dbt-model-style

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of documentation and a utility script intended for local use during data modeling workflows. No malicious behaviors were detected.
  • [REMOTE_CODE_EXECUTION]: The Python script scripts/validate_header.py parses YAML metadata from SQL comments using yaml.safe_load(). This is a secure method that prevents potential arbitrary code execution during the parsing of untrusted configuration data.
  • [DATA_EXFILTRATION]: No network-enabled operations or attempts to exfiltrate sensitive data were found. The script only interacts with local files specified by the user.
  • [CREDENTIALS_UNSAFE]: There are no hardcoded secrets or sensitive credentials in the skill files. The documentation promotes safe secret management practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 05:05 PM
Security Audit — agent-trust-hub — dbt-model-style