image-outpainting

Warn

Audited by Snyk on Feb 20, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md requires and demonstrates sending arbitrary public image URLs via the "image_urls" parameter (e.g., multiple curl examples) which the each::sense service fetches and interprets to drive outpainting decisions, so untrusted third-party content can influence generation behavior.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 20, 2026, 07:57 AM