pdf-construction

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill extracts and processes text from untrusted external PDF documents, presenting a surface for indirect prompt injection. \n- Ingestion points: Text is extracted in extract_rfi_data, extract_spec_sections, and split_drawing_package in SKILL.md. \n- Boundary markers: No explicit delimiters or instructions are used to separate extracted content from system instructions. \n- Capability inventory: The skill possesses filesystem access to read/write PDFs and Excel files via pypdf, pdfplumber, and pandas. \n- Sanitization: No validation or sanitization of extracted text is performed before it is processed by the AI.\n- [EXTERNAL_DOWNLOADS]: The skill requires several external Python libraries for its functionality. \n- Evidence: Documentation in SKILL.md specifies the installation of pypdf, pdfplumber, reportlab, and pandas from official package registries.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 04:29 AM