ljg-paper-flow

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected.
  • [COMMAND_EXECUTION]: The skill invokes local tools 'ljg-paper' and 'ljg-card'. These are identified as vendor resources from 'lijigang' and represent intended workflow functionality.
  • [PROMPT_INJECTION]: The skill processes external documents (papers and PDFs) which represents an indirect prompt injection surface.
  • Ingestion points: Reads user-supplied arXiv links, paper URLs, and PDF files.
  • Boundary markers: None identified in the workflow instructions.
  • Capability inventory: Triggers tool execution ('ljg-paper', 'ljg-card') based on the content of these external sources.
  • Sanitization: No explicit sanitization or filtering of external content is described before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 02:14 PM