Binding an AI Shopping Agent's Purchase Intent to the Actual Seller, Total Obligation, Payment, Order and Delivered Outcome

AI shopping agents discover products, assemble carts, accept terms, pay and track delivery across marketplaces, merchants, payment providers and carriers. Each step produces a signal that is easy to over-read: a recommendation is not consent, a consent screen may not show the shipping charge or the renewal hidden in a free trial, a valid payment credential is not authority for this product and this seller, a marketplace listing may route money to a different beneficiary, a captured charge is not an accepted order, an accepted order is not the right item delivered, and a timeout followed by a retry can charge twice. This disclosure describes a vendor-neutral, non-actuating method for checking that an autonomous purchase stays consistent from intent to outcome. Mandatory requirements are separated from preferences; product, variant, quantity, actual seller, merchant of record and payment beneficiary are bound to the approval; the complete obligation is recomputed in integer minor currency units (unit price × quantity + tax + shipping + fees − discounts) and compared with the checkout total, the approvedmaximum and the remaining shared budget. Admission requires the exact pending operation to match the approved checkout revision under current, unrevoked and unused authority; recurring and trial terms need their own authority; an unknown remote disposition blocks retry until non-acceptance is proven. Payment, order, fulfillment and refund outcomes are kept as separate evidence, and three verdicts — preflight, outcome and overall — are computed in three-valued logic so that a contradiction prevails over unknowns and an unknown outcome is never reported as success. The document provides a 60-stage, six-phase commerce workflow and a separate 52-stage payment-specific workflow, each with flowcharts, 72 plus 64 technical invariants, twelve technical variations, 60 engineering checkpoints, a comparison with existing approaches (agentic checkout protocols, network and wallet tokens, 3-D Secure and strong customer authentication, payment idempotency keys, authorization/capture separation, spending controls, marketplace and payee verification, and subscription regulation) and a non-actuating reference evaluator in Python, JavaScript and Go with 89 shared synthetic vectors, all passing. It never orders, pays or handles credentials; the laboratory is synthetic, self-authored and not peer reviewed. The author is open to acquihire or strategic collaboration. Keywords: AI safety; agentic commerce; AI shopping agents; autonomous purchasing; purchase intent; payment authorization; capture; merchant of record; payee verification; checkout integrity; hidden fees; subscriptions and free trials; idempotency; duplicate charges; shared budgets; multi-agent spending; fulfillment verification; three-valued logic