exploring-blockchain-data

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

• Third-party content exposure detected (high risk: 0.80). This skill autonomously fetches and ingests data from open/public third‑party sources (e.g., Etherscan/Polygonscan/BSCScan explorer APIs and CoinGecko) as shown in SKILL.md and implemented in scripts/chain_client.py (_explorer_call, get_contract_abi) and scripts/token_resolver.py (CoinGecko requests), and that untrusted, user-submitted explorer/ABI/metadata content is read and used to decode transactions and influence protocol identification and outputs—creating a clear path for indirect prompt injection.