research-figure
Warn
Audited by Snyk on Aug 22, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The runtime workflow reads an outsider-authored sketch (image/PPT/scan) and sends it to a vision LLM in Step 1 (“read sketch → extract structured layout JSON”), so attacker-supplied text can be ingested via the sketch content.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata