pov-vlog-presence-by-wat8s8
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a prompt engineering workflow that focuses on creating consistent visual styles for AI video generation. It utilizes a local reference file (
references/cases.md) containing descriptive prompts and metadata. - [EXTERNAL_DOWNLOADS]: The skill references external URLs for media and source documentation (e.g.,
goodcase.ai,x.com). These are used solely as reference material for the AI to analyze visual traits and do not involve execution of remote code or scripts. - [INDIRECT_PROMPT_INJECTION]: While the skill processes data from a reference file that contains natural language prompts, this is a standard design for template-based generation. The skill includes specific instructions to 'Preserve', 'Replace', and 'Avoid' certain traits, providing a framework that minimizes accidental obedience to embedded instructions in the source text.
- [DATA_EXFILTRATION]: There are no signs of sensitive data access or exfiltration. The instructions explicitly forbid publishing or making external account changes without user approval.
Audit Metadata