pdf-skill

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data by extracting text and metadata from PDF documents, creating a potential surface for indirect prompt injection attacks.
  • Ingestion points: Processes external PDF files for text, table, and data extraction (SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions found within parsed PDF content.
  • Capability inventory: The recommended tools include Puppeteer, Playwright, and PDFKit, which operate within environments capable of network and file system access.
  • Sanitization: No explicit sanitization or validation steps are defined for the extracted data before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 08:06 AM
Security Audit — agent-trust-hub — pdf-skill