skills/ninehills/skills/learn/Gen Agent Trust Hub

learn

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill mentions a fallback to curl for fetching web content when the specialized /read skill is unavailable. This is a standard utility for the skill's primary purpose of gathering research material from online sources.
  • [PROMPT_INJECTION]: The research process involves ingesting data from external URLs. While this creates a surface for indirect prompt injection (where instructions hidden in a webpage could influence the research output), the skill is designed for a multi-phase workflow with human-in-the-loop review (Phase 6), which significantly mitigates the impact of such automated manipulation.
  • [DATA_EXFILTRATION]: The skill performs network operations (web fetching) and writes files to ~/Downloads/. These operations are well-scoped to the stated goal of technical research and document preparation and do not involve sensitive system paths or credential locations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 05:57 PM