kometa-debug-review

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured instructions for the AI to perform static analysis of user-provided YAML configurations. It identifies common structural errors like nesting issues, incorrect top-level keys, and missing connectors.
  • [SAFE]: No evidence of data exfiltration or credential harvesting was found. The skill does not access sensitive system paths (e.g., .ssh, .aws) and does not contain any hardcoded secrets.
  • [SAFE]: The skill does not perform any network operations or external downloads. All references to 'evals/fixtures/' and 'references/cards/' appear to be internal documentation or example files used to provide context for the LLM.
  • [SAFE]: There is no executable code (Python, JavaScript, or Shell) included with the skill. The logic is entirely contained within natural language instructions and agent configuration metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 12:27 PM
Security Audit — agent-trust-hub — kometa-debug-review