bigquery-migration-data-loader

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [Command Execution]: The loader script utilizes the subprocess module to interface with Google Cloud SDK tools. The implementation uses list-based command construction, which effectively mitigates shell-related security considerations.
  • [Input Validation]: The skill incorporates defensive programming by validating environment identifiers such as Project IDs and Dataset IDs against strict regular expression patterns before execution.
  • [Indirect Prompt Injection]: The skill is designed to ingest and execute SQL view definitions from external migration metadata directories. Ingestion points: SQL and YAML files are read from the directory specified by --translation-metadata-dir. Boundary markers: None explicitly defined for the SQL statement content. Capability inventory: The script performs file operations and executes commands including bq query, bq load, and bq mk. Sanitization: The script validates core identifiers using regex to ensure they conform to expected formats.
  • [Dependency Management]: The skill uses the pyyaml library for configuration parsing, following best practices by pinning the version to 6.0.2 for environment stability.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 11:16 AM