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.pyparses YAML metadata from SQL comments usingyaml.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