Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad auto-invocation wording, but no hidden execution, credential use, persistence, or destructive behavior.
Installers should be aware this skill may be selected too broadly because it allows implicit invocation and uses generic terms like browser, automation, and cli. It is otherwise a low-risk workflow helper with no executable code or credential handling; tighten triggers if precise routing matters.
The invocation description says to use the skill when a user asks about broad categories like work-productivity, browser, headless, or automation, without clear boundaries. This ambiguous scope can cause inappropriate activation on benign or unrelated tasks, expanding the skill's effective authority and making prompt-routing behavior less predictable and easier to abuse.
The trigger list includes very generic terms such as "browser", "automation", and "cli", which can cause the skill to activate for many unrelated requests. Over-broad activation increases the chance that this skill intercepts prompts outside its intended scope, leading to misrouting, unintended instruction injection into workflows, or suppression of more appropriate skills.
Enabling implicit invocation without strong activation constraints allows the skill to auto-trigger based on loosely related user requests. Because this skill is positioned around browser-style workflows and automation support, unintended invocation could lead to unsafe guidance, unexpected workflow execution paths, or user confusion about which agent behavior is active.
The trigger phrases are broad, repetitive, and partially match generic requests about browser workflows, automation, or practical help, which can cause unintended activation of this skill. In an agent ecosystem, accidental invocation can route user requests into the wrong workflow, leading to irrelevant actions, unsafe browser automation guidance, or bypass of more appropriate specialized skills.
The trigger phrases are broad and loosely scoped to common terms like browser, automation, and workflow, which can cause unintended activation for unrelated user requests. In an agent skill ecosystem, this increases the chance that the wrong skill handles a task, leading to unsafe automation, irrelevant actions, or accidental execution of browser-oriented workflows where they were not explicitly requested.
The file identifies separate English and Chinese documentation variants, but this specific README is a Chinese-language file and does not indicate any user choice or opt-in for language selection. That can violate a language/locale policy when a skill experience is presented in a fixed language without explicit user preference handling.
The description defines activation in extremely broad terms, stating the skill should be used whenever a user asks about work-productivity, browser, headless, automation, or implementation support. Because these categories are common and underspecified, the skill may be invoked for tasks well beyond its validated scope, increasing the risk of incorrect guidance, privilege overreach, or interference with better-matched skills.
The trigger list is overly broad for a generic productivity/browser skill, including common terms like "browser," "automation," and "cli" that can match many unrelated user requests. This can cause the skill to activate outside its intended scope, leading to inappropriate instruction injection, workflow hijacking, or unsafe reliance on a mismatched skill in sensitive contexts.
The default prompt is broad and generic, using open-ended help phrasing tied to common topics like productivity, browser workflows, and automation. In combination with tool-routing or automatic skill selection, this can cause the skill to be invoked in situations the user did not explicitly intend, increasing the chance of overbroad actions or exposure to workflow-specific behavior.
The invocation examples are ambiguous, truncated, and syntactically inconsistent, which makes the skill's activation boundary unclear. This increases the chance of accidental triggering or misrouting by an orchestrator, especially because the skill targets browser/headless automation workflows that can be higher impact than passive text-generation tasks.
The trigger sentence is broad and malformed enough that it could match ordinary user phrasing unrelated to the intended skill scope. In an agent routing context, this can cause unintended invocation of the skill, leading to confused delegation, unreliable behavior, and possible execution of browser-oriented workflows in contexts where they were not requested.
