Audit any URL for AI search readiness. Twenty checks. No AI.
An AI readiness audit checks whether a website can be found, read, and cited by AI systems. This free auditor runs twenty technical checks against any URL: HTTPS, Time-To-First-Byte, robots.txt access for AI crawlers, robots.txt syntax, sitemap validity, mobile viewport, html lang, core semantic elements, HTML5 landmarks, canonical tag, Open Graph tags, image alt text, heading hierarchy, JSON-LD entity schema, schema required properties, content chunk size, content-to-HTML ratio, Last-Modified header, OKF bundle, and the four AI discovery files. The result is a score out of 100 with a specific remediation for each failure.
This auditor tests the requirements described in the AI-Ready Website Guide for Philippine Businesses.
| Target URL | Score | Load Time | TTFB | Timestamp |
|---|---|---|---|---|
| https://imanila.ph/ | 71/100 | 4.059s | 3.822s | 2026-10-15 07:49:33 |
| https://www.facebook.com/ | 31/100 | 0.509s | 0.14s | 2026-10-15 06:38:21 |
| https://anahawdigital.com/ | 76/100 | 1.9s | 0.967s | 2026-10-15 05:49:17 |
| https://www.securitybank.com/ | 33/100 | 6.963s | 6.806s | 2026-10-15 05:48:50 |
| https://www.wholesaledito.store/ | 100/100 | 0.078s | 0.078s | 2026-10-15 05:48:02 |
The auditor runs twenty checks. Each one is a specific technical condition that affects whether an AI system can read and cite the target page.
robots.txt and reports access for twelve AI crawlers grouped by training, search-index, and user-retrieval roles./sitemap.xml and /sitemap_index.xml and confirms the XML shape.<meta name="viewport">.lang attribute on the <html> tag.<title>, meta description, and primary <h1>.<main>, <section>, <article>, <nav>, and <aside>.og:title and related tags.@type values./.okf/index.md.llms.txt, llms-full.txt, /.well-known/ai.txt, and /.well-known/ard.json.Time-To-First-Byte is the interval between when a client sends an HTTP request and when the server begins to send the first byte of the response. It includes DNS resolution, TCP connection, TLS handshake, and server processing time. A TTFB above 1.5 seconds suggests that fast automated agents may time out.
llms.txt is a plain-text file at the domain root that gives large language models a curated map of a site's most important pages. It is a Markdown document with an H1 heading, a blockquote summary, and lists of links. It is a discovery aid for AI systems, not a search ranking factor.
llms-full.txt is the companion to llms.txt. Where llms.txt is a map, llms-full.txt contains the actual body text of the site's key pages so that a retrieval-augmented generation pipeline can ingest everything in one fetch.
ai.txt is a machine-readable usage policy for AI systems. It declares training, scraping, indexing, and caching permissions, path-level allow and deny rules, and per-agent overrides. The format is defined by the IETF Internet-Draft draft-car-ai-txt-wellknown-00.
ard.json is the manifest used by the Agentic Resource Discovery specification. It is published at /.well-known/ard.json and lists the AI-callable resources a domain offers, such as MCP servers, A2A agents, skills, APIs, and nested catalogs. The predecessor path is /.well-known/ai-catalog.json.
The Open Knowledge Format is a directory of Markdown files with YAML frontmatter, each describing a single concept. It is a newer discovery path that sits alongside llms.txt. The canonical location is /.okf/index.md. Most sites do not have this yet.
A JSON-LD entity schema is a structured data block that tells AI systems what an entity on a page is. Common types are Organization, LocalBusiness, WebSite, FAQPage, and WebApplication. AI systems and search engines use these blocks to categorize and cite content.
The matrix simulates whether each of twelve AI crawlers is permitted to fetch the target site. Ten are fully checked against robots.txt. Two (Google-Extended and Applebot-Extended) are robots.txt tokens only, not live crawler user-agents, and are reported accordingly.
A sequential heading hierarchy means that heading levels do not skip. A page that goes from an H1 straight to an H3 is flagged because the gap leaves ambiguity about the relationship between the sections. AI systems use headings as structural clues about content.
An AI readiness audit checks whether a website can be found, read, and cited by AI systems. It verifies twenty technical conditions covering infrastructure, semantic HTML, structured data, content structure, and AI discovery files.
The auditor checks for five discovery paths: llms.txt at the domain root, llms-full.txt at the domain root, ai.txt at /.well-known/ai.txt, ard.json at /.well-known/ard.json, and an OKF bundle at /.okf/index.md. The OKF bundle is optional and most sites do not yet have it.
No. The auditor is deterministic. Every check is a specific HTTP request, regex match, or DOM query. There is no AI, no inference, and no interpretation. The score is the sum of specific checks.
AI crawlers often have tighter timeouts than traditional search crawlers. If Time-To-First-Byte exceeds 1.5 seconds, faster crawlers may time out before the page begins to render. TTFB below 0.8 seconds is optimal.
The matrix simulates whether twelve AI crawlers are allowed to fetch the target site according to its robots.txt. Crawlers are grouped by role: Training (model training), Search-index (AI search answers), and User-retrieval (live user-triggered fetches).
llms.txt is a plain-text file at the domain root that gives AI systems a curated map of a site's most important pages. It is a discovery aid, not a ranking factor.
ard.json is the manifest used by the Agentic Resource Discovery specification. It is published at /.well-known/ard.json and lists the AI-callable resources a domain offers. The predecessor path is /.well-known/ai-catalog.json.
No. Passing the audit means the site has removed technical barriers that would prevent AI systems from reading it. It does not guarantee that AI systems will cite the site, rank it, or send traffic to it. AI visibility depends on content authority, structure, and citation by others.
Thresholds used. TTFB above 1.5 seconds is flagged. HTTPS is required for a pass. Content-to-HTML ratio below 10% is flagged. Content chunks over 150 words are flagged. Viewport and lang are binary checks. Schema property checks use a fixed map of common types and their required properties.
Deduction table. HTTPS 10. TTFB 5. robots.txt syntax 5. Sitemap 5. Viewport 5. html lang 3. Core semantic elements 10. HTML5 landmarks 5. Canonical 5. Open Graph 5. Image alt text 5. Heading hierarchy 5. Entity schema missing 15. Schema required properties 5. Oversized chunks 5. Content ratio 3. Last-Modified 2. OKF 2. Each missing discovery file 5.
Discovery files. The auditor performs a HEAD request against each discovery path. A 200 response is a pass.
What the audit does not do. It does not benchmark against other sites. It does not check JavaScript execution. It does not evaluate content quality. It does not verify that a referenced sitemap is the correct one for the domain.
Tool version 6.0. Last updated 2026-10-15.
Published by The Sniffer, the strategic insights blog of Wholesale Dito Store. Operated by Clickerwayne Zelle Solutions Inc, Forest Drive St., corner Country Drive, Country Homes, Binan, Laguna 4024, Philippines. Questions: Contact us.
This tool is part of the Ask W [Not AI] family. No AI is used. The logic and data are public. Every result can be reproduced by hand. The public tools use the same logic as the proprietary engine used internally by Clickerwayne Zelle Solutions Inc. The code and execution differ, but the logic is identical. The Ask W [Not AI] tools are provided as a free public service for Philippine MSMEs, freelancers, and small business owners. No Ads. No registration is required.