solidity-debug

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and decode public on-chain data via RPC calls (e.g., "cast receipt <tx_hash> --rpc-url $RPC_URL", "cast tx", "cast call", "cast run") and even references external analysis platforms ("Tenderly or Blocksec Phalcon"), which are untrusted, user-generated/public sources whose calldata or revert strings the agent is expected to read and that can materially influence debugging actions.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill is explicitly built for on-chain transaction handling and debugging using Foundry's cast. It includes concrete, actionable commands to send transactions (e.g., cast send <contract> ... --account <KEYSTORE_NAME> --rpc-url $RPC_URL), replay transactions (cast run), and manage keys/keystore. Those commands perform blockchain transactions and can move funds or sign transactions. This is a specific crypto/blockchain execution capability (not a generic tool), so it grants direct financial execution authority.