E-commerce MCP servers let AI agents interact with online stores — managing products, processing orders, searching catalogs, handling inventory, and powering AI-driven shopping experiences. Instead of clicking through admin dashboards or writing API calls manually, you describe what you want and the agent executes it against your store’s backend.

We’ve researched 30+ e-commerce MCP servers across the full landscape. This guide synthesizes all of that into one page: what’s worth using, what’s not, and where the gaps are.

Note: Our recommendations are based on documentation review, GitHub analysis, and community feedback — not hands-on testing of every server. Star counts were verified in March 2026.

The short version

Category Our pick Stars Runner-up
Shopify (storefront/shopping) Shopify Storefront MCP Official Shopify Dev MCP (414 stars, developer docs)
Shopify (admin/store management) GeLi2001/shopify-mcp 139 antoineschaller/shopify-mcp-server (22 tools)
WooCommerce techspawn/woocommerce-mcp-server 26 iOSDevSK/mcp-for-woocommerce (WordPress plugin)
Magento / Adobe Commerce boldcommerce/magento2-mcp 27+ elgentos/magento2-dev-mcp (dev assistant)
Amazon (seller) Seller Labs MCP — mattcoatsworth/AmazonSeller-mcp-server (SP-API)
BigCommerce BigCommerce Storefront MCP Beta CData BigCommerce MCP (read-only)
Headless (Saleor/Medusa) saleor/saleor-mcp Official minimalart/mcp-medusa (14 admin tools)

Why e-commerce MCP servers matter

E-commerce operations involve repetitive, data-heavy tasks across multiple systems. MCP servers turn these into natural-language commands.

The value comes in three forms:

  1. Store management automation. “Update the price of SKU-1234 to $29.99 and reduce inventory by 50” — instead of navigating to the product page, finding the variant, and making the change. At scale, agents can process bulk operations that would take hours manually.
  2. AI-powered shopping. Shopify’s Storefront MCP lets AI assistants like ChatGPT and Perplexity search products, manage carts, and initiate checkout directly. This is “agentic commerce” — customers shop through AI agents instead of browsing websites.
  3. Cross-platform analytics. Ask “What was my best-selling product last month?” or “Which ASINs lost margin this week?” and get answers pulled directly from your store’s live data.

The landscape splits into seven categories: Shopify storefront (official AI shopping layer), Shopify admin (community store management), WooCommerce (WordPress-based stores), Magento/Adobe Commerce (enterprise), Amazon (seller tools), BigCommerce (emerging), and headless platforms (Saleor, Medusa, and other developer-first commerce backends).

Shopify storefront servers (official)

Shopify is the clear leader in e-commerce MCP adoption. They’ve shipped two official servers — one for shopping, one for development — and the ecosystem has dozens of community servers filling the admin gap.

The winner: Shopify Storefront MCP (Official)

Status: GA, live on all Shopify stores | Auth: None required | Transport: Streamable HTTP

Shopify Storefront MCP is built into every Shopify store. Each store gets its own MCP endpoint that exposes storefront features to AI agents — product search, cart operations, policy Q&A, and checkout initiation.

Why it wins: Zero configuration. Every Shopify store automatically has an MCP endpoint. AI assistants like ChatGPT, Perplexity, and Copilot can connect to any Shopify store and help customers browse and buy. No API keys, no custom apps, no setup.

Key features:

  • Semantic product search — natural-language product discovery using vector embeddings
  • Cart operations — create carts, add/remove items, apply discounts
  • Policy Q&A — answers questions about return policies, shipping, FAQs
  • Checkout initiation — starts the checkout flow from within the AI agent
  • MCP UI components — interactive product cards and cart UI within the agent interface
  • Available on every Shopify store — part of Hydrogen Winter ‘26 Edition

The catch: Storefront only. Cannot manage products, inventory, orders, or any admin functions. Read-plus-cart operations, not full store control.

Best for: Merchants who want AI agents to help customers shop. This is Shopify’s “agentic commerce” play.

Also notable: Shopify Dev MCP

Stars: 414 | Language: TypeScript | License: — | Tools: 4

Shopify Dev MCP (@shopify/dev-mcp) is for developers building Shopify apps, not for managing stores. It provides docs search, GraphQL schema introspection, code validation, and theme validation. Zero config, no API key needed.

Key tools: search_dev_docs, introspect_admin_schema, fetch_docs_by_path, get_started

Best for: Developers building Shopify apps or themes who want AI-assisted development with validated Shopify API knowledge.

Shopify admin servers (community)

The biggest gap in Shopify’s MCP story: no official Admin API MCP server. For managing products, orders, inventory, customers, and fulfillment, you need community servers. Several exist, but adoption is still early.

The winner: GeLi2001/shopify-mcp

Stars: 139 | Forks: — | Language: TypeScript | License: — | Tools: 31+

GeLi2001/shopify-mcp is the most adopted community Shopify admin server. Comprehensive GraphQL Admin API integration covering products, customers, orders, metafields, inventory, and tags.

Why it wins: Most tools, most stars, active maintenance. Cursor-based pagination and Shopify query syntax pass-through on all list endpoints. Direct integration with the 2026-01 GraphQL Admin API version.

Key features:

  • Product management — full CRUD for products, variants, and options (8 tools)
  • Customer management — full CRUD, merge, and address management (8 tools)
  • Order management — smart lookup, cancel, close/open, mark as paid, fulfillment, refunds (10 tools)
  • Metafield management — get, set, and delete metafields on any resource (3 tools)
  • Inventory management — set absolute inventory quantities at locations
  • Tag management — add/remove tags on any taggable resource
  • Advanced filtering — pass-through Shopify query syntax on all list endpoints

The catch: Community-maintained with 139 stars — not battle-tested at scale. No official Shopify backing means it could break when APIs change. Requires creating a custom Shopify app for authentication.

Best for: Solo merchants and small teams who want AI-powered store management today and accept the risks of community tooling.

Also notable

Recommendation: Use GeLi2001/shopify-mcp for admin operations. Pair it with Shopify Storefront MCP for the shopping layer and Dev MCP for app development. This three-server combo covers most use cases — but an official Admin MCP from Shopify would instantly obsolete the community options.

Full review: Shopify MCP Servers

WooCommerce servers

WooCommerce powers ~36% of all online stores via WordPress. The MCP ecosystem is smaller than Shopify’s but growing, with both standalone servers and WordPress plugin approaches.

The winner: techspawn/woocommerce-mcp-server

Stars: 26 | Forks: 15 | Language: Python | License: — | Tools: 7+

techspawn/woocommerce-mcp-server provides WooCommerce store management through the WordPress REST API. Built with FastMCP framework.

Why it wins: Most complete standalone WooCommerce MCP implementation. Covers the core operations that store owners need: products, orders, customers, shipping, taxes, discounts, and inventory.

Key features:

  • Product management — list, search, and update products
  • Order management — list, get, update status, add notes
  • Inventory tracking — manage stock levels
  • Customer data — customer information and history
  • JSON-RPC 2.0 protocol — standard MCP communication
  • WooCommerce REST API — uses consumer key authentication

The catch: 26 stars — very early adoption. Fewer tools than the Shopify community servers. Python-based, which may not fit all WordPress deployment environments.

Best for: WooCommerce store owners who want basic AI-powered store management.

Also notable

WooCommerce’s MCP advantage: Full ownership of the server. For brands building proprietary AI models, customizing how the server interacts with an LLM is a competitive advantage that hosted platforms like Shopify can’t match.

Magento / Adobe Commerce servers

Magento (now Adobe Commerce) is the enterprise e-commerce platform. MCP server options are limited but functional.

The winner: boldcommerce/magento2-mcp

Stars: 27+ | Language: Node.js | License: — | Tools: 6+

boldcommerce/magento2-mcp by Bold Commerce connects to Magento 2’s REST API, allowing MCP clients to query and manage store data.

Why it wins: Only dedicated Magento 2 MCP server with meaningful adoption. Covers the essential commerce operations: products, customers, orders, and revenue analytics.

Key features:

  • Product management — fetch by SKU or ID, search with filters, read categories, update attributes and prices
  • Stock information — inventory data retrieval
  • Customer data — retrieve customer information and ordered products
  • Order analytics — order counts and revenue filtered by country and date ranges
  • REST API integration — standard Magento 2 API token authentication

The catch: Read-heavy. Product attribute and price updates are supported, but order creation and fulfillment management are limited. 27 stars means very small community.

Best for: Magento merchants who want AI-powered product and analytics queries.

Also notable

Amazon seller servers

Amazon seller MCP servers connect AI agents to Selling Partner API data — inventory, orders, listings, advertising, and analytics. The landscape splits between commercial products (Seller Labs) and open-source community servers.

The winner: Seller Labs MCP

Type: Commercial (SaaS) | Auth: Seller Labs account | Access: Read-only

Seller Labs MCP connects the Seller Labs Data Hub directly to Claude AI. The Data Hub collects, cleans, and structures Amazon data into 50+ detailed tables.

Why it wins: Production-grade data pipeline. Seller Labs has been in the Amazon tools space for years and has a deep understanding of seller data. Multi-account support makes it suitable for agencies. Natural-language queries like “Which ASINs lost Buy Box share last week?” return data-backed answers instantly.

Key features:

  • 50+ data tables — comprehensive Amazon business data
  • Natural-language analytics — ask questions about your business in plain English
  • Multi-account support — agencies and enterprise sellers with multiple stores
  • Read-only safety — analyzes data but cannot take destructive actions
  • Claude integration — designed specifically for Anthropic’s Claude

The catch: Commercial product — requires a Seller Labs subscription. Read-only by design (can’t create campaigns or change bids). Not open-source.

Best for: Serious Amazon sellers and agencies who want AI-powered business intelligence.

Also notable

  • mattcoatsworth/AmazonSeller-mcp-server — open-source SP-API integration covering catalog, inventory, orders, and reports. Requires Amazon SP-API credentials
  • jay-trivedi/amazon_sp_mcp — Amazon Seller Central MCP for sales data, inventory, returns, and reports
  • Fewsats/amazon-mcp — Amazon MCP for searching and buying products using L402 protocol (consumer-side, not seller)
  • Amazon Ads MCP Server — official Amazon server for ad campaign management. AI tools can plug directly into your ad account
  • Zapier Amazon Seller Central MCP — no-code integration connecting Seller Central to any MCP-compatible AI tool

BigCommerce servers

BigCommerce is developing its own Storefront MCP, currently in closed beta. Third-party options exist but are limited.

The winner: BigCommerce Storefront MCP (Beta)

Status: Closed beta | Type: Official

BigCommerce Storefront MCP enables AI-powered commerce experiences, following Shopify’s model. Combined with Google’s Agent Development Kit (ADK), developers can build commerce-aware merchant agents.

Why it wins: Official platform support signals long-term commitment. Following the same storefront-first approach as Shopify suggests interoperability across the agentic commerce ecosystem.

The catch: Closed beta — not generally available yet. Limited documentation. Feature set not fully disclosed.

Best for: BigCommerce merchants who want to be early adopters of agentic commerce.

Also notable

Headless platform servers (Saleor, Medusa, PrestaShop)

Developer-first headless commerce platforms are shipping MCP servers that give AI agents direct access to their APIs. These are particularly interesting because the platforms themselves are designed for programmatic control.

The winner: saleor/saleor-mcp (Official)

Language: — | License: — | Transport: Streamable HTTP | Compatibility: Saleor 3.21+

saleor/saleor-mcp is the official MCP server for Saleor Commerce (21,800+ stars for the main platform). Production instance deployed at mcp.saleor.app/mcp.

Why it wins: Official, hosted, production-ready. The Saleor team maintains it alongside their platform. Streamable HTTP transport means remote access without local installation. Compatible with Saleor Cloud instances.

Key features:

  • Product browsing — query and browse products with rich formatting
  • Order data — retrieve order information and history
  • Customer data — access customer records
  • Remote hosted — production instance at mcp.saleor.app/mcp
  • Streamable HTTP — modern transport protocol
  • Token-based auth — requires MANAGE_PRODUCTS and MANAGE_ORDERS permissions

The catch: Read-only — no mutations. Cannot create products, process orders, or modify inventory. Limited to data retrieval and browsing.

Best for: Saleor users who want AI-powered data exploration and customer support tooling.

Also notable: Medusa MCP servers

Medusa (27,400+ stars for the main platform) positions itself as “A Commerce Platform for Developers and Agents” — MCP is part of their core story.

  • minimalart/mcp-medusa — 14 comprehensive admin tools covering products, variants, categories, tags, types, orders (list, get, cancel, complete, archive, transfer, fulfillment), draft orders, and customers. The most complete Medusa MCP implementation
  • SGFGOV/medusa-mcp — MCP server for Medusa JS SDK with broader scope: carts, payments, shipping, pricing, returns, and fulfillment automation
  • Medusa Docs MCP — official docs server for AI-assisted Medusa development

Also notable: PrestaShop MCP servers

PrestaShop is popular in Europe and Latin America, with a growing MCP ecosystem:

Which e-commerce MCP server should you use?

“I run a Shopify store and want AI shopping for my customers” → Shopify Storefront MCP is already live on your store. No setup needed.

“I want AI to manage my Shopify products, orders, and inventory” → Use GeLi2001/shopify-mcp (139 stars, 31+ tools). Accept that it’s community-maintained and may need updates when Shopify changes APIs.

“I’m building apps on Shopify” → Use Shopify Dev MCP for docs, schema introspection, and code validation.

“I run a WooCommerce store” → techspawn/woocommerce-mcp-server is the most complete option. The WordPress plugin approaches (iOSDevSK, jlfguthrie) are worth watching if you prefer native WP integration.

“I use Magento / Adobe Commerce” → boldcommerce/magento2-mcp is your best (and really only) option. Covers products, customers, and order analytics.

“I sell on Amazon” → Seller Labs MCP for production-grade analytics (commercial). Open-source SP-API servers exist but are less mature.

“I use a headless platform” → saleor/saleor-mcp if you’re on Saleor (hosted, official). minimalart/mcp-medusa if you’re on Medusa (14 admin tools).

“I use BigCommerce” → Join the Storefront MCP beta if you can. Otherwise, CData’s read-only server or Zapier integration are your options.

1. Agentic commerce is real — Shopify proved the model

Shopify’s Storefront MCP is live on every store, powering shopping through ChatGPT, Perplexity, and Copilot. BigCommerce is following with their beta. This isn’t theoretical anymore — customers are discovering and buying products through AI agents. Every major e-commerce platform will need a storefront MCP endpoint within 12 months.

2. The admin gap is the biggest opportunity

Shopify has no official Admin API MCP server. WooCommerce, Magento, and BigCommerce admin tools are all community-driven with low adoption. The first platform to ship a comprehensive, official admin MCP server — covering products, orders, inventory, fulfillment, and analytics — gains a massive competitive advantage with AI-native merchants.

3. Universal Commerce Protocol (UCP) could unify the landscape

Shopify and Google co-developed the Universal Commerce Protocol — an open standard for AI-agent-to-commerce-backend communication. If UCP gains adoption, it could create a single protocol that works across Shopify, WooCommerce, Magento, and others, reducing the need for platform-specific MCP servers.

What’s missing

Despite 30+ servers, significant gaps remain:

  • No official Shopify Admin MCP — the most impactful missing piece. Community servers fill the gap but lack the reliability merchants need
  • No official WooCommerce MCP — Automattic (WooCommerce’s parent) has a WordPress MCP but no dedicated WooCommerce server
  • No official Magento/Adobe Commerce MCP — Adobe hasn’t entered the MCP space for commerce (only for Creative Suite via community servers)
  • No eBay MCP — one of the world’s largest marketplaces has no MCP server
  • No Etsy MCP — a large marketplace with an active developer API but no MCP integration
  • No multi-platform management — no server that unifies Shopify + Amazon + WooCommerce into a single interface
  • No inventory sync — no MCP server for cross-platform inventory synchronization
  • No shipping/logistics — no dedicated MCP servers for UPS, FedEx, USPS, or DHL APIs
  • No product information management (PIM) — no MCP servers for tools like Akeneo or Salsify
  • No returns/warranty management — limited support across all platforms
  • No A/B testing or conversion optimization — no MCP servers for tools like Optimizely in the commerce context

The e-commerce MCP ecosystem is maturing quickly on the storefront/shopping side (thanks to Shopify) but remains immature for backend operations. The pattern is clear: platforms ship storefront MCP first (customer-facing, low risk), then the community builds admin tools, and eventually official admin servers will follow.

Last updated: March 2026. Star counts and feature details may have changed since publication. See our individual reviews for the most detailed analysis: Shopify MCP Servers · E-Commerce & Shopping MCP