Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation wording, but it does not contain hidden execution, credential access, persistence, or data exfiltration behavior.
This skill appears safe to install for Agent Browser workflow guidance. Users or maintainers should consider narrowing its trigger phrases or disabling implicit invocation so ordinary browser, automation, CLI, or bug-fix requests are not routed here unintentionally.
The trigger sentence begins with a very broad everyday phrase ('Help me'), which can cause the skill to activate on many unrelated user requests. In an agent/browser automation context, overbroad activation is riskier because it may steer the system into offering automation or workflow guidance when the user did not actually intend to invoke this skill.
The trigger phrases are broad and map to common terms like 'browser', 'automation', and 'practical workflow', increasing the chance this skill activates for unrelated user requests. In an agent ecosystem, overbroad routing can cause the wrong skill to take control of a conversation, leading to unintended actions, misleading outputs, or unsafe workflow suggestions in contexts the skill was not designed for.
The activation guidance provides keywords and example prompts but does not define boundaries, prerequisites, or disallowed use cases. That ambiguity can cause accidental routing of generic browser, automation, or productivity requests into this skill, which is riskier because the skill targets browser-style workflows that may influence automation behavior and operational decisions.
The trigger phrase is broad enough to match ordinary user requests for help, which can cause the skill to activate outside its intended context. In an agent/browser automation setting, over-broad invocation increases the chance of unintended workflow execution, misleading responses, or the skill being selected when a more appropriate and safer capability should handle the request.
The description says to use the skill when a user asks for broad categories like 'browser', 'automation', or a 'practical workflow, artifact, checklist, analysis, or implementation support', which is effectively a catch-all activation condition. In a multi-skill agent environment, this ambiguity can cause over-selection of the skill, unintended handling of sensitive tasks, and reduced operator control over which instructions govern a conversation.
The trigger list is broad and includes generic terms such as 'browser', 'automation', 'cli', and 'bug fix', which are common across many unrelated requests. This can cause the skill to activate outside its intended scope, leading to inappropriate routing, prompt collisions with more specific skills, and increased exposure to unsafe or irrelevant browser-automation behaviors.
The description says the skill should be used when a user asks for broad categories such as work-productivity, browser, headless, or automation, without clear exclusion criteria. This ambiguous activation condition makes it easy for the skill to be invoked in contexts far beyond the validated use case, which can result in misrouting and inappropriate assistance.
The trigger keyword list is very broad and includes generic terms like "browser", "automation", "cli", and "bug fix", which can cause the skill to activate for many unrelated requests. Overbroad activation increases the chance that this skill intercepts prompts outside its intended scope, leading to unsafe routing, confused behavior, or accidental application of browser-automation guidance where it is not appropriate.
The default prompt is framed as a broadly applicable helper for common terms like work-productivity, browser, automation, and practical workflow support, while the policy also allows implicit invocation. That combination can cause the skill to trigger in ordinary user conversations without clear user intent, increasing the risk of unexpected prompt injection surface expansion, unintended tool usage, or unreviewed influence over responses.
The trigger sentence is ambiguous and insufficiently scoped, so the routing system may match it for requests that only loosely resemble the intended requirement. Because this skill covers browser-style workflows and automation support, ambiguous activation increases the chance of misrouting users into operational guidance they did not request, reducing safety and reliability.
This Chinese-language README contains substantial English content for core descriptive and trigger text, but it does not explicitly tell users that mixed-language interaction is expected or provide a language preference choice. That can amount to an implicit language policy constraint without clear user opt-in.
The file is a Chinese-language skill document, but the trigger examples and key phrases are presented in English only and appear to prescribe invocation language. This may create an implicit language constraint without stating that users can use other languages or locales.
No suspicious patterns detected.
