dbt-bigquery
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes several command-line interface (CLI) tools including
dbt,bq, andgcloud. These tools are used for legitimate project tasks such as environment verification, metadata retrieval (listing datasets/tables), data previewing, and compiling/testing dbt models. The operations are consistent with the skill's primary purpose. - [INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection as it processes untrusted data from external sources.
- Ingestion points: The agent reads data from BigQuery table previews (
bq head), schemas (bq show), Google Cloud logs (gcloud logging read), and local model files which may have been influenced by external sources. - Boundary markers: The instructions do not specify the use of clear delimiters or instructions to ignore embedded commands when processing this external data.
- Capability inventory: The skill possesses the capability to execute shell commands (
dbt,bq,gcloud) and write files to the local file system (dbt models, configuration files). - Sanitization: No explicit sanitization or validation protocols are mentioned for the content retrieved from external datasets or logs before it is used to generate or modify code.
Audit Metadata