agents-build

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

• Third-party content exposure detected (high risk: 0.75). High risk because the required Browser capability routes the agent through web browsing/scraping (“browser.md” Path A/B/C) and the agent tool outputs fetched page text back into the agent’s LLM context, which can originate from arbitrary third-party pages (public web content fetched at runtime).

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

• Direct money access detected (high risk: 1.00). The skill explicitly includes an "AgentCore Payments" capability to "pay for x402-protected APIs, tools, or content" and references payment-related resources (payment manager/connector, payment-connector). These are specific payment execution features (i.e., sending payments/microtransactions), which constitute direct financial execution authority.