nfd

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

• Third-party content exposure detected (high risk: 0.90). The SDK and docs show the agent will call public REST endpoints (e.g., https://api.nf.domains via NfdApiClient) and fetch user-generated NFD properties, search results, avatars/banners (including IPFS -> public gateway resolution) and metadata (properties.userDefined), which are open/public third‑party content that the agent is expected to read and display.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly designed for on-chain Algorand operations that move assets: it exposes minting, buying/claiming NFDs, managing linked addresses/metadata with an explicit signer API (nfd.setSigner(activeAddress, transactionSigner)), and vault send-to/send-from operations. It depends on algosdk and performs transaction-signing and asset transfer actions. These are specific crypto/blockchain transaction capabilities (wallet signing and sending assets), so it grants direct financial execution authority.