autonomous-agents

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [NO_CODE] (SAFE): The skill consists exclusively of markdown documentation. No Python, JavaScript, or shell scripts are included, nor are there any commands to execute.
  • [GENERAL] (SAFE): No patterns of prompt injection, data exfiltration, or obfuscation were identified. The content focuses on engineering best practices for agent reliability and safety.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 08:28 AM