tracing-downstream-lineage

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external data by reading DAG source code and database view definitions to identify lineage, creating a potential surface for indirect injection. \n
  • Ingestion points: Source code is retrieved via af dags source, and metadata is pulled from SQL information schemas. \n
  • Boundary markers: The instructions do not define specific delimiters to separate ingested data from agent instructions. \n
  • Capability inventory: The agent uses af CLI tools to list and read DAG files. \n
  • Sanitization: No sanitization of the retrieved code is specified before analysis. \n- [COMMAND_EXECUTION]: The skill utilizes the af (Astro CLI) tool for administrative discovery tasks (af dags list, af dags source). This tool is a vendor-provided utility for the Astronomer/Airflow environment and is used here in accordance with its intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:33 PM
Security Audit — agent-trust-hub — tracing-downstream-lineage