# Guides > How AI agents find, read and buy from online stores, and what to change on yours. Start with the agent-ready ecommerce guide if you are new to the topic. - Source: https://is-agentic-storefront.com/guides - [Agent-ready ecommerce: how to make your store work with AI agents](https://is-agentic-storefront.com/guides/agent-ready-ecommerce): An agent-ready (agent-friendly) ecommerce store is one AI agents can reach, read and operate: AI search bots and shopping agents aren't blocked, product facts are in the server HTML and structured data, key actions are real labelled controls, and ideally a protocol like UCP lets agents build carts and check out through an API. - [Agentic commerce protocols explained: UCP, ACP, MCP, A2A, AP2, WebMCP](https://is-agentic-storefront.com/guides/agentic-commerce-protocols): Agentic commerce protocols let AI agents discover a store, build carts, check out and pay. UCP (an open standard published at ucp.dev) is the shipped standard with a public discovery file; MCP is the usual transport; ACP (OpenAI and Stripe) is feed-based and invisible from a URL; A2A, AP2 and WebMCP are early-stage. - [AI crawler and agent user agents: the 2026 reference for stores](https://is-agentic-storefront.com/guides/ai-user-agents): AI bots fall into three classes: training crawlers (GPTBot, ClaudeBot, Google-Extended), AI search bots (OAI-SearchBot, Claude-SearchBot, PerplexityBot) and user-initiated agents (ChatGPT-User, Claude-User, Perplexity-User). Stores can block training crawlers without losing AI visibility, but blocking search bots or user agents hides products from AI answers and AI shopping agents. - [How the agent-readiness score works: methodology and limits](https://is-agentic-storefront.com/guides/scoring-methodology): The agent-readiness score is a 0–100 number computed from 29 checks run over plain HTTP against a store URL, its homepage and up to three product pages. Each check passes, partially passes or fails; checks are weighted High, Medium or Low within five categories, and categories are weighted into one overall score graded A to F. - [Agent readiness vs SEO: what's different for ecommerce](https://is-agentic-storefront.com/guides/agent-readiness-vs-seo): Agent readiness builds on ecommerce SEO but goes further. Both need crawlable pages, sitemaps, server-rendered content and Product structured data. Agent readiness adds robots and WAF rules for AI bots, pages browsing agents can operate, and machine interfaces such as UCP profiles, MCP endpoints, Markdown pages, llms.txt and checkout permalinks. - [Is Shopify agent-ready? What you get by default and what to fix](https://is-agentic-storefront.com/guides/shopify-agent-readiness): Shopify-hosted stores include several agentic-commerce files by default: a UCP profile, a shopper MCP endpoint, OAuth metadata, llms.txt and a sitemap. What Shopify does not guarantee is theme-level product data, such as complete Product JSON-LD and return and shipping structured data, or that no app or CDN blocks AI agents. - [Is WooCommerce agent-ready? Defaults and what to fix](https://is-agentic-storefront.com/guides/woocommerce-agent-readiness): WooCommerce core provides server-rendered product pages, Product JSON-LD, a core sitemap, add-to-cart forms and add-to-cart URLs. It does not output return or shipping structured data or ProductGroup variants, and it has no shopper-facing UCP profile or agent endpoint yet. WooCommerce lists UCP as coming soon. - [Is BigCommerce agent-ready? Defaults and what to fix](https://is-agentic-storefront.com/guides/bigcommerce-agent-readiness): BigCommerce Stencil storefronts render on the server, ship Product JSON-LD in the Cornerstone theme, and generate an XML sitemap. They lack return and shipping structured data, ProductGroup variants and llms.txt. UCP support runs through an early-access Google agentic checkout program, and Catalyst proxies /.well-known/ucp to BigCommerce. - [Is Medusa agent-ready? Build an agent-ready storefront you control](https://is-agentic-storefront.com/guides/medusa-agent-readiness): With Medusa, you own the storefront, so every agent-readiness fix is a change in your own code, not a theme or app workaround. The Next.js Starter server-renders product pages; Product JSON-LD, a sitemap, robots.txt, llms.txt and a form-based add-to-cart are short additions. Medusa has an official tutorial for OpenAI's Agentic Commerce Protocol. UCP and a shopper-facing MCP server can be built but are not included. - [Agentic commerce glossary](https://is-agentic-storefront.com/glossary) ## Run your store on a platform built for agentic commerce Medusa is open-source, headless commerce with a full Store API agents can call and storefronts you control end to end. ### Deploy on Medusa Cloud Open-source commerce with a full Store API, hosted and scaled for you. Your storefront, your data model. - [Deploy on Medusa Cloud](https://cloud.medusajs.com?utm_source=agent-friendly-scorer&utm_medium=referral&utm_campaign=markdown-cta&utm_content=%2Fguides) - [Medusa Cloud docs](https://docs.medusajs.com/cloud?utm_source=agent-friendly-scorer&utm_medium=referral&utm_campaign=markdown-cta&utm_content=%2Fguides) ### Storefront skills for coding agents Teach Claude Code, Cursor or Codex to build storefronts with the patterns these checks look for. ```bash npx skills add medusajs/medusa-agent-skills ``` [View the skill on GitHub](https://github.com/medusajs/medusa-agent-skills/tree/main/plugins/ecommerce-storefront/skills/storefront-best-practices) --- Scan your store for free at https://is-agentic-storefront.com/