walletconnect-pay

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

• Third-party content exposure detected (high risk: 0.85). The skill explicitly loads third-party pages from the payment API (e.g., the collectData.url in SKILL.md Step 5 and the various references/*.md WebView examples) and relies on JS bridge messages like "IC_COMPLETE" from those pages to proceed, so untrusted external content can directly influence the agent's actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly loads external collectData.url pages (the WebView URL returned by the pay.walletconnect.com API / payment options) at runtime to collect compliance data and relies on their postMessage (IC_COMPLETE / IC_ERROR) signals to proceed, meaning remote page content/JS can directly control the agent/app flow.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly a payments integration for WalletConnect Pay. It defines SDK/API methods and flows to fetch payment options, collect EIP‑712 signatures, and call confirmPayment (with statuses SUCCEEDED/PROCESSING/FAILED/etc.). It specifically targets USDC on named blockchains and instructs how to sign transactions and confirm payments — i.e., to initiate and finalize on‑chain/off‑chain payments. These are direct crypto financial execution capabilities (wallet signing + payment confirmation), so it meets the "Direct Financial Execution" criteria.