{
 "version": "1",
 "collection": "answers",
 "about": "Short answers to common questions, each 30 to 70 words, with the pages they're based on.",
 "updated": "2026-11-12",
 "count": 38,
 "items": [
  {
   "id": "what-is-a-good-agentscore",
   "url": "https://ghostagentlab.com/learn/answers/what-is-a-good-agentscore/",
   "question": "What is a good AgentScore?",
   "answer": "An AgentScore of 80 or more is in the Good band, which means AI agents can use your site well. Scores from 60 to 79 are Fair, 40 to 59 are Weak and below 40 are Poor. Treat your own trend as the real target, and clear any failure that caps your score first.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/reading-your-agentscore-report/",
    "https://ghostagentlab.com/agentscore/methodology/",
    "https://ghostagentlab.com/articles/agent-readiness-kpis/"
   ]
  },
  {
   "id": "is-agent-readiness-the-same-as-seo",
   "url": "https://ghostagentlab.com/learn/answers/is-agent-readiness-the-same-as-seo/",
   "question": "Is agent readiness the same as SEO?",
   "answer": "No. SEO decides whether you're found, and agent readiness decides whether an agent that found you can get in, read your prices and finish the task. Good SEO gives you a head start, so check how bot protection treats AI assistants and whether prices appear without JavaScript.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/seo-to-agent-readiness/"
   ]
  },
  {
   "id": "who-should-own-agent-readiness",
   "url": "https://ghostagentlab.com/learn/answers/who-should-own-agent-readiness/",
   "question": "Who should own agent readiness in my company?",
   "answer": "Name one accountable owner, usually the head of e-commerce or digital, who makes sure every fix has someone. The work itself spreads across SEO and content, developers, whoever runs bot protection or the CDN, and the e-commerce platform admin. Route each finding to the team that usually fixes it.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/who-owns-agent-readiness/",
    "https://ghostagentlab.com/blog/reading-your-agentscore-report/"
   ]
  },
  {
   "id": "how-long-does-it-take-to-make-a-website-agent-ready",
   "url": "https://ghostagentlab.com/learn/answers/how-long-does-it-take-to-make-a-website-agent-ready/",
   "question": "How long does it take to make a website agent-ready?",
   "answer": "Plan on about 90 days in three phases of roughly a month each. Spend the first month letting agents in, the second making pages readable and usable, and the third proving agents can finish your money journeys. Many access fixes are settings changes that take hours.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/90-day-agent-readiness-plan/",
    "https://ghostagentlab.com/agentscore/methodology/",
    "https://ghostagentlab.com/blog/explain-agent-readiness-to-your-board/"
   ]
  },
  {
   "id": "how-do-i-explain-agent-readiness-to-my-board",
   "url": "https://ghostagentlab.com/learn/answers/how-do-i-explain-agent-readiness-to-my-board/",
   "question": "How do I explain agent readiness to my board?",
   "answer": "Use one page with five headings: what's changing, why it matters, where you stand, what you're doing and what you need. Skip market forecasts and show agent visits from your own logs instead. Ask for one accountable owner and a small, staged budget with a review after each stage.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/explain-agent-readiness-to-your-board/",
    "https://ghostagentlab.com/articles/agent-readiness-kpis/"
   ]
  },
  {
   "id": "what-are-browser-agents",
   "url": "https://ghostagentlab.com/learn/answers/what-are-browser-agents/",
   "question": "What are browser agents?",
   "answer": "Browser agents are AI agents that open your site in a real browser and click, type and scroll like a person to book, sign up or buy. They're the agents that complete tasks, so they're where agent traffic turns into revenue. Give them named buttons, labelled fields, pop-ups they can close and a checkout without forced accounts.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/kinds-of-ai-agents/",
    "https://ghostagentlab.com/articles/agent-friendly-buttons-forms/"
   ]
  },
  {
   "id": "should-i-block-gptbot",
   "url": "https://ghostagentlab.com/learn/answers/should-i-block-gptbot/",
   "question": "Should I block GPTBot?",
   "answer": "Block GPTBot only if you don't want your content used to train OpenAI's models. It's a training crawler, so blocking it leaves ChatGPT search and ChatGPT's real-time page fetches untouched. Stores whose pages exist to sell products usually allow it. Publishers and others whose content is the product often block it.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/block-or-allow-ai-crawlers/",
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/"
   ]
  },
  {
   "id": "does-blocking-gptbot-remove-my-site-from-chatgpt",
   "url": "https://ghostagentlab.com/learn/answers/does-blocking-gptbot-remove-my-site-from-chatgpt/",
   "question": "Does blocking GPTBot remove my site from ChatGPT?",
   "answer": "No. GPTBot only collects pages that may be used to train OpenAI's models. ChatGPT search uses OAI-SearchBot, and ChatGPT-User fetches pages when a person asks about you. Block GPTBot if you want to opt out of training, and keep the other two allowed so ChatGPT can still read, cite and link to you.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/",
    "https://ghostagentlab.com/blog/block-or-allow-ai-crawlers/"
   ]
  },
  {
   "id": "what-is-chatgpt-user",
   "url": "https://ghostagentlab.com/learn/answers/what-is-chatgpt-user/",
   "question": "What is ChatGPT-User?",
   "answer": "ChatGPT-User is the OpenAI agent that fetches a page when someone using ChatGPT, or a custom GPT, asks about it or clicks through to it. It acts for one person in real time, so treat it like a customer visiting through ChatGPT. Almost always allow it, in robots.txt and in your bot protection.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/block-or-allow-ai-crawlers/",
    "https://ghostagentlab.com/articles/bot-protection-ai-agents/",
    "https://ghostagentlab.com/articles/verify-ai-crawlers/"
   ]
  },
  {
   "id": "will-ai-agents-respect-my-robots-txt",
   "url": "https://ghostagentlab.com/learn/answers/will-ai-agents-respect-my-robots-txt/",
   "question": "Will AI agents respect my robots.txt?",
   "answer": "Well-behaved AI crawlers follow robots.txt, and scrapers ignore it. Some operators say it may not apply when an assistant fetches a page because a person asked, and browser agents look like ordinary browsers, so rules by name don't reach them. Use robots.txt to state your policy, and enforce it at your CDN by verifying who's asking.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/",
    "https://ghostagentlab.com/blog/block-or-allow-ai-crawlers/"
   ]
  },
  {
   "id": "can-ai-agents-solve-captchas",
   "url": "https://ghostagentlab.com/learn/answers/can-ai-agents-solve-captchas/",
   "question": "Can AI agents solve CAPTCHAs?",
   "answer": "No. AI agents can't solve CAPTCHAs, and they aren't meant to. When an agent meets one, it typically stops or hands the page back to the person. Keep CAPTCHAs off product, pricing and policy pages, show challenges only when traffic looks abusive, and let verified AI assistants through.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/bot-protection-ai-agents/",
    "https://ghostagentlab.com/blog/what-ai-agents-cant-do-yet/",
    "https://ghostagentlab.com/articles/cdn-settings-ai-agents/"
   ]
  },
  {
   "id": "does-cloudflare-block-ai-agents",
   "url": "https://ghostagentlab.com/learn/answers/does-cloudflare-block-ai-agents/",
   "question": "Does Cloudflare block AI agents?",
   "answer": "Cloudflare can block AI agents, depending on your settings. It offers settings aimed at blocking AI crawlers, and keeps a verified bots list with categories such as AI Assistant and AI Crawler. Check exactly which agents any AI setting covers, allow verified assistants and AI search, and test your site from the outside.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/cdn-settings-ai-agents/",
    "https://ghostagentlab.com/blog/questions-for-your-cdn-vendor/",
    "https://ghostagentlab.com/platforms/shopify/"
   ]
  },
  {
   "id": "how-do-i-spot-a-fake-googlebot-or-gptbot",
   "url": "https://ghostagentlab.com/learn/answers/how-do-i-spot-a-fake-googlebot-or-gptbot/",
   "question": "How do I spot a fake Googlebot or GPTBot?",
   "answer": "Check where the request came from, because anyone can copy a crawler's name. Confirm Googlebot with forward-confirmed reverse DNS or Google's published IP lists, and GPTBot against OpenAI's published IP ranges. A request that claims the name but fails the check is an impostor, so challenge or block it.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/verify-ai-crawlers/"
   ]
  },
  {
   "id": "what-is-web-bot-auth",
   "url": "https://ghostagentlab.com/learn/answers/what-is-web-bot-auth/",
   "question": "What is Web Bot Auth?",
   "answer": "Web Bot Auth is a new standard that lets an AI agent sign each request, so your site can prove who sent it. The agent publishes a public key on its own domain, and your CDN checks the signature. Ask your CDN or bot protection provider whether it verifies these signatures, and turn it on where you can.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/signed-requests/",
    "https://ghostagentlab.com/articles/verify-ai-crawlers/",
    "https://ghostagentlab.com/articles/cdn-settings-ai-agents/"
   ]
  },
  {
   "id": "should-i-allow-ai-training-crawlers",
   "url": "https://ghostagentlab.com/learn/answers/should-i-allow-ai-training-crawlers/",
   "question": "Should I allow AI training crawlers?",
   "answer": "It's a real business choice. Lean towards allowing training crawlers if your pages mainly exist to sell something, such as product pages, help centers and pricing. Lean towards blocking if your content is the product, such as publishing, research or courses. Either way, block them by name so AI search and assistants can still reach you.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/block-or-allow-ai-crawlers/",
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/"
   ]
  },
  {
   "id": "does-shopify-block-ai-agents",
   "url": "https://ghostagentlab.com/learn/answers/does-shopify-block-ai-agents/",
   "question": "Does Shopify block AI agents?",
   "answer": "Shopify's defaults usually let AI agents in. Its default robots.txt allows them, and its always-on Cloudflare protection is usually fine. Problems tend to come from changes and settings you control, such as robots.txt edits, checkout bot protection, required sign-in and private mode. Check /robots.txt and those settings on your store.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/platforms/shopify/",
    "https://ghostagentlab.com/blog/agent-readiness-for-shopify-stores/"
   ]
  },
  {
   "id": "do-i-need-an-mcp-server-for-my-website",
   "url": "https://ghostagentlab.com/learn/answers/do-i-need-an-mcp-server-for-my-website/",
   "question": "Do I need an MCP server for my website?",
   "answer": "Probably not yet, unless your basics already work. An MCP server lets AI agents call named actions, such as searching products, instead of clicking through your pages. First make sure agents can get in and read your pages, then ask your commerce platform whether it offers MCP, and start with read-only tools.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/mcp-webmcp-for-websites/",
    "https://ghostagentlab.com/blog/should-you-build-an-mcp-server-yet/"
   ]
  },
  {
   "id": "do-ai-agents-run-javascript",
   "url": "https://ghostagentlab.com/learn/answers/do-ai-agents-run-javascript/",
   "question": "Do AI agents run JavaScript?",
   "answer": "Some do, and many don't. Training crawlers, AI search crawlers and assistant fetchers usually read only the HTML your server sends, while browser agents run JavaScript but are slow and impatient. Put product name, price, availability, key specifications, shipping, returns and plan pricing in the server's HTML so every kind of agent can read them.",
   "topic": "Readability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/javascript-content-ai-agents/",
    "https://ghostagentlab.com/articles/kinds-of-ai-agents/"
   ]
  },
  {
   "id": "can-ai-agents-see-prices-that-load-with-javascript",
   "url": "https://ghostagentlab.com/learn/answers/can-ai-agents-see-prices-that-load-with-javascript/",
   "question": "Can AI agents see prices that load with JavaScript?",
   "answer": "Many can't. Agents that read only the HTML your server sends never see a price that loads from an API after the page appears, so your product can drop out of a comparison. Put the price in the server's HTML as visible text and in structured data, and make sure the two agree.",
   "topic": "Readability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/machine-readable-prices/",
    "https://ghostagentlab.com/articles/javascript-content-ai-agents/",
    "https://ghostagentlab.com/articles/structured-data-ai-shopping-agents/"
   ]
  },
  {
   "id": "do-ai-assistants-read-structured-data",
   "url": "https://ghostagentlab.com/learn/answers/do-ai-assistants-read-structured-data/",
   "question": "Do AI assistants read structured data?",
   "answer": "Yes, when it's in the HTML your server sends. Structured data gives AI agents your product's name, price, currency, stock and return terms in a standard format, so they don't have to guess from your layout. Check what your platform already outputs, fill gaps such as stock status and product codes, and keep every value matching the page.",
   "topic": "Readability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/structured-data-ai-shopping-agents/",
    "https://ghostagentlab.com/articles/machine-readable-prices/"
   ]
  },
  {
   "id": "is-agent-readiness-the-same-as-accessibility",
   "url": "https://ghostagentlab.com/learn/answers/is-agent-readiness-the-same-as-accessibility/",
   "question": "Is agent readiness the same as accessibility?",
   "answer": "They overlap a lot, and neither covers the other fully. Named buttons, labelled fields, alt text and page landmarks help screen reader users and AI agents alike. Ask for your latest accessibility audit, since its unlabelled controls are agent readiness issues too, then check access and readability separately.",
   "topic": "Navigability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/accessibility-is-agent-readiness/"
   ]
  },
  {
   "id": "can-ai-agents-use-my-site-search",
   "url": "https://ghostagentlab.com/learn/answers/can-ai-agents-use-my-site-search/",
   "question": "Can AI agents use my site search?",
   "answer": "Yes, if your search box has a label, your filters are real controls and every search has its own web address. Browser agents type into search and click filters like shoppers, while assistants that fetch pages can only open links. Give each search and filtered view a readable URL that shows the same results when opened fresh.",
   "topic": "Navigability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/site-search-filters-ai-agents/",
    "https://ghostagentlab.com/articles/javascript-content-ai-agents/"
   ]
  },
  {
   "id": "do-pop-ups-and-cookie-banners-stop-ai-agents",
   "url": "https://ghostagentlab.com/learn/answers/do-pop-ups-and-cookie-banners-stop-ai-agents/",
   "question": "Do pop-ups and cookie banners stop AI agents?",
   "answer": "They can, because a banner or pop-up an agent can't dismiss ends its visit, and you never hear about it. Keep consent banners at the edge of the screen with named \"Accept all\" and \"Reject all\" buttons, label close icons, and show sign-up pop-ups only after someone engages, agreeing any change with your privacy owner.",
   "topic": "Navigability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/cookie-banners-popups-ai-agents/",
    "https://ghostagentlab.com/articles/agent-friendly-buttons-forms/"
   ]
  },
  {
   "id": "can-ai-agents-fill-in-forms",
   "url": "https://ghostagentlab.com/learn/answers/can-ai-agents-fill-in-forms/",
   "question": "Can AI agents fill in forms?",
   "answer": "Browser agents can fill in forms when every field has a visible label and choices use standard controls. They match what the person asked for, such as \"my email\", to each field's label, so placeholder-only fields and custom pickers trip them up. Show errors as text, avoid a CAPTCHA on every submission, and confirm in words.",
   "topic": "Task completion",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/agent-friendly-sign-up-booking/",
    "https://ghostagentlab.com/articles/agent-friendly-buttons-forms/"
   ]
  },
  {
   "id": "can-ai-agents-buy-from-my-online-store",
   "url": "https://ghostagentlab.com/learn/answers/can-ai-agents-buy-from-my-online-store/",
   "question": "Can AI agents buy from my online store?",
   "answer": "Yes, in two ways. Browser agents can shop your website like a person and usually hand back to the person to pay, while agentic commerce protocols let an assistant buy through a direct integration. Most agents still shop through the website, so make every step from product to payment usable and offer guest checkout.",
   "topic": "Task completion",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/agent-ready-checkout/",
    "https://ghostagentlab.com/articles/agentic-payments-ai-agents/",
    "https://ghostagentlab.com/articles/guest-checkout-ai-agents/"
   ]
  },
  {
   "id": "what-is-agentic-commerce",
   "url": "https://ghostagentlab.com/learn/answers/what-is-agentic-commerce/",
   "question": "What is agentic commerce?",
   "answer": "Agentic commerce means purchases an AI agent makes for a shopper. Agents buy either by using your website like a person or through protocols such as ACP, UCP and AP2 that skip your checkout pages. Fix your website journey and product data first, then ask your platform and payment provider which protocols they support.",
   "topic": "Task completion",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/agentic-commerce-protocols/",
    "https://ghostagentlab.com/articles/agentic-payments-ai-agents/",
    "https://ghostagentlab.com/articles/agent-ready-checkout/"
   ]
  },
  {
   "id": "how-do-i-know-if-ai-agents-visit-my-website",
   "url": "https://ghostagentlab.com/learn/answers/how-do-i-know-if-ai-agents-visit-my-website/",
   "question": "How do I know if AI agents visit my website?",
   "answer": "Look in your server or CDN logs, which record every request whether or not any JavaScript runs. Count requests by agent name, then verify each agent against its operator's published IP ranges or reverse DNS, because impostors copy well-known names. Track which agents visit, what they read and where they're blocked.",
   "topic": "Measurement",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/measure-ai-agent-traffic/",
    "https://ghostagentlab.com/articles/verify-ai-crawlers/",
    "https://ghostagentlab.com/blog/agent-traffic-is-not-bot-traffic/"
   ]
  },
  {
   "id": "why-doesnt-google-analytics-show-chatgpt-visits",
   "url": "https://ghostagentlab.com/learn/answers/why-doesnt-google-analytics-show-chatgpt-visits/",
   "question": "Why doesn't Google Analytics show ChatGPT visits?",
   "answer": "When ChatGPT reads your page for a user, it usually downloads the HTML without running your analytics tag, and analytics tools filter out known bots anyway. Find those visits in your server or CDN logs. People who click a link in ChatGPT's answer do appear in GA4, often as chatgpt.com under Referral.",
   "topic": "Measurement",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/measure-ai-agent-traffic/",
    "https://ghostagentlab.com/articles/ai-referral-traffic/"
   ]
  },
  {
   "id": "how-often-should-i-check-my-sites-agent-readiness",
   "url": "https://ghostagentlab.com/learn/answers/how-often-should-i-check-my-sites-agent-readiness/",
   "question": "How often should I check my site's agent readiness?",
   "answer": "Rescan at least monthly and after every significant release, because one theme update or bot protection change can undo months of work. Scan on the same day each month so the numbers compare fairly. Keep AI agent tests running on your money journeys, and benchmark against competitors every quarter.",
   "topic": "Measurement",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/blog/90-day-agent-readiness-plan/",
    "https://ghostagentlab.com/articles/agent-readiness-kpis/",
    "https://ghostagentlab.com/articles/benchmark-competitors-agent-readiness/"
   ]
  },
  {
   "id": "do-ai-assistants-send-real-customers-to-websites",
   "url": "https://ghostagentlab.com/learn/answers/do-ai-assistants-send-real-customers-to-websites/",
   "question": "Do AI assistants send real customers to websites?",
   "answer": "Yes. People click links in ChatGPT, Perplexity and other assistants' answers and arrive in their own browser like any other visitor. Many answers drive no click, so your reports show only part of the effect. Track AI referral visits and orders in your analytics and treat the number as a floor.",
   "topic": "Measurement",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/ai-referral-traffic/",
    "https://ghostagentlab.com/blog/agent-traffic-is-not-bot-traffic/",
    "https://ghostagentlab.com/articles/agent-readiness-kpis/"
   ]
  },
  {
   "id": "training-crawlers-vs-ai-search-crawlers-vs-assistant-fetchers",
   "url": "https://ghostagentlab.com/learn/answers/training-crawlers-vs-ai-search-crawlers-vs-assistant-fetchers/",
   "question": "What's the difference between training crawlers, AI search crawlers and assistant fetchers?",
   "answer": "Training crawlers collect pages to train AI models, AI search crawlers index pages so AI search can cite and link to you, and assistant fetchers fetch a page in real time because a person asked about it. Block training crawlers by name if you choose. Allow the other two in robots.txt and at your CDN.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/kinds-of-ai-agents/",
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/"
   ]
  },
  {
   "id": "browser-agents-vs-assistant-fetchers",
   "url": "https://ghostagentlab.com/learn/answers/browser-agents-vs-assistant-fetchers/",
   "question": "What's the difference between browser agents and assistant fetchers?",
   "answer": "An assistant fetcher reads a page in real time because a person asked an assistant about it. A browser agent opens your site in a real browser and clicks, types and fills in forms to book or buy for someone. Keep key facts in your HTML for fetchers, and keep journeys clear of CAPTCHAs and forced accounts for browser agents.",
   "topic": "Getting started",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/kinds-of-ai-agents/",
    "https://ghostagentlab.com/articles/test-journeys-with-ai-agents/"
   ]
  },
  {
   "id": "robots-txt-vs-llms-txt",
   "url": "https://ghostagentlab.com/learn/answers/robots-txt-vs-llms-txt/",
   "question": "What's the difference between robots.txt and llms.txt?",
   "answer": "robots.txt tells crawlers and agents which pages they may visit, and well-behaved ones follow it. llms.txt is a short Markdown guide that tells AI what your business does and where your key pages are. Use robots.txt to decide who gets in, use llms.txt to point agents at the right pages, and keep both at your domain root.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/",
    "https://ghostagentlab.com/articles/llms-txt/"
   ]
  },
  {
   "id": "mcp-vs-webmcp",
   "url": "https://ghostagentlab.com/learn/answers/mcp-vs-webmcp/",
   "question": "What's the difference between MCP and WebMCP?",
   "answer": "MCP and WebMCP both let AI agents call named actions, such as searching products, instead of clicking through your pages. An MCP server runs on your domain and follows a published, widely used specification, while WebMCP runs inside your web pages and is still an early proposal. Start with read-only MCP tools and treat WebMCP as an experiment.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/mcp-webmcp-for-websites/"
   ]
  },
  {
   "id": "bot-protection-vs-bot-verification",
   "url": "https://ghostagentlab.com/learn/answers/bot-protection-vs-bot-verification/",
   "question": "What's the difference between bot protection and bot verification?",
   "answer": "Bot protection blocks or challenges automated visitors to keep out scrapers, credential stuffers and fraud. Verification checks that a visitor claiming to be a known agent really is, using the proof its operator publishes. Use verification inside your bot protection, so verified AI assistants and AI search get through and impostors using their names don't.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/bot-protection-ai-agents/",
    "https://ghostagentlab.com/articles/verify-ai-crawlers/"
   ]
  },
  {
   "id": "claudebot-vs-claude-user-vs-claude-searchbot",
   "url": "https://ghostagentlab.com/learn/answers/claudebot-vs-claude-user-vs-claude-searchbot/",
   "question": "What's the difference between ClaudeBot, Claude-User and Claude-SearchBot?",
   "answer": "ClaudeBot, Claude-SearchBot and Claude-User are Anthropic agents, each with its own robots.txt name. ClaudeBot collects public content that may be used to train Anthropic's models; Claude-SearchBot crawls to improve Claude's search results; Claude-User fetches a page when someone using Claude asks about it. Block ClaudeBot if you choose, and allow the other two.",
   "topic": "Access",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/agents/companies/anthropic/",
    "https://ghostagentlab.com/articles/robots-txt-ai-agents/"
   ]
  },
  {
   "id": "structured-data-vs-product-feeds",
   "url": "https://ghostagentlab.com/learn/answers/structured-data-vs-product-feeds/",
   "question": "What's the difference between structured data and product feeds?",
   "answer": "Structured data is machine-readable product facts in your page's HTML, which agents read when they open the page. A product feed sends your whole catalog to a shopping platform, such as Google Merchant Center, on a schedule. A feed adds to your pages, so keep both and generate the feed, structured data and visible page from one product record.",
   "topic": "Readability",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/structured-data-ai-shopping-agents/",
    "https://ghostagentlab.com/articles/product-feeds-ai-shopping/"
   ]
  },
  {
   "id": "acp-vs-ucp-vs-ap2",
   "url": "https://ghostagentlab.com/learn/answers/acp-vs-ucp-vs-ap2/",
   "question": "What's the difference between ACP, UCP and AP2?",
   "answer": "ACP, UCP and AP2 are agentic commerce protocols that let an AI assistant buy from your store for a person. ACP (OpenAI and Stripe) covers checkout inside an assistant; UCP (Google) covers discovery and checkout; AP2 proves a person authorized the agent's payment. Most stores get support through their commerce platform or payment provider.",
   "topic": "Task completion",
   "verified": "2026-11-12",
   "basedOn": [
    "https://ghostagentlab.com/articles/agentic-commerce-protocols/"
   ]
  }
 ]
}
