Security audit
Security checks for vulnerabilities and agentic risk
This skill provides ordinary workflow guidance, with broad activation wording but no harmful behavior found in its files.
Expect general planning and implementation guidance; browser automation depends on your existing tools. Consider explicit invocation because generic keywords may cause this skill to activate for unrelated requests.
These keywords occur in many unrelated requests and do not reliably identify an Agent Browser task. Although the README lists more specific trigger sentences, it does not state that the generic keywords require Agent Browser context or exclude unrelated uses.
The description activates the skill for requests involving a practical workflow, artifact, checklist, analysis, or implementation support, while the trigger list includes generic terms such as tree, optimized, and bug fix. These criteria overlap with many unrelated tasks and have no explicit exclusions or requirement that the request concern Agent Browser workflows.
These keywords occur in many requests unrelated to Agent Browser workflows and could cause unintended activation. The skill provides no exclusion conditions or requirement that these terms appear in an Agent Browser-specific context.
The keyword list includes generic terms such as tree, optimized, cli, and bug fix, which commonly occur in unrelated requests. Although the subsequent trigger sentences mention Agent Browser, the file does not clarify whether these keywords can independently activate the skill.
The description and default prompt contain truncated demand-summary text rather than a clear statement of the tasks this skill handles. With implicit invocation enabled, these descriptions do not reliably distinguish when the skill should activate from unrelated browser or productivity requests.
No suspicious patterns detected.
