dev-data
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional guidelines and best-practice patterns for data engineering. It contains no executable code or malicious instructions.
- [EXTERNAL_DOWNLOADS]: The skill references documentation and repositories for established open-source projects, including Apache Airflow, dbt, pandas, Polars, and DuckDB. These references target well-known and trusted services and are documented neutrally.
- [DATA_EXFILTRATION]: The skill includes dedicated sections on PII handling, data classification, and compliance (GDPR/CCPA). It provides patterns for static/dynamic masking and tokenization to prevent unauthorized data exposure.
- [PROMPT_INJECTION]: The skill identifies and mitigates indirect prompt injection surfaces.
- Ingestion points: Processes external data formats including CSV, JSON, Parquet, and Excel.
- Boundary markers: Mandates the use of "Input contracts" and "Schema-first" design to define data boundaries.
- Capability inventory: Utilizes standard data processing libraries (Polars, DuckDB, Spark) and logging tools (MLflow).
- Sanitization: Recommends strict schema validation and PII masking as core pipeline components.
Audit Metadata