troubleshoot-errors

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

• Third-party content exposure detected (high risk: 0.70). The skill includes instructions and code to fetch and display live, public blockchain and node data (e.g., algorand.client.algod.pendingTransactionInformation, indexer.lookupTransactionByID, and examples referencing https://testnet-api.algonode.cloud and simulateResponse.txnGroups[0].failureMessage), which are untrusted, user-generated on the public network and are expected to be read/interpreted as part of the troubleshooting workflow.

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

• Direct money access detected (high risk: 1.00). The skill is specifically targeted at Algorand blockchain operations and includes explicit transaction-related APIs and actions (e.g., algorand.send.asset_opt_in(...), handling transaction rejections/overspend, funding sender accounts, transaction params). Those are concrete crypto/blockchain operations that can create/send on‑chain transactions (move assets), so this grants direct financial execution capability.