airflow-dag-patterns

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses globals() in resources/implementation-playbook.md to dynamically register DAG objects at runtime based on a configuration list. While this is a standard idiom for generating multiple similar pipelines in Apache Airflow, it involves dynamic manipulation of the global namespace.\n- [PROMPT_INJECTION]: The templates in resources/implementation-playbook.md describe data ingestion patterns that could be vulnerable to indirect prompt injection.\n
  • Ingestion points: Data is read from S3 buckets using pandas.read_csv in the extract tasks.\n
  • Boundary markers: The provided code samples do not include delimiters or instructions to ignore embedded commands within the processed data.\n
  • Capability inventory: The skill defines tasks executed via PythonOperator and BranchPythonOperator for data transformation and routing.\n
  • Sanitization: There is no evidence of input validation or sanitization for the data retrieved from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 10:57 PM
Security Audit — agent-trust-hub — airflow-dag-patterns