miyazaki-perspective
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains no malicious code, hidden instructions, or attempts to exfiltrate sensitive data. All instructions are focused on character development and persona consistency.
- [NO_CODE]: The skill is composed exclusively of markdown documentation and YAML metadata. There are no associated scripts (Python, JavaScript, shell) or binaries provided that could execute logic independently of the AI agent's inference.
- [PROMPT_INJECTION]: While the skill assigns a specific persona and mindset, it does not include commands to override safety protocols, bypass system filters, or ignore underlying instructions. It provides guidance on maintaining the persona while acknowledging boundaries like professional mental health advice.
- [EXTERNAL_DOWNLOADS]: The skill references repositories and installation commands within the 'nuwa-skills' organization. These are consistent with the skill's authoring environment and represent standard platform installation links rather than suspicious external dependencies.
Audit Metadata