deploying-on-azure

Warn

Audited by Snyk on Feb 15, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.90). The RAG implementation performs runtime vector searches against https://myaisearch.search.windows.net and injects the retrieved documents directly into the model prompt/context (thereby directly controlling the assistant's instructions), and the skill depends on that search endpoint at runtime.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 10:08 PM