# Legible report: Atlas Ceramics Studio - **Score:** 57/100, grade D (Hard to read) - **Store:** https://atlasceramics.example/ - **Product page:** https://atlasceramics.example/products/ember-stoneware-mug - **Platform:** Custom or headless (medium confidence) - **Scanned:** 2026-10-06 09:14 UTC (demo report, fictional store) - **Online report:** https://legible.digitalage.studio/report/demo-atlas-ceramics ## Pillars | Pillar | Score | Weight | Pass | Warn | Fail | | --- | --- | --- | --- | --- | --- | | Access | 87 | 30% | 7 | 1 | 0 | | Understanding | 38 | 30% | 0 | 5 | 2 | | Trust | 61 | 20% | 1 | 3 | 0 | | Transaction | 34 | 20% | 1 | 1 | 2 | ## Access (87/100) ### [WARN] AI search and shopping agents are allowed robots.txt blocks ChatGPT-User. Some of these may ignore the rule, but the intent turns shoppers' assistants away. _Why it matters:_ Search agents (OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot, Bingbot) build the indexes assistants search before recommending a product. User-triggered agents (ChatGPT-User, Claude-User, Perplexity-User) fetch your page while a shopper waits. Blocking either removes you from the answer. - OAI-SearchBot (OpenAI): Allowed - ChatGPT-User (OpenAI): Blocked on / by "Disallow: /" (line 10) - Claude-SearchBot (Anthropic): Allowed - Claude-User (Anthropic): Allowed - PerplexityBot (Perplexity): Allowed - Perplexity-User (Perplexity): Allowed - Googlebot (Google): Allowed - Bingbot (Microsoft): Allowed - Amzn-SearchBot (Amazon): Allowed **Fix:** Allow search and user-triggered agents explicitly. Keep blocking training crawlers if you want to. - Serve robots.txt as text/plain from the site root and append the agent groups below. `robots.txt (AI agents section)` ```txt # --- AI agents (generated by Legible) --- # Before pasting: delete any existing groups for the agents named below. # Groups for the same agent are merged, so old rules would still apply. # Search and user-triggered agents: allowed, so products can appear # in ChatGPT, Claude, Perplexity, Google AI Mode and Copilot answers. # A named group replaces the * group, so your * rules are repeated here. User-agent: OAI-SearchBot User-agent: ChatGPT-User User-agent: Claude-SearchBot User-agent: Claude-User User-agent: PerplexityBot User-agent: Perplexity-User User-agent: Googlebot User-agent: Bingbot User-agent: Amzn-SearchBot Allow: / Allow: / Disallow: /api/ Disallow: /cart Disallow: /checkout Disallow: /account # Training crawlers: your current choice is kept. Blocking them does # not remove you from AI search. User-agent: GPTBot Disallow: / User-agent: ClaudeBot User-agent: Google-Extended User-agent: Applebot-Extended User-agent: Amazonbot User-agent: meta-externalagent User-agent: CCBot User-agent: Bytespider Allow: / Allow: / Disallow: /api/ Disallow: /cart Disallow: /checkout Disallow: /account Sitemap: https://atlasceramics.example/sitemap.xml ``` ### [PASS] Homepage answers an agent HTTP 200. _Why it matters:_ Every agent starts with an HTTP request. Anything other than a fast 200 with HTML ends the visit before it begins. - Status: 200 - Final URL: https://atlasceramics.example/ - Redirects: 0 - Content-Type: text/html; charset=utf-8 ### [PASS] No bot wall in front of the store Bots and browsers receive the same page. _Why it matters:_ Bot-protection that challenges every non-browser client blocks AI agents even when robots.txt welcomes them. Agents cannot solve JavaScript challenges or CAPTCHAs. - As LegibleBot: HTTP 200 - As a browser: HTTP 200 - Size difference: 0 KB ### [PASS] robots.txt is valid robots.txt parses cleanly: 2 groups, 1 sitemap. _Why it matters:_ RFC 9309 says a robots.txt that errors with 5xx means "crawl nothing", and an HTML page served as robots.txt is noise. A clean file also points agents at your sitemap. - Groups: 2 - Sitemap lines: 1 - Unrecognised lines: none ### [PASS] XML sitemap lists your products Valid sitemap with 3 product URLs. _Why it matters:_ Crawlers that build AI search indexes discover product URLs from sitemaps far more reliably than from links, especially on large catalogues. - Location: https://atlasceramics.example/sitemap.xml - Found via: robots.txt - Type: URL set (6 URLs) ### [PASS] Fast first byte Slowest time to first byte: 420 ms. _Why it matters:_ User-triggered agents fetch pages while someone waits for an answer, with short timeouts. A slow server is a skipped store. - /: 96 ms - /products/ember-stoneware-mug: 420 ms ### [PASS] Lean HTML 119 KB of HTML for /products/ember-stoneware-mug. _Why it matters:_ Agents download and tokenise your HTML. Megabytes of inline scripts and styles bury product facts and can hit size caps before the content. - HTML size: 119 KB - Inline script: 118 KB in 4 tags - Visible words: 0 ### [PASS] Pages are not marked noindex No noindex directives on the pages scanned. _Why it matters:_ A noindex meta tag or X-Robots-Tag header tells search agents to drop the page from their index, even when robots.txt allows crawling. - Homepage: no directive - Product page: no directive ### [INFO] Training crawler policy 1 training crawler blocked, 7 allowed. _Why it matters:_ Training crawlers (GPTBot, ClaudeBot, Google-Extended, CCBot and others) collect content for future models. Blocking them is a legitimate business choice and does not affect AI search visibility. Legible reports your stance but does not score it. - GPTBot (OpenAI): Blocked - ClaudeBot (Anthropic): Allowed - Google-Extended (Google): Allowed - Applebot-Extended (Apple): Allowed - Amazonbot (Amazon): Allowed - meta-externalagent (Meta): Allowed - CCBot (Common Crawl): Allowed - Bytespider (ByteDance): Allowed ## Understanding (38/100) ### [FAIL] Product facts are in the raw HTML The product page is a JavaScript shell. Without running scripts an agent sees no product. _Why it matters:_ Most AI fetchers do not run JavaScript. If the name, price and description only appear after scripts run, the agent sees an empty page and moves on. - Visible words (no JS): 0 - Product name in text: No - Price in text: Not found - Shell signal: Only 0 words of visible text in the raw HTML - Shell signal: Empty Next.js root (#__next) - Shell signal: "Enable JavaScript" notice in the HTML **Fix:** Render product content on the server. - Server-render (SSR or static generation) the product name, price, availability and description. Hydrate interactivity afterwards. ### [FAIL] One clear H1 with the product name The product page has no H1. _Why it matters:_ Extractors use the heading outline to split a page into sections. One H1 naming the product, then H2s for details and reviews, makes the page easy to chunk. - H1: none - H2: 0 **Fix:** Mark the product name up as the page's only

. ### [WARN] Complete Product structured data Product schema is 73% complete. Missing: gtin / mpn, shippingDetails, hasMerchantReturnPolicy and aggregateRating. _Why it matters:_ Schema.org Product and Offer JSON-LD is the most reliable way for agents and shopping indexes to read price, currency, stock, identifiers, shipping and returns without guessing from layout. - name: Ember Stoneware Mug - image: https://atlasceramics.example/images/ember-mug.jpg - description: A 350 ml hand-thrown stoneware mug in a speckled ember glaze. - offers.price: 34 - offers.priceCurrency: EUR - offers.availability: https://schema.org/InStock - sku: ATL-MUG-EMB - gtin / mpn: missing - brand: Atlas Ceramics Studio - shippingDetails: missing - hasMerchantReturnPolicy: missing - aggregateRating: missing **Fix:** Add the missing Product fields. The template below is pre-filled with what Legible found. - Print one Product JSON-LD block per product page, server-side, generated from the same data that renders the visible price. `Product JSON-LD (product template)` ```html ``` ### [WARN] Organization and WebSite schema Organization schema found; WebSite missing. _Why it matters:_ Organization and WebSite markup tells an agent the store's official name, logo, URL and social profiles, so it can tell your shop apart from resellers and lookalikes. - Organization: Atlas Ceramics Studio - WebSite: missing - logo: present - sameAs: present **Fix:** Add an Organization (or OnlineStore) and WebSite graph to the homepage. `Homepage ` ```html ``` ### [WARN] OpenGraph product preview Missing og:description and og:url. _Why it matters:_ Chat interfaces render link previews and many agents fall back to OpenGraph when JSON-LD is missing. og:title, og:image and og:type are the minimum. - og:title: Ember Stoneware Mug - og:description: missing - og:image: https://atlasceramics.example/images/ember-mug.jpg - og:url: missing - og:type: product **Fix:** Add og:description and og:url to the product template. ### [WARN] Canonical URL and language is missing. _Why it matters:_ A canonical URL collapses filter and tracking variants into one product, and a lang attribute tells the agent which language and market the price and policies apply to. - canonical: missing - html lang: en **Fix:** Add . ### [WARN] llms.txt guide for agents No llms.txt. The server answers /llms.txt with an HTML page and HTTP 200, so agents get a page instead of a 404. _Why it matters:_ llms.txt is a proposed markdown index of a site's most useful pages. No major assistant has confirmed it uses the file, so it carries little weight here, but it is cheap and gives agents a clean map of products and policies. - llms-full.txt: not found **Fix:** Publish /llms.txt. This one is generated from the pages Legible found on your site. - Serve /llms.txt as text/plain or text/markdown from the site root. `/llms.txt` ```markdown # Atlas Ceramics Studio > Handmade stoneware from a two-person studio in Split. Mugs, plates, bowls and vases, glazed in small batches. ## Products - [Ember Stoneware Mug](https://atlasceramics.example/products/ember-stoneware-mug) - [Tide Dinner Plate](https://atlasceramics.example/products/tide-dinner-plate) - [Cove Bud Vase](https://atlasceramics.example/products/cove-bud-vase) ## Categories - [Tableware](https://atlasceramics.example/collections/tableware) - [Vases](https://atlasceramics.example/collections/vases) ## Policies - [Shipping](https://atlasceramics.example/help/shipping) - [Returns](https://atlasceramics.example/help/returns) - [Privacy](https://atlasceramics.example/privacy) - [Contact](https://atlasceramics.example/contact) ## Optional - [Journal](https://atlasceramics.example/journal) ``` ## Trust (61/100) ### [WARN] Policy pages are linked Terms not linked. _Why it matters:_ Before recommending a purchase, assistants look for shipping costs, the return window and who to contact. In the EU, consumers have a 14-day right of withdrawal for most online purchases (Directive 2011/83/EU, Art. 9), so returns terms are a deciding fact. - Shipping: https://atlasceramics.example/help/shipping - Returns: https://atlasceramics.example/help/returns - Privacy: https://atlasceramics.example/privacy - Terms: no link found - Contact: https://atlasceramics.example/contact **Fix:** Link terms from the site footer. - Link every policy page from a site-wide footer with plain anchor links agents can follow. ### [WARN] Business identity is machine-readable Organization schema found, but address and legal name or VAT ID are missing. _Why it matters:_ Agents weigh whether a merchant is real. A legal name, postal address and a contact point in Organization schema are the strongest signals short of a marketplace listing. - address: missing - email / telephone / contactPoint: present - legalName / vatID: missing **Fix:** Add address and legal name or VAT ID to the Organization node. ### [WARN] Visible price matches structured data JSON-LD says EUR 34.00, but no price is visible in the raw HTML. _Why it matters:_ When the price a shopper sees differs from the price in JSON-LD, agents quote the wrong number or distrust both. Google Merchant Center suspends listings for mismatches, and under EU consumer law the price shown must be the price charged. - JSON-LD price: EUR 34.00 - Current prices on page: none found - Microdata price: none ### [PASS] HTTPS everywhere, with HSTS HTTPS with HSTS, and HTTP redirects to HTTPS. _Why it matters:_ Agents that hand a shopper off to checkout will not send them to an insecure page. HSTS proves the store never falls back to plain HTTP. - Final URL scheme: https - http:// redirects to https: yes (308) - Strict-Transport-Security: max-age=63072000; includeSubDomains; preload ## Transaction (34/100) ### [FAIL] UCP business profile UCP profile found with 2 spec errors: Required "transport": one of rest, mcp, a2a, embedded. _Why it matters:_ The Universal Commerce Protocol (UCP), announced by Google in January 2026, lets agents discover a store's capabilities from /.well-known/ucp and complete checkout through a standard API. It powers buying in Google AI Mode and Gemini. Current spec version: 2026-08-25. - ucp.version: 2026-01-11 - Services: dev.ucp.shopping - Transports: none - Capabilities: none - Payment handlers: none - Signing keys: none - $.ucp.version: Version 2026-01-11 is valid but not the latest (2026-08-25). Agents negotiate down, so this still works. - $.ucp.services.dev.ucp.shopping: Service bindings should be an array (one entry per transport). - $.ucp.services.dev.ucp.shopping[0].transport: Required "transport": one of rest, mcp, a2a, embedded. - $.ucp.capabilities: Capabilities must be an object keyed by capability name. An array is not the published form. - $.keys: No signing keys published. Recommended so agents can verify signed responses. **Fix:** Publish a UCP profile backed by a working checkout implementation. - Implement the UCP REST or MCP binding for checkout, then publish the profile at /.well-known/ucp. `/.well-known/ucp (skeleton)` ```json { "ucp": { "version": "2026-08-25", "services": { "dev.ucp.shopping": [ { "version": "2026-08-25", "spec": "https://ucp.dev/2026-08-25/specification/overview/", "transport": "rest", "endpoint": "https://atlasceramics.example/ucp/v1", "schema": "https://ucp.dev/2026-08-25/services/shopping/rest.openapi.json" } ] }, "capabilities": { "dev.ucp.shopping.checkout": [ { "version": "2026-08-25", "spec": "https://ucp.dev/2026-08-25/specification/shopping/checkout/", "schema": "https://ucp.dev/2026-08-25/schemas/shopping/checkout.json" } ], "dev.ucp.shopping.fulfillment": [ { "version": "2026-08-25", "spec": "https://ucp.dev/2026-08-25/specification/shopping/extensions/fulfillment/", "schema": "https://ucp.dev/2026-08-25/schemas/shopping/fulfillment.json", "extends": "dev.ucp.shopping.checkout" } ] }, "payment_handlers": {} }, "keys": [] } ``` ### [FAIL] Add-to-cart works without scripts No add-to-cart control in the server HTML. _Why it matters:_ Browser-driving agents (ChatGPT agent, Gemini in Chrome and similar) click through your real checkout. A plain form or link for add-to-cart in the server HTML is the most robust path when no protocol is available. - Cart form in HTML: no - Add-to-cart link: no - Add-to-cart button: no **Fix:** Render a real form that posts to your cart endpoint and works before scripts load; let JavaScript enhance it. - Server-render (SSR or static generation) the product name, price, availability and description. Hydrate interactivity afterwards. `Product template` ```html
``` ### [WARN] Machine-readable catalogue One catalogue signal: Product URLs in the sitemap (3). _Why it matters:_ Agent channels (ChatGPT shopping, Google AI Mode, Copilot) prefer a structured product feed they can refresh over crawling pages. A feed keeps price and stock current between crawls. - Signal: Product URLs in the sitemap (3) **Fix:** Expose a product feed and submit it to agent channels. - Expose a product feed (Google Merchant Center format works for most agent channels) and a product sitemap. ### [PASS] Agentic Commerce Protocol readiness (signal) All ACP prerequisites are in place. Enrolment itself happens with OpenAI and cannot be detected from outside. _Why it matters:_ OpenAI and Stripe's Agentic Commerce Protocol (ACP) launched in September 2025 to let shoppers buy inside ChatGPT. There is no public discovery file: merchants are onboarded through OpenAI and submit a product feed. In March 2026 OpenAI moved checkout back to merchants' own sites, so discovery in ChatGPT now depends on that feed and on crawlable pages. Legible scores the prerequisites, not enrolment. - OAI-SearchBot allowed: yes - Product structured data: yes - Feed source available: yes ### [INFO] Platform and remediation path Detected Custom or headless (medium confidence). _Why it matters:_ What you can fix, and how fast, depends on the platform. Shopify merchants can switch most of this on; WooCommerce, Magento and custom stores need it built or configured. - Signal: Next.js application - Signal: x-powered-by: Next.js ## What an agent extracts From https://atlasceramics.example/products/ember-stoneware-mug, about 0 tokens from 119 KB of HTML. | Fact | Found | Source | | --- | --- | --- | | Product name | Ember Stoneware Mug | json-ld | | Price | EUR 34.00 | json-ld | | Availability | InStock | json-ld | | Brand | Atlas Ceramics Studio | json-ld | | GTIN / EAN | missing | — | | Shipping | missing | — | | Returns | missing | — | | Rating | missing | — | ## Requests made - 200 https://atlasceramics.example/ (Homepage (as LegibleBot), 116 ms) - 200 https://atlasceramics.example/ (Homepage (as a browser, for bot-wall comparison), 116 ms) - 200 https://atlasceramics.example/robots.txt (robots.txt, 60 ms) - 308 http://atlasceramics.example/ (HTTP to HTTPS redirect probe, 42 ms) - 200 https://atlasceramics.example/llms.txt (llms.txt, 80 ms) - 200 https://atlasceramics.example/llms-full.txt (llms-full.txt, 80 ms) - 200 https://atlasceramics.example/.well-known/ucp (UCP business profile, 95 ms) - 200 https://atlasceramics.example/sitemap.xml (XML sitemap, 90 ms) - 200 https://atlasceramics.example/products/ember-stoneware-mug (Product page (as LegibleBot), 440 ms) Evaluated against UCP 2026-08-25 and the agent list of 2026-10-07. Agent categories: Training crawler, Search / index, User-triggered. --- Generated by Legible, a Digital Age product (https://digitalage.studio/en/).