| name | seo-ecommerce |
| description | E-commerce SEO — product-page on-page audit, Product schema (JSON-LD) requirements and scoring, Google Shopping / Amazon marketplace intelligence, organic-vs-Shopping keyword gaps, and UCP (Universal Commerce Protocol) agentic-checkout profile audit. |
| allowed-tools | Read, Write, Glob, Grep, WebSearch, WebFetch |
SEO Ecommerce — Product Pages, Marketplace Intelligence, UCP
A product page competes on three fronts at once: the SERP (on-page + schema), the marketplace (Google Shopping / Amazon pricing and seller data), and — starting in 2026 — AI shopping agents (UCP). This skill scores all three and never treats "no marketplace data" as a blocker.
When this skill activates
Implicit: "ecommerce SEO", "product SEO", "product page audit", "Google Shopping", "marketplace SEO", "product schema", "Amazon SEO", "product listings", "shopping ads", "merchant SEO", "UCP profile", "agentic checkout".
Explicit: "Use the seo-ecommerce skill to [task]."
Routed from: [[seo]] orchestrator (ecommerce industry detection), product-page and category-page SEO contexts, [[seo-schema]] for the Product schema type specifically, [[seo-page]] for page-level SEO that turns out to be a product page, /mk:seo ecommerce actions.
Scope
Covers:
- Product-page on-page SEO checklist and weighted scoring (title, meta, headings, images, internal linking, content quality).
- Product schema (JSON-LD) — required + recommended properties, validation rules, completeness scoring.
- Marketplace intelligence — Google Shopping and Amazon competitive analysis (pricing, sellers, listing quality).
- Organic-vs-marketplace keyword gap analysis.
- UCP (Universal Commerce Protocol) profile audit — agentic-checkout discoverability for AI shopping agents.
Does NOT cover:
- General JSON-LD generation/validation mechanics for non-Product types → [[seo-schema]] (this skill owns the Product-specific field rules; [[seo-schema]] owns the taxonomy and validator).
- Page-level on-page SEO for non-product pages → [[seo-page]].
- Category-page template generation at scale, faceted-navigation and pagination crawl-budget management, canonicalization rules for large catalogs → [[seo-programmatic]].
- Product image file optimization itself (alt text, format, compression) → [[seo-images]]; this skill only checks that image optimization happened.
- Site-wide technical crawl/index health (Core Web Vitals, sitemaps, JS rendering) → [[seo-technical]].
- DataForSEO cost/quota guardrails as a general concern → [[seo-dataforseo]].
1. Product page on-page audit
Fetch the page (WebFetch) and score these product-specific signals:
Title tag — primary product keyword + brand name, under 60 characters, format [Product Name] - [Key Feature] | [Brand].
Meta description — product keyword + benefit, price or "from $XX" (triggers rich-snippet interest), a call to action (Shop now / Buy / Free shipping), under 155 characters.
Heading structure — single H1 matching the primary product name; H2s for Features / Specifications / Reviews / Related Products; no duplicate H1s across product variants.
Product images — alt text includes product name + distinguishing feature; descriptive filenames (not IMG_001.jpg); WebP with JPEG fallback; at least 3 images (hero, detail, lifestyle); >= 800px for Google Shopping eligibility; lazy-loading on below-fold images only. (Delegate the actual fix to [[seo-images]] — this skill checks, doesn't optimize.)
Internal linking — breadcrumb Home > Category > Subcategory > Product; related-products cross-sell/upsell section; keyword-rich anchor back to the category page; reviews link to a full review page if separate.
Content quality — unique product description (not manufacturer copy-paste); >= 200 words in the description body; a specs table, not just prose; on-page user reviews (UGC signal).
Scoring
| Category | Weight | Criteria |
|---|
| Schema completeness | 25% | Required + recommended Product fields (§2 below) |
| Title & meta | 15% | Keyword placement, length, format |
| Image optimization | 20% | Alt text, format, sizing, count |
| Content quality | 20% | Unique description, specs, reviews |
| Internal linking | 10% | Breadcrumbs, related products, categories |
| Technical | 10% | Page speed, mobile rendering, canonical |
2. Product schema (JSON-LD)
Required (Google Merchant): name, image, offers. Use Offer, not AggregateOffer, for merchant listings.
{
"@context": "https://schema.org",
"@type": "Product",
"name": "",
"image": [""],
"offers": {
"@type": "Offer",
"url": "",
"priceCurrency": "USD",
"price": "0.00",
"availability": "https://schema.org/InStock"
}
}
Recommended (enhance rich results): sku; description, brand, offers.seller; gtin13/gtin14/mpn; aggregateRating; review (minimum 1); variant attributes color/material/size; shippingDetails (a ShippingDetails object with rate + delivery time — merchant-level shipping via ShippingService is also supported, and shipping/returns can be set in Search Console without a Merchant Center account); hasMerchantReturnPolicy (a MerchantReturnPolicy with type + days); hasAdultConsideration — required for adult-oriented products, value must be https://schema.org/SexualContentConsideration.
Validation rules:
price is a bare number string ("29.99"), never "$29.99".availability uses the full https://schema.org/... enum.image is an array with >= 1 high-resolution URL.priceCurrency is ISO 4217 (USD, EUR, GBP...).- If
brand is present, brand.name is non-empty and not "N/A".
priceValidUntil (if present) is ISO 8601.- If
aggregateRating is present, both ratingValue and reviewCount are required.
Completeness scoring: all required fields = 50/100; + aggregateRating = 65/100; + sku/gtin/mpn = 75/100; + shippingDetails = 85/100; + hasMerchantReturnPolicy = 90/100; + 3+ reviews = 100/100.
Delegate JSON-LD generation itself and cross-type validation to [[seo-schema]]; this skill supplies the Product-specific field list and the score above.
3. Marketplace intelligence — Google Shopping & Amazon
Live pricing/seller data needs a paid provider. Use a preference ladder, same as the rest of the marketing kit (see [[seo-writing]] references/research-tools.md): DataForSEO Merchant API (via MCP, /ck:use-mcp) → manual WebSearch/WebFetch spot-check of the Shopping tab or Amazon listing → skip and label [marketplace data unverified]. Never fabricate a price, rating, or seller name to fill the gap. Cost/quota guardrails for DataForSEO itself are a [[seo-dataforseo]] concern — check budget approval before a paid call.
When a provider is available, produce:
Pricing intelligence — price distribution (min, max, median, P25, P75); outliers (> 2 std. dev. from median); price-to-rating correlation; currency normalized to USD or the user's currency.
Seller landscape — top 10 sellers by listing count; merchant rating distribution; free-shipping prevalence; new vs. established sellers.
Product listing quality — title keyword patterns in top listings; average rating/review-count benchmarks; image count per listing; availability status distribution.
Amazon results additionally carry asin, is_prime, is_best_seller. Full endpoint/field reference (for when a DataForSEO Merchant API MCP is wired up): references/marketplace-endpoints.md.
Cross-marketplace report:
| Metric | Google Shopping | Amazon |
|---|
| Avg price | $ | $ |
| Median rating | X.X | X.X |
| Avg review count | N | N |
| Top seller share | % | % |
| Free shipping % | % | % |
4. Marketplace keyword gaps
Cross-reference organic visibility against Shopping/marketplace visibility for the same keyword set:
- Organic rankings for the domain (via [[seo-dataforseo]] MCP, or a
site:domain + keyword WebSearch spot-check).
- Shopping/marketplace presence for the same top organic keywords (§3 above).
- Cross-reference into gap types.
| Gap type | Meaning | Action |
|---|
| Organic only | Ranks organically, no Shopping presence | Set up a Merchant Center feed; bid on these keywords |
| Shopping only | Shopping visibility, weak/no organic | Write content (buying guides, comparisons) targeting these keywords |
| Both present | Visible in both channels | Optimize: keep price consistent across channels, enhance schema |
| Neither | No visibility either way | Low priority unless search volume is high |
Output as two tables: Organic → Shopping opportunities and Shopping → Organic opportunities, each with keyword / position-or-rank / volume / CPC / recommended action.
5. UCP — Universal Commerce Protocol
A Google-initiated open standard (co-developed with Shopify, Etsy, Wayfair, Target, Walmart; payment partners Visa/Mastercard/Stripe/Adyen/Amex) letting AI shopping agents discover, negotiate, and transact with merchants without one-off integrations. Google has a confirmed first reference implementation for conversational buying in AI Mode in Search; broader "Universal Cart" rollout claims trace to keynote coverage, not a Google-owned source — treat those as hedged. UCP uses date-based versioning (YYYY-MM-DD, e.g. 2026-04-08), not semver — a literal "version": "1.0" is invalid. Two integration paths: Native checkout (default, full agentic potential) and Embedded checkout (iframe-based, Google-approved merchants only). Pairs with AP2 (Agent Payments Protocol) for cryptographic proof of user consent. Merchants stay Merchant of Record under UCP.
A merchant already on Google Merchant Center with clean Product schema can declare a UCP profile at /.well-known/ucp in a sprint. Full spec context, capability examples, and the AP2 relationship: references/ucp-universal-commerce-protocol.md.
Audit procedure (WebFetch the profile, no dedicated tool needed):
- Presence — does
/.well-known/ucp resolve to valid JSON? Missing profile is an opportunity, not a failure.
- Capability coverage — which capabilities are declared (
dev.ucp.shopping.checkout, .fulfillment, .discount, .cart)? Flag missing ones as opportunities.
- Endpoint reachability — are declared endpoints HTTPS with valid TLS, not 5xx? A plain
WebFetch against each declared endpoint is enough to catch dead links; full negotiation testing is out of scope.
- Version coherence — is
version a valid date (YYYY-MM-DD)? Flag a literal "1.0" or any non-date string as invalid.
- Integration path — does the profile imply Native or Embedded checkout?
Audit posture by tier:
| Tier | Guidance |
|---|
| E-commerce site already on Merchant Center | Recommend declaring a UCP profile as a forward-looking opportunity |
| DTC site not on Merchant Center | Don't recommend UCP yet — Merchant Center is the prerequisite |
| Informational / B2B site | Ignore UCP — except hospitality/restaurant sites: UCP is expanding into Lodging and Food (hotel booking in AI Mode, food delivery via Google Maps) |
Never score the absence of a UCP profile as a critical failure — the protocol itself is live, but broad merchant adoption is still early.
Key concepts
- Merchant Center vs. on-page schema — Product schema on the page and a Merchant Center feed are two different surfaces carrying overlapping data; this skill audits the page, not the feed (feed validation is a Merchant Center / Merchant API concern, not this skill's).
- Preference ladder for marketplace data — MCP/paid provider first,
WebSearch/WebFetch spot-check second, honest [unverified] label last. Never invent a price, rating, or seller.
- Keyword-channel gap — a keyword can rank organically, appear in Shopping, both, or neither; each combination implies a different next action (§4).
- UCP is a discovery/negotiation layer, not a payment processor — it sits next to Merchant Center feeds and Google Business Profile as a third agent-era discovery surface; it doesn't replace either.
Output
plans/marketing/<campaign>/seo-ecommerce.md — the e-commerce SEO report: overall score, product-page sub-scores, schema completeness, marketplace intelligence (if a provider was available, else labeled unverified), UCP audit, and ranked recommendations.- Inline recommendations in the conversation for single-page or single-keyword requests.
Cross-references
plans/marketing-context.md — required hub.claude/workflows/marketing-rules.md — content quality + truth-only rules (no fabricated prices, ratings, or reviews).claude/skills/marketing/README.md — full kit overview.claude/skills/marketing/seo/SKILL.md — orchestrator (parent)- [[seo-schema]] — JSON-LD generation/validation mechanics this skill's Product rules plug into
- [[seo-page]] — page-level on-page SEO for non-product pages
- [[seo-programmatic]] — category-page templates at scale, faceted nav, pagination
- [[seo-images]] — product image optimization itself
- [[seo-technical]] — site-wide crawl/index health
- [[seo-dataforseo]] — DataForSEO provider integration + cost guardrails
- [[seo-google]] — GSC indexation/Performance data for product URLs (not Merchant Center feed validation)
Provenance
Imported from AgriciDaniel/claude-seo (seo-ecommerce, v2.2.4; original author Matej Marjanovic, Pro Hub Challenge) and adapted for KitForge. Adaptations: dropped all claude-seo run <script>.py invocations and the DataForSEO cost-guardrail script calls — ClauKit ships no script runtime for this kit, so marketplace/UCP data fetches now route through the same MCP-first/WebSearch-WebFetch-fallback preference ladder used elsewhere in the kit (see [[seo-writing]] references/research-tools.md), and DataForSEO cost/quota concerns are pointed at [[seo-dataforseo]] instead of a bespoke script. Product-page checklist, Product schema field rules + scoring, keyword-gap framework, and UCP audit criteria are preserved substantively. Two reference files ported (marketplace-endpoints.md, ucp-universal-commerce-protocol.md) with script-invocation lines replaced by the MCP/WebFetch ladder and the cost-tiers.md cross-reference (not present in this port) replaced by a pointer to [[seo-dataforseo]]. For full original content see the source repo.
