extracting-tables

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill serves as a guide for using the xberg utility to extract structured data from documents. The provided examples for CLI usage and Python/Node.js integration are standard and do not contain malicious patterns.
  • [COMMAND_EXECUTION]: The instructions include legitimate commands for the xberg tool and jq. These are scoped to the primary purpose of data extraction and transformation.
  • [PROMPT_INJECTION]: The skill describes processing external documents (PDFs, images, spreadsheets). While this presents a surface for indirect prompt injection from untrusted document content, this is the core functionality of the tool and is treated as a safe ingestion point in this context. Ingestion points: xberg extract <file> commands (SKILL.md). Boundary markers: Not specified in the documentation examples. Capability inventory: File reading, table layout analysis, and Markdown/JSON generation (SKILL.md). Sanitization: None explicitly mentioned in the usage examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 10:22 PM
Security Audit — agent-trust-hub — extracting-tables