skill-persona-check
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs its stated purpose of persona consistency checking using local profile files. It does not attempt to execute external code, exfiltrate data, or bypass system constraints.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user content (the 'content to be checked') as its primary input. While this represents a potential injection surface, the skill does not possess any dangerous capabilities—such as file writing, shell command execution, or network requests—that could be exploited via malicious input. The instructions also explicitly state that the tool should only warn the user and never block publication, maintaining user control.
- [EXTERNAL_DOWNLOADS]: The documentation in
EASEL-META.mdreferences several well-known industry services and tools (Taskade, ZBrain, Siteimprove, GitHub) for attribution and methodology reference. These are static, informational links and do not involve dynamic code execution or automated downloads.
Audit Metadata