1inch-aqua

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

• Third-party content exposure detected (medium risk: 0.30). Skill aqua read actions (maker_stats, list_maker_strategies, strategy_overview/activity/volume, list_opened) can ingest outsider-authored free text when the runtime workflow accepts user-supplied parameters (e.g., maker/app/strategyHash and feed selection) to fetch analytics/feeds from the provider.

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

• Direct money access detected (high risk: 1.00). The skill explicitly exposes crypto transaction actions: "build_ship", "build_dock", "build_swap" and describes non-custodial execution where the user's wallet signs and, when connected via WalletConnect, transactions are sent through the connected wallet (user approves each prompt). These are specific blockchain wallet/swap/signing capabilities (direct financial execution).