Agentspay

Explainer

Agentic Commerce Protocol: OpenAI Agentic Commerce Protocol, Stripe ACP and Agentic Checkout

ACP is the open standard behind agentic checkout in ChatGPT. It tells a merchant how to sell to an AI agent. It says nothing about whether your agent should have made the purchase.

Agent Payments Console

Pick an agent

Payment intent

intent:

Policy evaluation

Human approval required

This spend is over your approval threshold. Approve it to issue a scoped card, or deny it.

Scoped virtual card issued

Agentspay

single-use

Wallet budget

spent of

Audit trail

In short

The Agentic Commerce Protocol (ACP) is an open standard that defines how an AI agent completes a purchase with a merchant on a buyer behalf. The specification is maintained by OpenAI and Stripe, is published openly on GitHub at agenticcommerce.dev, and covers five building blocks: agentic checkout, cart and feed, delegate payment, delegate authentication, and orders and webhooks. It powers Instant Checkout inside ChatGPT, where a shopper buys without leaving the chat. ACP standardizes the sell side of agent commerce, so it handles how a transaction happens but not whether a given agent was allowed to spend that money.

What the Agentic Commerce Protocol actually standardizes

Before ACP, an AI agent that wanted to buy something had to be integrated store by store, because every merchant exposed a different catalog format and a different checkout. ACP replaces that with one interaction model between three parties: the buyer, the agent acting for them, and the business selling. The current specification, dated 2026-04-17, defines five composable building blocks. Agentic checkout creates, updates and completes a checkout session with cart contents, fulfillment options and payment. Cart and feed let an agent browse a product catalog and assemble an order before checkout. Delegate payment passes a scoped payment token between the buyer, the agent and the business through a payment handler. Delegate authentication uses OAuth 2.0 so an agent can act on a signed-in buyer behalf. Orders and webhooks carry lifecycle updates back: confirmation, shipping, delivery, refunds.

How agentic checkout works, step by step

A shopper asks an agent for something. The agent reads the merchant product feed, finds matching items and opens a checkout session against the merchant ACP endpoint. It adds line items, gets back real shipping options, taxes and totals, and shows the shopper a confirmation surface. When the shopper approves, a delegated payment token is created and handed to the merchant, who charges it through their own processor and their own merchant account. The order is created on the merchant side, and webhooks push status changes back to the agent so the shopper can ask about the package later in the same conversation. The merchant keeps the customer relationship, the payment processing and the fulfillment obligation. The agent is the storefront, not the seller of record.

Delegated payment tokens: the part that matters for risk

The most important design choice in ACP is that the agent never holds raw card data. Instead, a payment handler mints a delegated token that is scoped to a specific purchase and passed to the specific merchant completing it. If that token leaks it is close to useless, because it is bound to one transaction context rather than being a reusable credential. This is the same instinct behind scoped virtual cards, and it is why agentic checkout is meaningfully safer than the alternative people reach for first, which is pasting a corporate card number into an agent config. What the token does not do is decide whether the purchase is a good idea, whether it fits a budget, or whether the agent was manipulated into making it.

ACP, AP2 and x402 solve different layers

These three standards get compared as rivals and they are not. ACP covers checkout: the commercial conversation between an agent and a merchant. AP2, the Agent Payments Protocol led by Google, covers authorization, using cryptographically signed mandates to prove an agent had real user permission to buy within stated limits. x402, introduced by Coinbase, covers settlement, reviving the HTTP 402 status code so machines can pay inline for a resource, typically in stablecoins and typically for tiny amounts. A serious agent payments stack will touch more than one of them, which is why our breakdown of AP2 vs ACP vs x402 treats them as a layer cake rather than a bake-off. A fourth standard now sits above all three: the Universal Commerce Protocol, co-developed by Google and Shopify and announced in January 2026, which covers catalog search, cart, checkout, and order lifecycle in one merchant-hosted specification. UCP is the closest thing ACP has to a direct rival, and most US merchants will end up implementing both, because ACP is how you reach ChatGPT and UCP is how you reach AI Mode in Google Search and the Gemini app.

What a merchant has to implement to support ACP

Three pieces have to exist: a product feed the agent can read, checkout endpoints that follow the ACP specification, and a payment integration that accepts a delegated token. How much work that is depends entirely on your stack. Shopify merchants enable a ChatGPT sales channel from the admin with no code. Stripe and PayPal both ship ACP server support, which covers a large share of US sellers. Everyone else implements the REST endpoints from the open specification on GitHub, or exposes the same capabilities as an MCP server, which the 2026-04-17 release added as an alternative to REST. The card networks are moving in too: both Visa Intelligent Commerce and Mastercard Agent Pay can route ACP-initiated payments with their own tokenized credentials. Instant Checkout in ChatGPT currently requires a US merchant and a US buyer, and OpenAI has publicly confirmed a 4 percent fee on completed Instant Checkout orders, charged to the merchant on top of normal processing costs. We broke the setup and the economics down in our guide to ChatGPT Instant Checkout for merchants, and the wider channel, catalog and fraud work in our readiness guide to agentic commerce for merchants.

The gap ACP leaves open: buy-side spend governance

ACP is written from the merchant point of view. It answers how do I let an agent buy from me. If you are the company whose agents are doing the buying, none of your questions are answered by it. Which agent is allowed to spend, up to what amount, with which counterparties, how often, and who signs off above a threshold are all buy-side decisions that live outside the protocol. That gap gets expensive fast, because an agent that retries a failed order, misreads a quantity, or gets steered by prompt injection will happily complete a perfectly protocol-compliant purchase you never wanted. It is the same problem whether the buyer is a shopping agent filling a cart or a procurement agent renewing a contract. Standards make the transaction possible. Policy makes it accountable.

How Agentspay governs ACP purchases

Agentspay is the rail-neutral control plane that sits in front of whatever standard moves the money. Every agent gets its own funded wallet with hard limits rather than access to a shared corporate card. Each intended purchase is checked against policy before a credential exists: per-transaction ceiling, budget over a window, merchant allowlist, velocity rules. Spend above your threshold pauses for a human approval instead of going through. What the agent receives is a scoped virtual card or a delegated credential valid for that one purchase, and every decision lands in an immutable audit trail tied to the agent, its human owner, the intent and the policy that allowed it. Adopt ACP for checkout, AP2 for mandates, x402 for machine settlement, and keep one place that never moves money without policy.

Stripe ACP: what Stripe actually ships

Stripe is a founding maintainer of the specification alongside OpenAI, and it is also the fastest path to implementing it, which is why stripe acp is one of the most searched forms of this topic. Stripe ships server-side ACP support so a merchant already processing on Stripe does not hand-roll the checkout endpoints, and its Shared Payment Tokens are the delegated credential that carries the purchase. Two published details are worth committing to memory because they bound what an agent purchase can be: the minimum is 0.50 dollars for a card charge via a Shared Payment Token and 0.01 USDC for stablecoin settlement, and stablecoin acceptance covers every US state except New York. In March 2026 Stripe extended Shared Payment Tokens to Mastercard Agent Pay and Visa Intelligent Commerce, along with Affirm and Klarna, so the same token now fronts several networks. The wider Stripe agentic stack, including the Checkout Sessions API and Radar agentic signals, is covered on our Stripe agentic commerce page. Worth being clear about one thing: implementing ACP does not require Stripe. Businesses processing elsewhere can adopt the open specification with their existing payment provider.

Who maintains ACP, OpenAI or Stripe?

Both. ACP was released on September 29, 2025 and is co-maintained by OpenAI and Stripe as founding maintainers, with Meta also contributing to the standard. It is published under Apache 2.0, which means you can implement, fork or extend it without a commercial agreement with either company. The distinction that confuses people is between the protocol and the product. ACP is the open standard. Instant Checkout was OpenAI’s consumer feature built on top of it. Searches for openai agentic commerce protocol and acp openai are usually really asking about one or the other, and the answer differs: the protocol is alive and still being developed, while the specific in-chat buying experience was deprioritized in early March 2026 in favor of purchasing inside merchant apps. OpenAI named ACP as infrastructure it continues to build on at the same time.

Where is the ACP specification on GitHub?

The specification lives in a public repository under the agentic-commerce-protocol organization on GitHub, with documentation at agenticcommerce.dev. It uses date-based versioning rather than semantic version numbers, and as of August 2026 the current stable snapshot is still 2026-04-17, the release that added cart, feed, orders, authentication and the MCP transport option. The repository carries an unreleased directory where in-progress changes land before a dated snapshot, which is the place to watch if you are implementing against it, because a change appears there before it appears in any coverage. The protocol remains in beta. Practical advice for anyone building: pin the dated version in your integration rather than tracking the tip of the repository, and read the OpenAPI definitions and JSON schemas rather than the marketing pages, because the announcement-era write-ups still describe an earlier shape of the standard.

Is ACP still worth implementing in 2026?

Yes, with a clear head about what you are buying. The case against is real: Instant Checkout, the flagship consumer surface, did not work as hoped. Walmart executive Daniel Danker said in-chat purchases converted at roughly one third the rate of click-through, calling the result unsatisfying, and Forrester’s Emily Pfeiffer counted only about 30 Shopify merchants live in February 2026. Those are sobering numbers and anyone selling you ACP as a traffic channel is overselling. The case for is structural. ACP is the standard by which agents reach merchants, it is Apache-licensed and not controlled by a single vendor, and both card networks can now route ACP-initiated payments with their own tokens. The competing standard, the Universal Commerce Protocol from Google and Shopify, covers the same ground for AI Mode in Google Search and the Gemini app, and most US merchants will implement both rather than choose. Treat ACP as table stakes for being purchasable by agents, not as a demand generator.

ACP and the machine payment standards are not the same thing

A distinction worth drawing, because these get filed together and they serve different buyers. ACP is retail checkout: a human wants a product, an agent completes the purchase with a merchant, and there is a cart, a shipping address and a refund path. The machine payment standards handle a different transaction entirely, where software pays software for a resource with no human in the loop and no cart at all. That is what x402 does, and what the Machine Payments Protocol co-authored by Stripe and Tempo does with a broader payment method surface covering stablecoins, cards and Lightning. If your agents are buying goods, you want ACP. If your agents are buying API calls, compute or data, ACP is the wrong layer and you want a machine payment protocol. Plenty of teams need both, and the mistake is assuming one standard covers the other.

Whatever standard moves the money, Agentspay is the rail-neutral control plane that keeps it governed. See how it works and the control surfaces that enforce policy, approvals, and audit on every transaction.

Building blocks

What each ACP building block covers

The five components in the 2026-04-17 specification, and the decision each one leaves to you.

Building block What it does What it does not decide
Agentic checkout Creates, updates and completes a checkout session with cart, fulfillment and totals. Whether this agent should be buying this item at all.
Cart and feed Exposes a merchant catalog so an agent can browse and build an order. Which merchants your agents are allowed to transact with.
Delegate payment Passes a scoped payment token from buyer to merchant via a payment handler. The spend limit, budget window or velocity cap behind that token.
Delegate authentication Uses OAuth 2.0 so an agent can act for a signed-in buyer. Whether a human needs to approve this specific amount.
Orders and webhooks Reports confirmation, shipping, delivery and refund events back to the agent. Attribution of the spend to an agent, owner and policy for audit.

Standard by standard

ACP compared with the other agent commerce standards

What each one covers, who maintains it, and whether it can stop an agent overspending.

Standard Maintainer Layer Licence Enforces a budget?
ACP OpenAI and Stripe Retail checkout with a merchant Apache 2.0 No
UCP Google and Shopify Catalog, cart, checkout and order lifecycle Apache 2.0 No
AP2 FIDO Alliance since April 2026 Authorization via signed mandates Apache 2.0 Per mandate only
x402 x402 Foundation, Linux Foundation Machine settlement on public chains Open No
MPP Stripe and Tempo, IETF track Machine payment negotiation over HTTP 402 Open No

Frequently asked

Questions people ask about the Agentic Commerce Protocol

What is the Agentic Commerce Protocol?

The Agentic Commerce Protocol is an open standard for how an AI agent buys from a business on a shopper behalf. It defines a shared interaction model covering product feeds, checkout sessions, delegated payment tokens, OAuth-based authentication and order webhooks, so an agent can transact with any compliant merchant without a custom per-store integration.

Who created the Agentic Commerce Protocol?

ACP was created by OpenAI and Stripe, with Meta also contributing to the open standard. The specification is maintained by OpenAI and Stripe and published openly, with the OpenAPI specs and JSON schemas on GitHub and documentation at agenticcommerce.dev. Anyone can implement it without a commercial agreement with either company.

Is the Agentic Commerce Protocol open source?

Yes. The specification, OpenAPI definitions and JSON schemas are published in a public GitHub repository under the agentic-commerce-protocol organization, and versioned by release date. The current release is 2026-04-17, which added cart, feed, orders, authentication and an MCP transport option. The protocol is free to implement.

What is the difference between ACP and MCP?

MCP, the Model Context Protocol, is a general standard for connecting a model to tools and data. ACP is specific to commerce: it defines the checkout, payment and order semantics an agent needs to complete a purchase. They are complementary, and since the 2026-04-17 release a merchant can expose ACP capabilities through an MCP server instead of REST endpoints.

Does ACP power Instant Checkout in ChatGPT?

It did, and ACP outlived the feature. Instant Checkout, the buy-in-chat experience OpenAI launched on September 29, 2025, ran on ACP: Etsy sellers went live first, Shopify merchants followed, and PayPal brought additional small businesses through its own ACP server. In early March 2026 OpenAI deprioritized Instant Checkout and moved purchasing into merchant apps, while naming ACP as the infrastructure it continues to build on. The full timeline is on our ChatGPT Instant Checkout page.

What is Stripe ACP?

Stripe ACP refers to Stripe’s implementation of the Agentic Commerce Protocol, which Stripe co-maintains with OpenAI. Stripe ships server-side ACP support so merchants processing on Stripe do not build the checkout endpoints themselves, and Shared Payment Tokens act as the delegated credential. Published minimums are 0.50 dollars for a card charge and 0.01 USDC for stablecoin settlement.

Do I need Stripe to implement the Agentic Commerce Protocol?

No. ACP is an open Apache 2.0 specification and businesses can adopt it with their existing payment provider. Stripe and PayPal both ship ACP server support, which shortens the work considerably for merchants already on those platforms, and Shopify merchants can enable a ChatGPT sales channel from the admin. Everyone else implements the REST endpoints or exposes the same capabilities through an MCP server.

What is the current version of the ACP specification?

The current stable snapshot is 2026-04-17, still the latest as of August 2026. ACP uses date-based versioning rather than semantic versions, and that release added cart, feed, orders, authentication and an MCP transport option alongside the original checkout and delegated payment blocks. The protocol remains in beta. Pin the dated version in your integration rather than tracking the repository tip.

Is ACP the same as the Machine Payments Protocol?

No. ACP handles retail checkout, where a human wants a product and an agent completes the purchase with a merchant, including cart, shipping and refunds. The Machine Payments Protocol, co-authored by Stripe and Tempo, handles software paying software for an API call or a resource with no cart and no human. Teams buying goods want ACP; teams buying compute or data want a machine payment protocol.

Does ACP stop an AI agent from overspending?

No. ACP standardizes how a purchase is completed, not whether it should be. Spend limits, budget windows, merchant allowlists, velocity caps and human approval above a threshold are buy-side controls that sit outside the protocol. If your agents are the ones spending, you need a control plane enforcing policy before any payment credential is issued.

Keep reading

More explainers

ServiceNow AI Control Tower

ServiceNow AI Control Tower

ServiceNow AI Control Tower is the most complete agent inventory and risk console a large US enterprise can buy, and it now reaches across AWS, Google Cloud and Azure. We read the schema ServiceNow ships to developers to answer the one question the rollout meeting always ends on: can it stop an agent from spending money? It cannot, and the reason is written into the data model.

Read

Gemini Enterprise

Gemini Enterprise

Google did something in August 2026 that the other agent platforms have not done: it shipped a hard monthly spend cap that genuinely stops usage instead of emailing you about it. That deserves credit, and it also moves the interesting question one step along. A cap that stops something is only as useful as the thing it is scoped to, so we went and measured what Google can actually point that cap at, in the API model Google publishes for anyone to read.

Read

Salesforce Agentforce

Salesforce Agentforce

Agentforce is the largest agent platform any US enterprise is likely to already own, and it moved to consumption billing, which means the meter now runs on what your agents do rather than on how many seats you bought. That raises a finance question the rollout deck rarely answers: when an Agentforce agent is loose in production, what actually stops it spending. We went and measured the answer in Salesforce own published object model rather than guessing at it.

Read

AWS AgentCore

AWS AgentCore

Amazon shipped the missing piece in August 2026. Bedrock AgentCore Payments went generally available, and it is a real payments product: an agent can now hold a wallet, meet an HTTP 402, pay, and carry on reasoning without a human in the loop. So the question a platform lead has to answer stopped being whether AWS gives agents money and became a narrower, more awkward one: how much of a spend policy did AWS actually ship? We went and measured it, property by property, in the API model AWS publishes.

Read

Microsoft Agent 365

Microsoft Agent 365

Microsoft shipped a control plane for AI agents, and it is a good one. It gives every agent an identity, a registry entry, an owner, a sponsor and a Conditional Access policy. Then somebody in finance asks the obvious follow-up question: fine, but what stops the agent from spending money? This page answers what Agent 365 costs, what it governs, and what we measured when we went looking for a dollar amount anywhere in Microsoft's agent governance surface.

Read

QuickBooks MCP Server

QuickBooks MCP server

Connecting an accounting system to an AI assistant is now a ten minute job. Deciding what that assistant is allowed to do once it is connected is the part nobody writes about, and it is the part your controller will ask about first. This page compares what the official QuickBooks, NetSuite and Xero MCP servers actually hand a model, measured rather than summarized from marketing pages.

Read

Payment MCP Servers

payment MCP servers

Every large payment company shipped an MCP server in the last eighteen months, and almost every write-up of them is a setup tutorial. The setup is the easy part. The question worth answering before you connect one to a production account is narrower and much less comfortable: what, exactly, can the model on the other end of that connection do to your money?

Read

PayPal Agentic Commerce

PayPal Agentic Commerce

PayPal made a bet that most merchants would rather not implement a commerce protocol at all. Where Stripe and OpenAI shipped a spec for you to build against, PayPal shipped two products that sit on top of the checkout you already have, and then bought a company to make the catalog half work. That choice is the whole story: it explains why Agent Ready needs almost no engineering from you, why there is nothing for an agent to discover about your store on the open web, and why the thing PayPal will not do for you is the thing that gets expensive later.

Read

Shopify Agentic Commerce

Shopify Agentic Commerce

Shopify switched agentic commerce on by default, so your store is probably already selling to AI assistants whether or not anyone on your team configured it. Instead of restating the announcement, we checked something you can check too: on September 2, 2026 we requested the machine-readable capability file that Shopify publishes for real storefronts, on fourteen well-known US brand domains, and read what it exposes to an agent. Eleven answered correctly. The three that did not share one trait, and it is quietly costing them agent traffic.

Read

Web Bot Auth

Web Bot Auth

Web Bot Auth is the reason your agent either gets served or gets throttled with the scrapers. Almost everything written about it repeats the same architecture diagram, so we did something different: on September 1, 2026 we fetched the published key directories of more than twenty major AI operators and infrastructure vendors to see who is genuinely signing their traffic. Four were. The results are in the first table.

Read

Tempo Blockchain

the Tempo blockchain

Tempo is the payments chain Stripe and Paradigm built, and it shipped with a protocol that lets software pay for things on its own. It settles machine payments in under a second. It has nothing at all to say about whether your agent should have paid.

Read

AI Agent Governance

AI agent governance

Every agentic AI governance framework published so far governs the same four things: identity, tools, data and prompts. Not one of them carries a budget. Here is what the real frameworks say, which guardrails actually bind at runtime, and what to do about the last mile none of them reach.

Read

A2A Protocol

A2A Protocol

Most explanations of the A2A protocol stop at the sentence that agents can now talk to each other. That was true in April 2025 and it is no longer the interesting part. A2A shipped version 1.0 in April 2026 under Linux Foundation governance, it runs in production inside Azure AI Foundry and Amazon Bedrock AgentCore, and the questions engineers actually get stuck on are narrower: what an Agent Card commits you to, when to reach for MCP instead, and what happens the first time one of your agents has to pay another one for the work. That last question has a specific answer, and it is not in the core spec.

Read

Mastercard Agent Pay

Mastercard Agent Pay

Nearly every article about Mastercard Agent Pay is a retelling of one press release from April 2025, the one where Mastercard said AI agents would be able to shop with Agentic Tokens and named Microsoft as the first platform. That was sixteen months ago, and four more things have shipped since. Reading only the launch coverage leaves you with roughly a quarter of the picture, and the missing three quarters are the parts that decide whether you can actually put this into production.

Read

Visa Intelligent Commerce

Visa Intelligent Commerce

Almost everything written about Visa Intelligent Commerce is a retelling of the April 2025 announcement, when Visa said AI agents would be able to pay with a Visa credential. Three more things have shipped since, including an open agent-identity protocol built with Cloudflare that most coverage does not mention at all. This page is the current version, checked against Visa’s own developer documentation and newsroom in August 2026.

Read

Stripe agentic commerce

Stripe agentic commerce

Most writing about Stripe and agentic commerce is still a retelling of the September 2025 launch week, when Stripe and OpenAI shipped Instant Checkout and published the Agentic Commerce Protocol together. Stripe has built a good deal more since then, and some of it points in a direction the launch coverage never anticipated. This page is the current version, checked against Stripe’s own documentation in August 2026.

Read

ChatGPT Instant Checkout

ChatGPT Instant Checkout

Almost every guide to ChatGPT Instant Checkout still reads like it was written the week it launched, walking merchants through how to apply and what the fee will be. OpenAI changed course in March 2026. Here is the accurate version: what Instant Checkout was, what the numbers actually looked like, what replaced it, and which parts of the stack are still very much alive.

Read

Google AP2

Google AP2

Most guides to Google AP2 still describe an Intent Mandate and a Cart Mandate, because most of them are rewrites of the September 2025 launch post. The specification moved. Here is what the Agent Payments Protocol actually defines today, and the one question it deliberately does not answer.

Read

Human in the loop AI

Human in the Loop AI

Every guide to human in the loop AI describes the same shape: the agent pauses, a person decides, the agent continues. The shape is right. What almost none of them ask is a harder question, which is where the pause is enforced, because a pause written into the agent's own code is a pause the agent is trusted to honor.

Read

AI agent cost

AI Agent Cost

Every cost guide for AI agents answers the same two questions: what does it cost to build, and what does it cost to run. Both are answerable, and both are on somebody's invoice. The third question is the one that ends up in a variance report, because the agent also spends your money, and nobody sends you a bill for that.

Read

Agentic checkout

Agentic Checkout

Nearly every guide to agentic checkout is written for the merchant who wants to receive these orders. Far fewer are written for the company whose agents are placing them, which is odd, because agentic checkout quietly removes the one screen where spending used to get a second look.

Read

API monetization

API Monetization

Most guides to API monetization argue about which pricing model wins. The harder question in 2026 is who is calling. An API priced for a signed-up developer with a key behaves very differently when the caller is an agent that showed up once, wants one record, and has no account.

Read

x402 protocol

x402 Protocol

x402 took the one HTTP status code the web never used and turned it into a payment rail machines can drive. The protocol is elegant and genuinely small. The part it deliberately leaves to you is the budget.

Read

AI procurement agents

AI Procurement Agents

Every major procurement suite shipped agents during 2026. Almost none of them answer the question your controller will ask first, which is what happens when the agent is wrong about a purchase and the money has already moved.

Read

Agentic payments

Agentic Payments

Agentic payments move money with no human at the checkout. The rails to do it all shipped during 2026. The part most teams have not solved is deciding, before the money moves, whether the agent was allowed to spend it.

Read

AI agent monetization

AI Agent Monetization

Every AI agent company is rewriting its price list. The models that survive are metered. The ones that quietly fail are the ones where nobody measured what a single task costs to serve.

Read

Agent payment platforms

AI Agent Payment Platforms

Five different kinds of product now call themselves an AI agent payment platform, and they solve five different problems. Picking the wrong category is the expensive mistake, not picking the wrong vendor inside a category.

Read

Universal Commerce Protocol

the Universal Commerce Protocol (UCP)

Google and Shopify shipped UCP as an open standard so an AI agent can check out at any merchant that supports it. Here is what the specification actually defines, where it is live for US buyers, and the one thing it deliberately leaves to you.

Read

MCP Payments

MCP Payments

MCP payments are how an AI agent discovers a payment tool and calls it to move money. The catch: the Model Context Protocol carries the tool call, not the spending decision, so nothing in the stack asks whether the purchase should have happened.

Read

Visa Intelligent Commerce vs Mastercard Agent Pay

Visa Intelligent Commerce vs Mastercard Agent Pay

Visa Intelligent Commerce and Mastercard Agent Pay are the two big card networks racing to let AI agents pay. They take different routes to the same idea, and neither one decides whether a given purchase should have happened.

Read

AP2 vs ACP vs x402

AP2 vs ACP vs x402

AP2, ACP, and x402 are the three standards shaping how AI agents pay. They solve different layers of the problem, and most real systems will touch more than one.

Read

Machine payments protocol

Machine payments protocol

As software starts paying software, machine payments protocols define how value moves without a human at the keyboard. The harder question is how to keep that spending governed.

Read

Know Your Agent (KYA)

Know Your Agent

KYA, or Know Your Agent, extends the idea of customer due diligence to autonomous software. When an agent spends, you need to know which agent, on whose authority, and under what limits.

Read

Keep agent spending governed

Add policy, hard limits, human approval, and an immutable audit trail across any protocol or rail. Start in the sandbox today.

Never moves money without policy