agent-tuning

Pass

Audited by Gen Agent Trust Hub on May 27, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill content is purely informational, serving as a reference for configuring agent environments. It describes various security postures (Restrictive, Balanced, Permissive) and encourages best practices such as starting with minimal permissions.
  • [NO_CODE]: There are no scripts, executables, or automated commands included in the skill. The provided shell commands (e.g., cat ~/.config/opencode/opencode.json) are examples for manual debugging by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
May 27, 2026, 02:35 PM
Security Audit — agent-trust-hub — agent-tuning