environmental-science-and-technology

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a purely instructional resource for journal submission guidance. It contains structured text describing journal scope, evidence bars, and formatting requirements.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides a framework for the agent to process untrusted user data (manuscripts). While this constitutes an ingestion point, the skill possesses no exploitable capabilities such as file system writing, network requests, or shell command execution, rendering the surface area for injection attacks benign.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 11:33 AM
Security Audit — agent-trust-hub — environmental-science-and-technology