databricks-vector-search

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes official Databricks SDKs and CLI tools for managing Vector Search resources, adhering to standard vendor practices.
  • [PROMPT_INJECTION]: The skill documents RAG patterns that ingest data from Delta tables into model prompts. This represents an indirect prompt injection surface inherent to the RAG use case.
  • Ingestion points: references/end-to-end-rag.md (retrieval from Delta indexes).
  • Boundary markers: Markdown headers are used to delimit documents within the prompt.
  • Capability inventory: Operations are limited to API-based data retrieval and model inference.
  • Sanitization: No content sanitization is demonstrated in the provided code examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 08:56 PM
Security Audit — agent-trust-hub — databricks-vector-search