Oracle Agent Spec Expert
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The spec explicitly includes tools/nodes that fetch and consume external web/API content—e.g., the Multi-Agent "researcher" AgentNode with tools "web_search" and "fetch_document" and the ResearchPipeline pattern's use of web_search/fetch_docs, plus APINode examples calling public endpoints—so the agent is expected to ingest arbitrary public/untrusted third-party content.
Audit Metadata