Magento/Adobe Commerce and AI Agents: Platform Guide for Merchants

Magento (now branded as Adobe Commerce in its paid version) is the enterprise-grade open-source e-commerce platform. It powers some of the largest online stores in the world, including brands that process millions of orders annually. When it comes to AI readiness, Magento occupies a unique position: it has the technical depth to support anything, but nothing comes configured out of the box. For Magento merchants, AI readiness is a development project, not a toggle.

Platform Overview

Magento exists in two forms. Magento Open Source (formerly Community Edition) is the free, self-hosted version. Adobe Commerce (formerly Magento Commerce/Enterprise) is the paid version with additional features, cloud hosting options, and Adobe’s AI tools (Adobe Sensei).

The platform is built on PHP with a complex, modular architecture. Magento stores are typically run by mid-market to enterprise businesses with dedicated development teams or agency partners. The average Magento store is larger and more complex than stores on Shopify or WooCommerce, with bigger catalogs, more sophisticated pricing rules, and multi-store, multi-language configurations.

This complexity is both an asset and a liability for AI readiness. Magento can be configured to output perfect structured data, serve optimized product feeds, and even implement custom protocol endpoints. But none of this happens by default. Every AI readiness feature requires deliberate development work.

Adobe has been integrating AI features into Adobe Commerce, primarily through Adobe Sensei (now being rebranded under Adobe’s generative AI initiatives). These features focus on personalization, search, and merchandising within the store itself - not on external AI agent discovery. The gap between Adobe’s internal AI features and external AI agent compatibility is significant.

AI Agent Compatibility

Magento does not have native support for ACP, MCP, or UCP. AI agents interact with Magento stores through the same channels as any other website:

1. Structured data on product pages. Magento themes can output JSON-LD, but the default Luma theme provides only basic structured data.
2. Magento REST and GraphQL APIs. Magento has robust APIs, but they require authentication for most product data access. The GraphQL API is powerful but not designed for public AI agent consumption.
3. Product feeds. Extensions generate product feeds for Google Shopping, Facebook, and other channels.

Magento’s API layer is actually one of the most capable in e-commerce. The GraphQL API supports complex queries across products, categories, pricing, and inventory. If a protocol adapter were built on top of this API - exposing a public-facing MCP or ACP endpoint - Magento could have best-in-class AI agent support. This is technically feasible but requires custom development.

For now, Magento stores appear in AI shopping results only when they have strong organic SEO, good structured data, and comprehensive product content. The platform’s technical capabilities are not the bottleneck - configuration and implementation are.

Structured Data Support

Magento’s default structured data output through the Luma theme is minimal. It includes basic product name, price, and availability, but misses many fields that AI agents rely on.

For comprehensive structured data, Magento merchants typically use:

• Third-party extensions from the Magento Marketplace (Amasty, Mageplaza, and others offer rich snippets/structured data extensions)
• Custom theme development to add JSON-LD directly to product templates
• Adobe Commerce’s built-in features (the paid version includes some enhanced structured data capabilities)

Magento’s product data model is one of the most comprehensive in e-commerce. The platform supports:

• Configurable products with variant-level attributes
• Grouped and bundled products
• Custom attributes (unlimited)
• Multiple images with roles (base, small, thumbnail, swatch)
• EAN, UPC, and custom identifier fields
• Tiered pricing and customer group pricing
• Multi-source inventory (MSI)

The data is there. The challenge is getting it into structured data format on the frontend. Many Magento stores have rich product data in their backend that never makes it into JSON-LD because the theme doesn’t output it.

Protocol Support

Protocol	Status	Notes
ACP (Agentic Commerce Protocol)	Not supported	No native or extension support. Custom development possible.
MCP (Model Context Protocol)	Not supported	Could be built on Magento’s GraphQL API. No standard extension exists.
UCP (Universal Commerce Protocol)	Not supported	No integration available.
JSON-LD / Schema.org	Extension required	Default theme output is minimal. Extensions or custom development needed.
robots.txt	Full control	Fully customizable via admin panel or file system.
llms.txt	Manual	Can be added to the pub/static directory or web root.
ai.txt	Manual	Can be added to the web root.

Magento’s technical architecture could support any protocol - the GraphQL API and extension system are flexible enough. The gap is that no one has built standardized, widely-adopted protocol extensions yet. This will likely change as agentic commerce matures, but for now, Magento merchants must rely on structured data and product feeds.

Optimization Checklist

• Install a structured data extension. This is the highest-impact step. Choose an extension that outputs comprehensive Product schema including brand, identifiers, reviews, and variant data.
• Map your product attributes to schema properties. Magento’s custom attribute system is powerful but disconnected from structured data by default. Ensure your structured data extension maps attributes like brand, material, color, and size to their Schema.org equivalents.
• Fill in product identifiers. SKU is usually populated, but EAN/GTIN, UPC, and MPN fields are often empty. Fill these in for every product. Bulk import tools can help for large catalogs.
• Write detailed product descriptions. Magento supports short description and full description. Both should be comprehensive. For AI agents, the full description is particularly important - include specifications, use cases, comparisons, and materials.
• Configure your product feed. Install a feed extension and generate Google Shopping feeds. Ensure the feed includes all product attributes, not just the minimum required fields.
• Expose your catalog through a public endpoint. If you have development resources, consider creating a public GraphQL endpoint or REST API that serves product data without authentication. This makes your catalog directly accessible to AI agents.
• Add llms.txt and ai.txt files. Place these in your web root to guide AI crawlers. For Magento, this is typically the pub/ directory or the document root configured in your web server.
• Review your robots.txt. Magento’s default robots.txt can be restrictive. Ensure product pages, category pages, and your sitemap are accessible to AI crawlers.
• Handle multi-store carefully. If you run multiple store views (common in Magento), ensure each store view outputs correctly localized structured data with the right currency, language, and availability.
• Monitor page load performance. Magento stores can be slow, and AI crawlers have timeouts. Ensure your product pages load quickly enough for AI agents to crawl efficiently.

Related Terms

• JSON-LD - The structured data format for communicating product information to AI agents.
• Structured Data - Machine-readable markup that helps AI systems understand page content.
• Product Feed - Standardized product data exports consumed by shopping channels and AI systems.
• AI Visibility Score - A measure of how visible your products are to AI shopping agents.
• Product Schema - Schema.org vocabulary for describing products in structured data.