documenting-dbt-models
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the local dbt project, specifically SQL models and
schema.ymlfiles, which could theoretically contain malicious instructions. - Ingestion points: Reads
.sqlfiles in themodels/directory andschema.ymlfiles throughout the project (SKILL.md, Steps 1-3). - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within these files.
- Capability inventory: The skill reads files (
cat,grep), searches the filesystem (find), and executes dbt documentation commands (dbt docs generate). - Sanitization: No specific sanitization or validation of the ingested file content is performed.
- Context: This behavior is essential to the skill's primary purpose of generating documentation from existing codebases.
Audit Metadata