refactoring-dbt-models
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a logical process for refactoring dbt models using inspection commands like cat and grep along with dbt CLI validation steps. These actions are standard for the intended data engineering tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve reading local SQL files from the models/ directory, which constitutes a data ingestion surface. While this could theoretically be used for indirect prompt injection if the source code contained malicious comments, it is a standard and expected operation for a refactoring tool.
- Ingestion points: Workflow steps 1 and 3 in SKILL.md use cat to read dbt model files.
- Boundary markers: No specific delimiters are mentioned for the file content analysis.
- Capability inventory: The skill uses shell commands (cat, grep), dbt CLI operations (ls, compile, build, show), and file modification capabilities.
- Sanitization: No explicit sanitization of file content is required for this analysis and refactoring use case.
Audit Metadata