dbt-test

Pass

Audited by Gen Agent Trust Hub on Oct 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses the altimate-dbt deps and altimate-dbt add-packages commands to manage dbt dependencies. These packages are typical for dbt development (e.g., dbt-utils, dbt-expectations) and are managed through standard dbt registry protocols.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of dbt model SQL and YAML schema definitions which are passed to the altimate_core_testgen and altimate_core_validate tools. While this represents a potential injection surface, the risk is typical for development-focused skills and is mitigated by the intended use-case of processing local project files.
  • Ingestion points: Reads project files via read <yaml_file> and model SQL files from the models/ directory.
  • Boundary markers: Not explicitly defined in the manual instructions, but typically handled by the underlying platform tools.
  • Capability inventory: Includes bash command execution, filesystem write and edit, and internal validation functions.
  • Sanitization: Relies on the dbt compilation and validation logic provided by the vendor's engine.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 17, 2026, 07:18 AM