listener-first
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional text and academic references for drafting persuasive and honest messages. It does not include any executable code, scripts, or system commands.
- [SAFE]: No external dependencies, package installations, or remote code execution patterns were identified in the skill files.
- [SAFE]: The skill does not request access to sensitive data, environment variables, or network resources. It operates purely within the text-processing context of the agent.
- [SAFE]: No obfuscation techniques, hidden URLs, or prompt injection patterns were found. The instructions emphasize honesty and transparency through the 'Admission rule' and 'full-disclosure test'.
- [SAFE]: The skill defines clear boundaries for its application, excluding adversarial contexts and visual formatting, and provides guidance for handling power asymmetries.
Audit Metadata