aave

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly performs public on-chain queries (see "Prerequisites" noting markets is a public on-chain query and Rule 3 requiring npx fibx@latest aave status to read the Health Factor), so it ingests untrusted third‑party blockchain data that directly influences decisions like whether to proceed with borrows/repays.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill repeatedly requires running "npx fibx@latest" at runtime, which fetches and executes remote code from the npm registry (e.g., https://registry.npmjs.org/fibx or https://www.npmjs.com/package/fibx), so the externally fetched package directly controls execution/behavior.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly designed to perform on-chain financial operations for Aave V3 on Base. It provides specific commands to supply, borrow, repay, and withdraw assets, requires wallet authentication and ETH for gas, auto-handles ETH↔WETH conversion, and instructs use of max (MAX_UINT256) for full repayments/withdrawals. These are direct crypto transaction capabilities (signing/sending blockchain transactions) — not generic tooling — so it grants direct financial execution authority.