ponder

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflows (SKILL.md) show it directly ingesting public, user-submitted blockchain data via RPC endpoints (e.g., PONDER_RPC_URL_<CHAIN_ID>) and processing contract events/call traces and readContract results (e.g., tokenURI, event.args) as part of its indexing logic, so untrusted third‑party content can influence runtime actions.