GEOAssetGenerator
Instantly generate the core files needed to communicate with AI Agents.
Configure Bot Access
Control which AI models are allowed to learn from your data.
# Robots.txt for yourdomain.com User-agent: * Allow: / # AI Agents & Crawlers User-agent: GPTBot Allow: / User-agent: ClaudeBot Allow: / User-agent: Google-Extended Allow: / Sitemap: https://yourdomain.com/sitemap.xml
Technical Disclaimer
This tool generates standardized baseline configurations suitable for general web properties. Complex single-page applications (SPAs), enterprise e-commerce sites, or sites with specific legal compliance needs may require custom implementation by a technical SEO specialist. Always test generated code in a staging environment before deploying to production.
Coverage
Everything covered, deterministically
No LLM calls, no API limits. Type in, copy out, ship it.
JSON-LD Schema
BlogPosting, HowTo, FAQPage, Organization, Product, SoftwareApplication.
FAQ Blocks
Schema + matching HTML so AI engines can verify the schema matches visible content.
llms.txt File
Follows the emerging convention. Tells AI assistants what your site is and what to prioritize.
Meta Tags
SEO meta + Open Graph + Twitter Card, with character counters so you stay within limits.
Reference
Quick reference for each asset type
When to use which, what AI engines do with each, and the gotchas worth knowing before you ship.
Schema markup (JSON-LD)
What it does. Tells AI engines what type of content lives on the page (article, product, software, organization, FAQ). Schema gives engines structured fields they can pull into rich results, AI Overviews, and citation cards without parsing the visible HTML.
When to use which type. Use BlogPosting on articles. Use HowTo on step-by-step guides; it powers AI Overview step cards. Use FAQPage wherever you have a real FAQ section (don't fake one). Use Organization on your homepage with a complete sameAs array linking to your LinkedIn, X, GitHub. Use Product for e-commerce items and SoftwareApplication for SaaS pricing pages.
The most common mistake. Schema that doesn't match the visible page. Google's documentation explicitly warns against schema-only content (Q&A that exists in JSON-LD but nowhere on the page). AI engines penalize this aggressively. Always make schema mirror what users see.
FAQ blocks (schema + matching HTML)
What it does. Wraps your Q&A content in both FAQPage JSON-LD and matching itemscope HTML attributes. AI engines pull FAQ blocks into snippet results and AI summaries with high frequency. A well-structured FAQ section consistently outperforms long-form prose at the same word count.
What makes FAQs cite-worthy. Questions should be phrased the way real users ask them, not the way you'd phrase them in a marketing brief. Answers should be direct: the first sentence answers the question, the rest provides supporting context. Avoid hedging language. AI engines prefer atomic, decisive answers.
What this tool generates. The schema block goes in your <head>. The HTML block goes wherever you want the FAQ visible on the page. Both are required; AI engines verify the schema by parsing the HTML.
llms.txt
What it does. A markdown file at the root of your domain that tells AI assistants what your site is and which pages matter most. Proposed by Jeremy Howard in September 2024 as a complement to robots.txt and sitemap.xml. Where robots.txt says what can be crawled, llms.txt says what should be prioritized.
The honest 2026 picture. Adoption sits around 10% of domains. Major AI crawlers (GPTBot, ClaudeBot, PerplexityBot) overwhelmingly skip the file and crawl HTML directly. SEO-citation studies show no measurable improvement from adding one. So why ship it? IDE agents fetch it routinely. MCP servers route on it. The agentic-web ecosystem treats it as infrastructure, even when search engines don't.
When the upside lands. When users prompt Cursor or Claude Code with "look at the docs for this library", the agent fetches library.com/llms.txt first to know which pages to read. If your site doesn't have one, the agent guesses. If it does, you control the routing.
Meta tags (SEO + OG + Twitter)
What it does. Three layers in one block: primary SEO (title, description, canonical, robots), Open Graph (controls how the page renders when shared on Facebook, LinkedIn, Discord, Slack), and Twitter Card (controls how the page renders on X/Twitter). AI engines use all three; the title and description directly affect AI snippet text.
The constraints worth knowing. Titles over 60 characters get truncated in Google search results. Descriptions over 155 characters get truncated. The tool's character counters turn red when you cross these limits. Open Graph images should be 1200x630 pixels; smaller images get cropped or skipped on most platforms.
The fields that punch above their weight. Canonical URL is critical when your CMS generates multiple URLs for the same content. og:image drives social click-through more than any other single factor. Twitter handle in twitter:site attributes posts to your account when shared.
FAQ
Quick questions
Why use JSON-LD instead of Microdata or RDFa?+
Google explicitly prefers JSON-LD. It lives in a single script block, doesn't pollute your HTML, and is easier to maintain. Microdata and RDFa are still supported but get less reliable handling from AI engines.
Is llms.txt a real standard?+
It's an emerging convention proposed by Jeremy Howard in September 2024, not a W3C standard. As of mid-2026, adoption sits around 10% of domains and the measured SEO citation impact is low. However, IDE agents (Cursor, Claude Code, Windsurf, GitHub Copilot) and MCP servers route on it routinely. The cost of adding one is trivial; the upside compounds as the agent ecosystem grows.
Does this tool call an LLM or external API?+
No. Everything generated here is deterministic template output. The values come from what you type in. No API calls, no rate limits, no surprises. For LLM-powered content generation (where the LLM writes the article and FAQ answers from your topic), use the GIST content generator instead.
Can I trust the output without validating?+
The tool checks for required fields and JSON validity before showing output, but you should always run schema through Google's Rich Results Test before deploying to production. Validation is one step you should not skip.
Which schema types should every site have?+
At minimum: Organization schema on your homepage, BlogPosting on every article, BreadcrumbList on non-root pages. Add FAQPage where you have FAQ sections, HowTo for step-by-step guides, Product or SoftwareApplication on commercial pages. Nest multiple types into a single @graph wrapper to give AI engines a clean entity map.