Agent Payments
An agent hits an HTTP 402 and has to pay for what it wants, without holding a wallet key or an unlimited budget. Amazon Bedrock AgentCore Payments makes that a governed action instead of a stored credential — here's what I ran against it.
Calling an API is easy compared with letting a non-human actor spend money without handing it private keys, an unlimited budget, or a prompt-level spending rule it can be talked out of. AgentCore Payments makes spend a governed action — session budgets, Cedar policy, and a real audit trail — instead of a stored card number. This is the order I actually verified it in: the payment lifecycle and the guardrails first, then the browser-specific wrinkle, then how it compares to the competing protocol.
-
An Agent That Can Pay for Its Tools
Payment is not a tool call—it is a governed financial action. An agent that pays requires user consent, credential isolation, session budgets, policy checks, and a proof path the model cannot bypass.
-
AgentCore Browser Meets x402: A Browser That Pays
Updated:A browser gets an HTML x402 challenge where a script gets JSON. Extract it from the DOM, ask AgentCore Payments for proof, then retry.
-
MPP vs x402: Two Competing Answers to the Same HTTP 402
AgentCore Payments speaks both protocols natively. I tested x402's metered settlement and built a spec-compliant MPP challenge to compare what each actually buys: simplicity and on-chain audit trails versus multi-rail reach and high-frequency billing.