Skip to content
SSyntalith
PLEN
Book a review

Blog · Published · Updated

MCP, A2A, UCP, ACP and AP2: AI agent protocols compared

Short answer

MCP connects an agent to tools and data, A2A to other agents, UCP and ACP describe shopping in a store, and AP2 secures a payment an agent makes on your behalf. All are open standards at different stages. We build an MCP server with your catalogue and availability from €3,000 net, and a readiness audit costs from €600 net.

How we workDetails ↓

  1. 01Call30 minutes with an engineer, free
  2. 02PrototypeA working system on your process and data, free
  3. 03ImplementationTarget and fixed price in the contract; 50% at signing
  4. 04Acceptance50% once the agreed target is met
net per agent for one process, paid once
from €3,000
from signing the contract to acceptance
6–16 weeks
free prototype on your cases
€0
Book a free process review (30 min)

We belong to the OpenAI and Anthropic partner programmes and learn through their official courses. What this means ↓

Author: Syntalith · 3 min read

Example: the morning request inbox
Agent · ready

Statistics

Agent protocols: statistics 2026

5

AI agent protocols in October 2026: MCP for tools, A2A between agents, UCP and ACP for shopping, AP2 for payments. A2A has been at version 1.0 since 12 March 2026.

Figures checked: 22 October 2026

The AI agent protocols at a glance

ProtocolWho develops itWhat it is forStatus in October 2026
MCP (Model Context Protocol)Agentic AI Foundation at the Linux Foundation, created by Anthropicgives an agent tools and datawidely used in assistants and company tools
A2A (Agent2Agent)Linux Foundation, created by Googlepassing tasks between agents of different companiesversion 1.0 since 12 March 2026
UCP (Universal Commerce Protocol)Google and Shopifyproduct discovery, basket, checkout and after-salesbuying in AI Mode and Gemini for offers from the US, Canada and Australia
ACP (Agentic Commerce Protocol)OpenAI and Stripe, betaproducts and promotions in ChatGPTsince 24 March 2026 focused on product discovery
AP2 (Agent Payments Protocol)FIDO Alliance, created by Googlesigned confirmations of the customer's consent to payhanded to the FIDO Alliance in April 2026

MCP and A2A: tools and other agents

MCP standardises the way an AI assistant uses tools: a database, a catalogue, a calendar or an order system. A company builds one MCP server that different assistants use. Since 9 December 2025 the Agentic AI Foundation at the Linux Foundation has developed MCP. In the browser version, WebMCP, a website offers tools to an agent in the user's browser. Shopify launched such tools for checkout in September 2026, as the article on Shopify WebMCP describes.

A2A helps when the other side takes its own decisions. An agent publishes an Agent Card describing its skills and how to authenticate, and other agents give it tasks. The version 1.0 announcement describes signed Agent Cards, support for many tenants on one endpoint and a technical committee with representatives of 8 companies. According to that announcement, A2A complements MCP: MCP works at the level of one agent's tools, A2A at the level of agents working together.

How the protocols fit into one purchase
  1. 01Customerasks an assistant for a product within their budget
  2. 02UCP or ACPthe assistant finds the product in the shop's data
  3. 03MCPa tool checks availability and delivery date
  4. 04A2Athe shop's agent passes a task to the carrier's agent
  5. 05AP2the customer signs a payment mandate

UCP and ACP: shopping in Google and ChatGPT

Google and Shopify announced UCP on 11 January 2026. A shop publishes a profile at /.well-known/ucp listing the features it supports, for example checkout, discounts and delivery. The agent fetches the profile, agrees the shared capabilities and runs the purchase through a REST API, MCP or A2A (Google Developers Blog). Buying in AI Mode and Gemini requires a Merchant Center account (Merchant Center help).

OpenAI and Stripe released ACP in September 2025, together with buying without leaving ChatGPT. On 24 March 2026 OpenAI extended ACP to product discovery: merchants pass product feeds and promotions through it, also through providers such as Salesforce and Stripe. The specification has dated versions in the ACP repository. Preparing the feed is covered in shopping in ChatGPT.

AP2: a payment an agent makes for you

AP2 answers the question of who is responsible for a purchase made by an agent. The customer signs a digital confirmation, a mandate, with the basket and the amount, or with the terms of a future purchase, for example a price limit. The bank, the payment operator and the merchant receive proof of consent, which helps with disputes (Google Cloud). In April 2026 Google handed AP2 to the FIDO Alliance (Google). For a shop, AP2 is today a matter for the payment operator.

Which protocol matters for your company

Where to start

  • A shop on Shopify

    Agentic Storefronts settings and the quality of product data

    configuration

  • A shop on another platform, selling in the EU

    a feed in Merchant Center, data on the site, measuring traffic from assistants

    audit from €600

  • A wholesaler or a B2B manufacturer

    an MCP server with catalogue and availability, A2A with a specific partner

    from €3,000

  • A service company

    an MCP server with the offer and available dates, a clear page with prices

    from €3,000

Before you build an integration
  • Are the product or service data on the site complete and current
  • Which operations the agent may perform and which it may only read
  • Who approves an order or a booking placed by an agent
  • How you measure traffic and sales from AI assistants

According to Gartner (2025), of the thousands of agentic AI vendors only about 130 offer real agentic solutions, so the name of a protocol in a vendor's offer proves nothing yet. An example of the work on our own website is in syntalith.ai, ready for AI agents, and preparing a store in agentic commerce. We build interfaces for agents as part of AI agents. Status as of 11 October 2026, with A2A and ACP checked again on 22 October 2026.

Worked example · Scenario

How we count the payback: an online shop

We run the same count for your shop in the free process review.

€2,792

a month for handling 800 orders that bypass the checkout

Payback: 1.3 mo.

6 minutes to retype an order or answer a question and checking in 15% of the old time are our assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat, for companies with 10 or more employees. Model cost: reading the order in our system (about €0.0006) plus matching the lines (Claude Sonnet 5.5, about 3,000 + 500 tokens), at the ECB rate of 1.1205 USD per euro on 22 October 2026.

See the full calculation →

This is how we implement it in companies

All case studies →

Client results.

All projects →
foap.comcustomer service

foap.com customer service: 95% of replies are sent automatically.

Users from several countries write to foap.com, in several languages. The system reads the entire conversation and the terms of that person, and then sends a response in that person's language. Questions about money and exceptions go to the team with a ready draft, so they have to check and send them.

the system sends the answers itself, without human typing
95%
time from the arrival of the e-mail to the ready response (median)
30 s
e-mails that require a response receive a ready-made draft
100%
less work for every 1000 answers (estimate)
47 h

Data from foap.com production, October 20-21, 2026. Estimate: 3 minutes for handwritten response.

How foap.com customer service works →
Sample messages. Money matters always come to the person.

Other implementations

jaktako.design

studio projektowe

−87%

Time to look for new customers, every day

4 hours → 30 min

approx. 73 hours less per month (21 days of work)

according to the studio

How the agent loop works →
Signature Estates

real estate

approx. −90%

Operator time spent browsing portals, per day

2–3 hours → 20 min

approx. 45 hours less per month (21 days of work)

respect · contact the seller 5 minutes after the announcement

How the agent finds property listings →
Kuba Koziej, CEO and co-founder of MoreGrowth and board member at Natu.Care

Kuba Koziej

CEO and co-founder, MoreGrowth

„We started with my own AI agent. During the training that followed, I learned to build new applications myself. Today I am bringing this way of working with AI into my business. Working with Syntalith gave me exactly the skills and tools I needed.”

View the original training review on LinkedIn ↗

See the agent on your own cases before you commit.Book a free process review (30 min) →

How we build automation and agents

From a conversation to a working system in four steps.

  1. 01 · 30 min

    Free process review

    €0

    You're talking to an engineer. You receive the written plan within 2 business days.

  2. 02 · 5 days

    Free prototype

    €0

    The system works on 20-100 cases that your team has already handled. You see every result.

  3. 03 · 2–16 weeks

    Build

    50%

    Fixed price and goal written in the contract. First installment upon signing.

  4. 04 · after reaching the goal

    Reception

    50%

    You pay the second installment when the system reaches its target on your cases.

How a free prototype is created

Mon.

We set a goal

What result do you consider good?

Tue–Wed

The system is working

It goes through your affairs, one by one.

Thu

We compare

We check every difference with the team's work.

Fri

Result and price

List of errors and fixed implementation price.

You choose the matters. We anonymize them upon request, they stay in the EU and we delete them after the test.

On the market, such a test usually costs from PLN 3,000 to PLN 8,000.

Book a free process review (30 min)

We establish a prototype during a free analysis.

Data checked on October 21, 2026.

„Is my data safe?”

Yes. You choose where they go: from your company account to the model on your server.

  1. 00Employee's private account

    Customer and company data ends up in an account that the company cannot see or control. There is no data processing agreement, so it is a violation of the GDPR. When an employee leaves, the chat history goes with them.

  2. 01ChatGPT or Claude corporate account

    The provider does not use your data to train models. A copy of the queries remains in his logs for up to 30 days.

  3. 02API with zero retention (ZDR)

    The provider processes the request and does not save anything. It requires his consent, we prepare the application together.

  4. 03Processing in the EU

    Queries are counted and stored in Europe: OpenAI in the EU region, Claude via clouds with the EU region.

  5. 04Model at your place

    The open model runs on your server or in your cloud in the EU. The data does not go outside the company.

What we do with every implementation

  • Confidentiality agreement and data processing agreement before we receive any data.
  • You choose the cases for the prototype and we anonymize them upon request.
  • The data stays in the EU and we delete it after the test.
  • Corporate accounts with an administrator on your side.

When data cannot leave the company

The open model runs on your server or in your cloud in the EU. To choose from:

White-tailed eagle
Polish model, SpeakLeash and ACK Cyfronet AGH
PLLUM
Polish model, HIVE consortium commissioned by the Ministry of Digitization
Qwen
Alibaba
GLM
Zhipu AI
DeepSeek
DeepSeek

We select a model for the task and check it on your cases. We run it, tune it and maintain it.

Source: Bielik and PLLuM, ACK Cyfronet AGH ↗ · PLLuM, Ministry of Digitization ↗

Do you have a lot of your own data? We will fine-tune the small model.

  1. 1 Your examples
  2. 2 tuning
  3. 3 test on new cases
  4. 4 model at your place

When you have thousands of similar emails, documents or reports, we teach a small open model based on your examples. It works for you, responds quickly and costs less to maintain than a large model. We measure quality on issues that the model has not seen while learning.

Przemysław and Piotr did it in large programming and consulting companies. Artem did it before and does it today at Syntalith. It's about models on your own infrastructure, tuning and quality measurement.

Private AI models →

Source: OpenAI, training on data from private accounts ↗ · OpenAI, who sees conversations ↗ · Anthropic, changes for private accounts (August 2025) ↗ · OpenAI, data in API ↗ · AWS Bedrock, EU profile ↗

You choose the protection level during the interview. We sign the data processing agreement before we see any data.Schedule a free analysis →

AI agents

See the agent on your own cases before you commit.

The agent takes the case through the next steps in your systems, and the actions take effect until the consent of the indicated person is obtained. A prototype for 20-100 cases that your team has already handled costs €0. We write down the purpose and test set in the contract.

See AI agents →
net per agent for one process, paid once
from €3,000
from signing the contract to acceptance
6–16 weeks
free prototype on your cases
€0
Book a free process review (30 min)

or call+48 888 78 48 78WhatsAppMon–Fri 9:00–17:00 CET

Questions

  • How does MCP differ from A2A?

    MCP gives an agent tools and data, for example a catalogue or a calendar. A2A passes tasks between agents, when the other side takes its own decisions and needs time to answer.

  • Which protocol matters for a shop selling in the EU?

    Buying through UCP in AI Mode and Gemini currently covers eligible offers from the US, Canada and Australia. In the EU, start with a feed in Merchant Center, product data on your site and measuring traffic from assistants.

  • What is AP2?

    A protocol for payments an agent makes on your behalf. The customer signs a mandate with the basket and the amount, or with the terms of a future purchase, and the bank and the merchant receive proof of consent. Since April 2026 the FIDO Alliance develops it.

  • How much does an MCP server for a company cost?

    At Syntalith from €3,000 net, for example a server with your catalogue and product availability. An agent readiness audit costs from €600 net and takes 5 working days.

Let's talk

Just write, call or choose a date.

We work fast because we hate corporate cotton wool. We don't have salespeople or managers to keep an eye on managers. You talk immediately to engineers who then build your system. We respect each other's time and yours too.

We will accept 3 more companies by the end of the year.

Write
contact@syntalith.ai
Call
+48 888 78 48 78
WhatsApp
write on WhatsApp

Mon–Fri 9:00–17:00 CET

What our OpenAI and Anthropic partnerships mean for you.

OpenAI makes ChatGPT and Anthropic makes Claude. We belong to the partnership programs of both companies, so we work with their tools every day and learn from the creators themselves.

OpenAI Select PartnerClaude Partner Network
01We will select a tool for the task
We implement both ChatGPT and Claude. We check both on your documents, and you choose the one that makes fewer corrections and saves more time.
02Knowledge straight from the creators
We learn from official OpenAI and Anthropic courses and materials. Your team learns about the features and settings that work today.
03Licenses without our margin
You buy accounts directly from OpenAI or Anthropic and receive an invoice from them. We do not add anything to the license.
04Certified engineers
On the team, we have an AWS certification in machine learning and an Anthropic certification in systems architecture on Claude.

Participation in programs does not mean that OpenAI or Anthropic endorses our services.