Sign InOpen Brain
AI EngineerVideoSource Linked

Teaching agents to pay — Anna Spysz, Stripe

Agent checkout needs more than payment tools: enforce user-set limits outside the agent, expose structured merchant data, honor cancellation, and log every decision.

AI Engineer · Sep 1, 2026
Open Source Open MarkdownOpen JSON
Source Summary

The demo uses **UCP** as a shared interface for agents and merchants. Stores expose a capabilities manifest and structured catalog; checkout uses a **shared payment token**, so the agent receives a token rather than the raw card number.

Practical Implication

Put spending and token validity checks at the payment provider, not inside a prompt. Also disclose the agent, show fees, honor stop or cancel, reject totals above the user's maximum, avoid pressure tactics, and keep auditable decision logs.

Agent-Ready Context
The demo uses **UCP** as a shared interface for agents and merchants. Stores expose a capabilities manifest and structured catalog; checkout uses a **shared payment token**, so the agent receives a token rather than the raw card number.

Put spending and token validity checks at the payment provider, not inside a prompt. Also disclose the agent, show fees, honor stop or cancel, reject totals above the user's maximum, avoid pressure tactics, and keep auditable decision logs.

The hostile-salesperson example shows that a **system prompt can materially alter behavior**, even when the transaction plumbing is unchanged. These safeguards are presented as a practical checklist, not evidence that every manipulation or commerce failure is covered.
Connected Context · Feed7 Judgment

This makes the external-control pattern concrete for agent payments: UCP structures merchant interaction, tokenization withholds raw card data, and the provider enforces amount and validity limits. It also adds a distinct warning that safe transaction plumbing does not neutralize manipulative prompting, so disclosure, cancellation, fee visibility, and audit logs remain separate behavioral controls.

Your Agent Just Authorized What?! — Jay Mok & Ben Coumes, PaypalPayPal’s scoped, time-bound mandate complements the provider-side amount and token checks here; together they place payment authority in independently enforceable transaction controls rather than prompt compliance.From coding to Knowledge work agents — Karan Vaidya, ComposioThe candidate’s preflight checks for irreversible actions are instantiated here as provider-enforced spending limits, token validation, fee disclosure, cancellation handling, and decision logs.AI Agents Are Just Distributed Systems Now — Salman Munaf, TikTokDistributed-systems safeguards remain necessary once payment mutates external state; this Signal adds commerce-specific authorization and anti-manipulation controls but does not address ambiguous outcomes after timeouts.Healthcare’s Agent Bytecode: X12 as the Harness for AI Agents — Vasant Kearney, OnlayUCP and X12 both show how a shared domain protocol can make agent interactions structured and checkable, while each still requires controls beyond protocol conformance.
Context Map
agentsecurity#tool-use#harness-engineering#agent-reliability
Uncertainty
The hostile-salesperson example shows that a **system prompt can materially alter behavior**, even when the transaction plumbing is unchanged. These safeguards are presented as a practical checklist, not evidence that every manipulation or commerce failure is covered.