thermo-nuclear-performance-review

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructional content focused on software performance auditing. It does not contain any executable code, shell commands, or network operations. The metadata correctly uses disable-model-invocation: true, which is a safety best practice for instruction-only skills. No malicious patterns such as prompt injection, obfuscation, or data exfiltration were found. The eval files are standard testing configurations for the skill's logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 11:52 AM
Security Audit — agent-trust-hub — thermo-nuclear-performance-review