Skills
skills/betterpromptme/skills/05-indoor-selfie-with-any-outfit/Gen Agent Trust Hub

05-indoor-selfie-with-any-outfit

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the betterprompt CLI to process inputs and generate images. This is the intended primary function of the skill.
  • [EXTERNAL_DOWNLOADS]: The skill specifies the installation of the betterprompt package from a standard registry and links to documentation on the vendor's repository.
  • [DATA_EXFILTRATION]: The skill processes user-provided images and text for image generation; no patterns for exfiltrating sensitive local files or credentials were identified.
  • [PROMPT_INJECTION]: The skill ingests untrusted text data via the outfit parameter in SKILL.md. This data is interpolated into the generation process without explicit boundary markers or sanitization, though the risk is limited to the image generation context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 04:46 PM