pdf-analysis
Warn
Audited by Snyk on Jul 3, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via the OCR/caption runtime path:
extract_scanned_pdf()/extract_hybrid_pdf()renders each page to PNG and then runscaption.pywith a--promptand the image content, where the captioner’s output (derived from the PDF’s page text) is then appended intoall_textand fed to the agent’s LLM context.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata