gold-layer-design
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not contain any malicious instructions, prompt injections, or obfuscated code. It follows the intended functionality of a data engineering design orchestrator.
- [COMMAND_EXECUTION]: The skill includes a Python utility (
scripts/generate_lineage_csv.py) to process YAML schemas and generate CSV/Markdown lineage reports. This script is well-defined and performs standard file processing operations without executing arbitrary shell commands. - [DATA_EXPOSURE]: The skill reads schema metadata from user-provided files in the
context/directory (e.g.,Wanderbricks_Schema.csv). This data is used for classification and relationship inference. No access to sensitive system files, environment variables, or hardcoded credentials was found. - [PROMPT_INJECTION]: The skill parses metadata such as column comments to infer relationships. This data ingestion is handled programmatically via regex and does not influence the agent's core instructions or safety guidelines.
Audit Metadata