ambiguity can lead orchestrators or users to apply the skill in contexts far outside its intended purpose, reducing predictability and potentially exposing user tasks to the wrong workflow logic.
The example trigger phrases are written as ordinary help requests that resemble everyday user language, so they do not safely distinguish this skill from many unrelated support queries. If used by a routing layer or copied by authors, these examples encourage accidental activation and make the skill easier to invoke in unintended contexts.
The default invocation phrase is broad and generic enough to overlap with ordinary user requests about productivity, workflows, or implementation help, which can cause unintended skill activation. Because implicit invocation is enabled, this increases the chance the agent routes unrelated conversations into this skill, potentially expanding its influence beyond user intent and creating prompt-routing or policy-bypass opportunities.
The invocation examples do not define when the skill should and should not trigger, so routing logic may select it for ambiguous requests. This can lead to misfires, user confusion, and accidental handling of prompts that belong to other skills, especially because the skill is framed around broad work-productivity and Google-related tasks.
This file is presented as a Chinese README, but key user-facing content including the requirement description, workflow description, and trigger phrases is written in English. That creates an implicit language constraint for users without indicating opt-in or offering an alternative, which can violate language/locale policy expectations.
