NYC

multi-model

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 skill sends prompts to external models specified by ARENA_MODELS and ARENA_PROVIDER_*_BASE_URL (e.g., public OpenAI/Ollama endpoints) and ingests their generated outputs as candidate answers and critiques in the multi-model loop, which exposes the agent to untrusted third-party content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 03:38 AM