Skills
Navigate
skills/nimbleway/agent-skills/nimble-databricks-data-products/Gen Agent Trust Hub

nimble-databricks-data-products

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands using the Databricks CLI via scripts/ingest.sh and scripts/build_dashboard.py for legitimate API interactions such as submitting SQL statements and managing dashboards. The Python implementation uses the subprocess module with argument lists and temporary files for JSON payloads, which effectively mitigates command injection risks.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external web sources through Nimble agents, creating a standard surface for indirect prompt injection. The skill mitigates this by focusing on structured data extraction and using defensive SQL patterns like try_cast and regexp_replace to sanitize values during ingestion into Delta tables.
  • [EXTERNAL_DOWNLOADS]: Documentation references setup scripts located in the vendor's official GitHub repository (Nimbleway). These resources are consistent with the skill's purpose and are intended for manual installation by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 08:34 PM
Security Audit — agent-trust-hub — nimble-databricks-data-products