sandwich-attack-investigator-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and decodes public, user-generated blockchain content (e.g., "pull surrounding txs/bundle peers", "Historical blocks / bundles from public APIs", and toolchain entries like "Jito explorers, block traces, Dune, Flipside") and requires the agent to read and interpret those third‑party sources as part of its analysis, so untrusted external content can materially influence decisions and workflow.