ask-pdf-processing
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill has a vulnerability surface for indirect prompt injection via the processing of untrusted PDF content. Malicious instructions inside PDFs could be extracted and interpreted by the agent.\n
- Ingestion points: PDF content is ingested in
scripts/extract_text.py(via pdfplumber) andscripts/fill_form.py(via pypdf).\n - Boundary markers: Absent. The scripts do not use any delimiters or specific instructions to isolate extracted content from the agent's system instructions.\n
- Capability inventory: The skill includes file-writing capabilities in
scripts/extract_text.py(Path.write_text),scripts/fill_form.py(writer.write), andscripts/merge_pdfs.py(writer.write). No network or shell execution capabilities are implemented within the scripts themselves.\n - Sanitization: Absent. Extracted text is returned to the agent without validation, filtering, or escaping.
Audit Metadata