urizen

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text via the skill’s news and filings research endpoints (e.g., GET /api/quant/news?symbol=... and GET /api/quant/filings?symbol=... under https://urizenfund.com/api), where returned headlines/filing text can originate from external user/public sources.

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

• Direct money access detected (high risk: 1.00). The skill explicitly provides blockchain trading endpoints that produce ready-to-sign transactions for buying the $URI token. It documents a specific GET quote endpoint that returns tx:{to,data,value}, requires chainId/token/router/calldata validation, and names the exact chain (4663), token address, and router address/selector. This is a targeted crypto execution capability (non-custodial swap quoting) — i.e., direct financial execution authority for on-chain trades.