agent-protocol

Warn

Audited by Snyk on Mar 10, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly shows an A2A client that fetches agent cards from arbitrary base URLs (discover() calling {base_url}/.well-known/agent.json) and a ProtocolBridge that sends tasks to external A2A agents and ingests their artifacts/text parts, so untrusted third‑party agent content can be fetched and directly influence tool selection and subsequent actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 10, 2026, 12:13 AM