table-extractor

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the camelot-py and pandas Python packages, as well as system dependencies like ghostscript and tcl-tk via package managers (pip, brew, apt-get). These are standard requirements for the library's stated functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to parse external PDF documents, which creates a potential surface for indirect prompt injection if an adversary embeds malicious instructions within a PDF to influence the agent's behavior during or after extraction.
  • Ingestion points: Processes external PDF files via camelot.read_pdf in SKILL.md.
  • Boundary markers: No explicit markers are defined to delimit or sanitize the extracted content from the rest of the prompt.
  • Capability inventory: The skill has file-writing capabilities through methods such as to_csv, to_excel, to_json, to_html, and fig.savefig (as documented in SKILL.md).
  • Sanitization: The instructions do not specify any sanitization, validation, or filtering of the extracted content before it is processed or exported.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 08:53 PM