Skip to content
Headless / Custom

Headless commerce visible to AI agents — without rewriting your stack

Whether you're on Medusa, commercetools, Saleor, or a custom Rails/Django catalog, SignalixIQ ingests your product data via REST, GraphQL, or JSON feed and exposes it through a hosted MCP server.

About this integration

Custom commerce stacks (Next.js + Stripe, Remix + Medusa, custom Rails) and headless commerce platforms (commercetools, Saleor, Spree). SignalixIQ ingests catalog via a flexible REST/GraphQL adapter or direct JSON feed.

Why Headless / Custom Commerce stores need MCP

Headless setups are agent-invisible by default

Headless commerce often relies on client-side rendering, which AI agent crawlers handle inconsistently. Even with SSR, custom product schemas usually lack canonical fields like GTIN and structured availability. SignalixIQ bypasses this by ingesting from your data source directly.

Multi-frontend setups create catalog inconsistency

Headless merchants often run multiple frontends (web, mobile, partner portals) against the same catalog. Agents see different data depending on which frontend they hit. SignalixIQ unifies by pulling from your single source of truth.

Custom commerce gives you the cleanest agent integration possible

Unlike platform-specific quirks (Shopify's variant pricing, WooCommerce's missing GTIN), headless stores can map exactly to SignalixIQ's canonical schema. The result is the highest GEO scores in the customer base.

How to connect Headless / Custom Commerce to AI shopping agents

  1. 1

    Choose your ingestion mode

    REST endpoint (recommended), GraphQL endpoint, or signed JSON feed delivered on schedule. The SignalixIQ dashboard guides you through the right choice based on your stack.

  2. 2

    Map your fields to the canonical schema

    Use the SignalixIQ field-mapping UI to map your custom field names (e.g., 'productSlug', 'priceCents') to canonical fields (title, price). Validation runs in real-time.

  3. 3

    Set sync cadence

    Polling (every 5/15/60 min) or webhook push. Push is recommended for real-time inventory and price accuracy.

  4. 4

    MCP endpoint live

    Same as the platform integrations — registered with ChatGPT, Claude, Perplexity within 24-72 hours.

Headless / Custom Commerce → SignalixIQ field mapping

Exactly which Headless / Custom Commerce fields SignalixIQ reads and how they map to the canonical agent-product schema.

SignalixIQ fieldHeadless / Custom Commerce fieldNotes
product.titleYour product name fieldRequired.
product.descriptionYour description field (markdown OK)Markdown auto-converted to clean prose.
product.gtinYour GTIN/UPC/EAN field if availableStrongly recommended. If absent, SignalixIQ runs heuristic matching against GS1.
product.brandYour brand fieldRequired for agent product disambiguation.
product.priceYour price field (cents or dollars; both supported)Configurable unit.
product.availabilityStock status enum or numeric quantityEither format accepted.
product.urlCanonical product URL on your storefrontUsed in agent click-through.
product.imagesImage URL arrayUp to 8 images surfaced to agents.

Common Headless / Custom Commerce issues that hurt agent discoverability

  • Field naming inconsistency across data sources (catalog API vs search index)
  • Stale data when sync cadence is too long (recommend ≤15 min polling or webhook push)
  • Currency unit confusion (price in cents vs dollars vs subunit)
  • Image URLs requiring auth or hotlink protection — must be publicly accessible
  • Missing brand fields on white-label or generic products

SignalixIQ's free GEO score audits your store and ranks these issues by agent-discoverability impact. Run a free scan →

Frequently asked

Does this work with Medusa, Saleor, commercetools, Spree, or other open-source commerce?

Yes. All major headless platforms expose REST or GraphQL endpoints that SignalixIQ can ingest. Medusa, Saleor, commercetools, and Spree have battle-tested setup guides. Custom Rails/Django/Laravel commerce works via the JSON-feed mode.

What if my product data is split across multiple services?

Common in mature headless setups (catalog service + pricing service + inventory service). SignalixIQ can pull from multiple sources and merge based on product ID. Configuration is per-customer for the first multi-source ingestion; subsequent ones are self-serve.

Can I push to SignalixIQ instead of having you pull?

Yes — webhook push is the recommended approach for headless setups. SignalixIQ provides a signed-webhook endpoint; you POST product changes when they happen. Lower latency and avoids polling overhead.

Make your Headless / Custom Commerce store discoverable

SignalixIQ hosts the MCP server so AI shopping agents can find your products. Free GEO score in 60 seconds — no signup required.

Other platform integrations