pdf

Warn

Audited by Snyk on Feb 16, 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 ingests and processes arbitrary user-supplied PDFs and derived images that the agent is explicitly expected to read and interpret (e.g., text/table extraction with pdfplumber/pypdf, convert_pdf_to_images.py + forms.md which instructs Claude to visually inspect page PNGs and produce fields.json), which exposes it to untrusted third‑party content that could carry indirect prompt injections.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 01:36 PM