paper-analysis-assistant

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads PDF documents from arXiv's official website using the requests library in scripts/download_pdf.py.
  • This operation is performed through the arxiv.org domain, which is a well-known academic repository.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted external PDF content and interpolating it into an HTML report without sanitization.
  • Ingestion points: External data enters the system through scripts/download_pdf.py (fetching from arbitrary arXiv URLs) and is processed by scripts/extract_text.py to create paper.txt.
  • Boundary markers: No delimiters or isolation instructions are used when embedding the extracted paper content into the HTML template in scripts/generate_html.py.
  • Capability inventory: The skill includes scripts capable of network requests (scripts/download_pdf.py), file system writes across multiple formats (scripts/generate_html.py, scripts/generate_ppt.py), and system speech engine interaction (scripts/text_to_speech.py).
  • Sanitization: There is no evidence of HTML escaping or content filtering in scripts/generate_html.py before the extracted paper text is written to the output file.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 09:42 AM
Security Audit — agent-trust-hub — paper-analysis-assistant