creating-dbt-models
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses standard development practices for dbt (data build tool), utilizing basic shell commands like
catandfindalongside the officialdbtCLI to create and verify data transformations. No malicious patterns, obfuscation, or unauthorized access attempts were identified.- [INDIRECT_PROMPT_INJECTION]: The skill has a functional requirement to read external project configuration (dbt_project.yml) and preview database data (dbt show), which presents a standard indirect prompt injection surface area common to data engineering tasks. - Ingestion points: Local files (dbt_project.yml, *.sql) and database table previews.
- Boundary markers: None; the agent processes the files as part of the development context.
- Capability inventory: Execution of dbt commands (build, compile, show) and file system discovery.
- Sanitization: The skill relies on the user's local project environment and does not implement additional sanitization of the data models it analyzes.
Audit Metadata