skills/databricks-solutions/databricks-exec-code-mcp/databricks-data-engineering/Gen Agent Trust Hub
databricks-data-engineering
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill provides well-structured templates and best practices for developing Databricks data pipelines following the medallion architecture. All external tools and CLI patterns used are standard for the Databricks ecosystem and appropriate for the skill's stated purpose.
- [PROMPT_INJECTION]: The skill documents an attack surface for indirect prompt injection as it facilitates the ingestion of external, potentially untrusted data (e.g., JSON files from
/mnt/source/). However, the risk is mitigated by the skill's primary focus on structured data processing via PySpark, which includes schema enforcement and validation logic. - Ingestion points:
SKILL.md(Phase 2: Bronze Layer Development, Phase 3: Silver Layer Development) - Boundary markers: None present in the data ingestion logic.
- Capability inventory:
Bash,Write,Edit, andReadtools;databricksCLI usage; execution of PySpark code on remote clusters viadatabricks-testing. - Sanitization: The Silver layer implementation (Phase 3) includes explicit sanitization through type casting (
double,int,timestamp), null field removal (isNotNull()), and domain-specific range validation (amount >= 0,quantity > 0).
Audit Metadata