Security audit
Security checks for vulnerabilities and agentic risk
This is a low-risk workflow-helper skill with overly broad triggers, but it does not include code, persistence, credential use, or hidden data access.
Before installing, be aware that this skill may activate for broad Google or productivity requests. It appears safe as a guidance-only helper, but users who want precise routing may prefer narrowing or disabling implicit invocation.
The skill description is broad enough to match many ordinary productivity or Google-related requests, which can cause the skill to be invoked outside its intended scope. In an agent system, overbroad routing can misapply instructions, create prompt-collision behavior, and steer users into an irrelevant or lower-safety workflow without clear consent.
The keyword list includes very generic terms such as 'google', 'workspace', 'cli', 'gmail', 'calendar', and 'drive', which are common across many unrelated user requests. This makes unintended invocation collisions likely, allowing the skill to hijack routing from more appropriate skills and potentially produce incorrect guidance or unsafe actions in the wrong context.
The trigger list includes very common terms such as "google", "workspace", "cli", and "bug fix", which are likely to match many unrelated requests. This can cause the skill to activate outside its intended scope, leading to confused delegation, unsafe overreach into unrelated tasks, or accidental precedence over more appropriate skills.
The trigger sentence begins with a very generic phrase ('Help me') and then appends a long requirement description, which creates weak activation boundaries and increases the chance the skill is invoked for unrelated requests. In an agent ecosystem, over-broad routing can misdirect user tasks into this skill, causing unintended handling of Gmail/Google Workspace or workflow-related requests and reducing predictability and safety.
The trigger phrases are broad and generic, combining common productivity terms with long natural-language prompts that could cause the skill to activate in situations outside its intended scope. In an agent environment, ambiguous activation boundaries can lead to accidental invocation, inappropriate handling of unrelated tasks, or chaining into higher-risk workflows involving Google Workspace artifacts without clear user intent.
The trigger phrases are broad enough to match common requests about productivity, Google Workspace, CLI, or bug fixing, which can cause the skill to activate outside its intended scope. Over-broad invocation increases the chance of unintended capability routing, user confusion, and accidental use of a workflow that may access or shape sensitive work-productivity tasks without clear user intent.
The example triggers model broad natural-language requests that lack strong boundaries, which can train or encourage routing systems to activate this skill for everyday help requests. While not directly exploitable as code execution, this increases the chance of systematic misrouting and inappropriate application of the skill's instructions.
The activation description says to use the skill for broad categories like work-productivity, gog, google, workspace, and CLI, or whenever a user needs practical workflow or implementation support. That scope is ambiguous and expansive, increasing the chance of accidental invocation on generic productivity or Google-related requests that do not fit this skill's real purpose.
The skill enables implicit invocation without any narrowing conditions, so the agent may auto-select it in loosely related contexts. Because the skill is broadly described and its default prompt steers behavior toward workflow help, this can cause unintended activation, prompt-surface expansion, and execution of guidance in situations the user did not explicitly request.
This trigger sentence is still broadly worded ('I need a practical workflow for...') and does not clearly constrain invocation to specific Gog/Google Workspace use cases. Because the skill is framed for practical workflow help, a vague trigger can cause accidental activation on ordinary productivity requests, leading to incorrect routing and unintended execution paths in higher-privilege agent environments.
This README is labeled as Chinese (README.zh-CN.md), but key descriptive content is written in English, which can impose a language preference on users without opt-in. The file does not explain that mixed-language content is intentional or provide an explicit language choice.
The activation examples in a zh-CN user document are written only in English, which can imply that users must interact in English to use the skill. There is no note offering equivalent Chinese triggers or clarifying that language is flexible.
