gpt-image-edit
Warn
Audited by Snyk on Apr 30, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill explicitly accepts up to 10 "publicly-fetchable HTTPS" image URLs that the RunComfy model server fetches and uses to guide edits (see the "images" field in SKILL.md and the "Third-party content" note), so untrusted user-provided images can embed content that influences model behavior and enable indirect prompt injection.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata