Agentic Commerce Architecture

Overview

Agentic commerce represents a fundamental shift in how consumers discover and purchase products. Instead of browsing websites directly, consumers increasingly delegate shopping tasks to AI agents like ChatGPT, Perplexity, Google AI Mode, and specialized shopping assistants.

This architecture document explains the technical layers involved in making ecommerce stores "agent-ready" — visible, understandable, and transactable by AI systems.

The Three Layers

Layer 1: Discovery

Before an AI agent can recommend your products, it needs to know you exist. The discovery layer encompasses all the ways agents find and index your store.

Key Components

• Robots.txt Configuration

  Explicitly allow AI crawlers (GPTBot, PerplexityBot, etc.) to access your product pages.

• XML Sitemaps

  Comprehensive, well-structured sitemaps that prioritize product pages and include last-modified dates.

• UCP Manifest

  The Universal Commerce Protocol manifest declares your store's capabilities to AI agents.

User-agent: GPTBot
Allow: /products/
Allow: /collections/
Disallow: /checkout/
Disallow: /account/

User-agent: PerplexityBot
Allow: /products/
Allow: /collections/

User-agent: Google-Extended
Allow: /

Layer 2: Comprehension

Once an agent discovers your store, it needs to understand what you sell. This is where structured data becomes critical. AI agents parse machine-readable schemas to build accurate mental models of your products.

Schema.org Markup

The foundation of comprehension is Schema.org Product markup. This tells agents the essential facts: name, description, price, availability, brand, reviews, and specifications.

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Wireless Noise-Canceling Headphones",
  "description": "Premium over-ear headphones with...",
  "brand": {
    "@type": "Brand",
    "name": "AudioTech"
  },
  "offers": {
    "@type": "Offer",
    "price": "299.99",
    "priceCurrency": "USD",
    "availability": "https://schema.org/InStock",
    "priceValidUntil": "2026-12-31"
  },
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.7",
    "reviewCount": "2847"
  }
}

Beyond Basic Schema

• •Product variants — Colors, sizes, configurations
• •Shipping information — Delivery times, costs, regions
• •Return policies — Duration, conditions, process
• •FAQ schema — Common questions about products

Layer 3: Transaction

The most advanced layer enables AI agents to complete purchases on behalf of users. This is still emerging, but protocols like ACP (Agentic Commerce Protocol) are defining how this will work.

What Transaction Enablement Includes

• •Cart APIs — Programmatic cart creation and management
• •Checkout delegation — Secure checkout initiation by agents
• •Payment handling — Integration with user payment credentials
• •Order tracking — Status updates back to agents

Implementation Priority

Not all layers are equally urgent. Here's how we recommend prioritizing: