Skills
Navigate
skills/legout/data-agent-skills/building-data-pipelines/Gen Agent Trust Hub

building-data-pipelines

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources such as Parquet and CSV files, which constitutes a vulnerability surface for indirect prompt injection if the ingested data contains instructions for the agent.\n
  • Ingestion points: Data is loaded via polars.scan_csv and polars.scan_parquet in templates/complete_etl_pipeline.py and references/pipeline-patterns.md.\n
  • Boundary markers: While the skill lacks natural language delimiters for prompt isolation, it implements structural validation through schema checks.\n
  • Capability inventory: The pipeline scripts have capabilities for file system writes and database interactions in templates/complete_etl_pipeline.py.\n
  • Sanitization: The skill includes a validate_schema function in templates/complete_etl_pipeline.py that validates column names and data types against expected definitions, mitigating risks from malformed data structures.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 01:39 PM