visual-scrape

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities detected. The skill performs PDF image extraction and analysis as described.
  • [PROMPT_INJECTION]: The skill processes untrusted text from user-provided PDF files which creates a surface for indirect prompt injection.
  • Ingestion points: PDF content is ingested into the agent context via the Read tool in Step 2 of SKILL.md.
  • Boundary markers: No delimiters or isolation instructions are used when processing the text.
  • Capability inventory: The skill utilizes Bash, Write, Edit, and Read tools.
  • Sanitization: No sanitization of the PDF text is performed prior to analysis.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a local Python script (extract_visuals.py) for PDF processing. The execution is controlled and the script's source code is provided and verified as safe.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 02:14 PM
Security Audit — agent-trust-hub — visual-scrape