anslyansly
Free AuditWhy DGOSHow It WorksModulesDashboard PreviewOutcomes
PricingBlog
Sign In
anslyansly

Ansly DGOS — the operating system for digital growth. Start free with the AEO Audit tool.

@tryansly

Product

  • Audit
  • Modules
  • How It Works
  • Pricing
  • Blog
  • Changelog

Company

  • About
  • Privacy Policy
  • Terms of Service
  • Contact Us
© 2026 ansly. All rights reserved.
PrivacyTermsContact
anslyansly
Free AuditWhy DGOSHow It WorksModulesDashboard PreviewOutcomes
PricingBlog
Sign In
  1. Home
  2. /
  3. Blog
  4. /
  5. AEO Checklists
  6. /
  7. AI Crawler Access
AEO Category17% of overall AEO score

AI Crawler Access Checklist (2026)

AI Crawler Access is about permissions — making sure you have not accidentally blocked the AI crawlers that power ChatGPT, Perplexity, Gemini, and Claude from reading your content. Worth 17% of your AEO score, this category is the easiest to fix and the most commonly broken.

Category weight17%
12
Total checks
7
High priority
10
Quick wins

High Priority

7 checks

Create a valid robots.txt

high priorityQuick win

Publish a robots.txt file at the root of your domain. It must be accessible at https://yourdomain.com/robots.txt.

How to implement ▸
Create a robots.txt file. At minimum: User-agent: *\nAllow: /\nSitemap: https://yourdomain.com/sitemap.xml

Why AI cares: Without robots.txt, AI crawlers fall back to default behavior and may not index everything. An explicit file signals you are in control.

Allow OpenAI's search crawlers

high priorityQuick win

Ensure OAI-SearchBot and ChatGPT-User are not blocked in robots.txt.

How to implement ▸
Remove any "Disallow:" rules under "User-agent: OAI-SearchBot" and "User-agent: ChatGPT-User".

Why AI cares: OAI-SearchBot builds the index ChatGPT cites from and ChatGPT-User fetches pages live during a conversation. These decide whether ChatGPT can cite you. GPTBot is a third, separate agent used only for model training — blocking it is a legitimate choice that costs no citation visibility.

Allow Anthropic's search crawlers

high priorityQuick win

Ensure Claude-SearchBot and Claude-User are not blocked.

How to implement ▸
Remove any "Disallow:" under "User-agent: Claude-SearchBot" and "User-agent: Claude-User".

Why AI cares: Claude-SearchBot indexes for Claude's search and Claude-User fetches pages live when someone asks. These decide whether Claude can cite you. ClaudeBot is a separate training crawler — blocking it costs no citation visibility.

Allow Perplexity's crawlers

high priorityQuick win

Ensure PerplexityBot and Perplexity-User are not blocked. Perplexity is one of the highest-citation-rate AI search engines.

How to implement ▸
Remove any "Disallow:" under "User-agent: PerplexityBot" and "User-agent: Perplexity-User".

Why AI cares: Perplexity is the AI search engine most likely to cite your content directly in answers. Blocking it removes you from a major citation channel.

Remove X-Robots-Tag: noai headers

high priorityModerate effort

Check that your server is not sending X-Robots-Tag: noai or X-Robots-Tag: noimageai HTTP headers.

How to implement ▸
Audit your server/CDN response headers. Remove any noai directives from nginx, Apache, or CDN configs.

Why AI cares: The noai header directive explicitly tells AI systems not to use your content — even if robots.txt allows them to crawl it.

Remove noindex meta tags from key pages

high priorityQuick win

Ensure your homepage, pricing, product, and about pages do not have <meta name="robots" content="noindex">.

How to implement ▸
Check page source for <meta name="robots"> tags. Remove or change to "index, follow" on all public pages.

Why AI cares: Noindex blocks both search engine and AI crawler indexing. It is commonly accidentally applied to important pages.

Ensure key pages return HTTP 200

high priorityQuick win

Homepage, /pricing, /about, /blog, and /contact should all return 200 status codes.

How to implement ▸
Test with: curl -I https://yourdomain.com/pricing. Fix any 404s, 403s, or redirect loops.

Why AI cares: AI crawlers skip non-200 pages entirely. A broken pricing page means AI models cannot cite your pricing.

Medium Priority

3 checks

Reference sitemap.xml in robots.txt

medium priorityQuick win

Add a Sitemap: directive pointing to your sitemap at the bottom of robots.txt.

How to implement ▸
Add "Sitemap: https://yourdomain.com/sitemap.xml" to the end of robots.txt.

Why AI cares: Sitemap discovery via robots.txt is the most reliable mechanism for crawlers to find your complete content inventory.

Set a permissive max-snippet policy

medium priorityQuick win

Avoid max-snippet:0 — set max-snippet:-1 (unlimited) or a large value like max-snippet:200.

How to implement ▸
In robots.txt or meta robots: "max-snippet: -1". In Next.js: robots: { index: true, "max-snippet": -1 }

Why AI cares: max-snippet:0 tells AI models they cannot quote any text from your page, effectively preventing citations.

Fix redirect chains

medium priorityModerate effort

Ensure there are no chains of 3+ redirects before reaching a final page.

How to implement ▸
Test with a redirect checker tool. Simplify to a single 301 redirect wherever possible.

Why AI cares: Long redirect chains slow crawlers and increase the chance they abandon before indexing the final URL.

Lower Priority

2 checks

Decide your Google-Extended opt-out

low priorityQuick win

Google-Extended controls Gemini and Vertex AI training and grounding. It does not affect Google Search or AI Overviews.

How to implement ▸
Leave "User-agent: Google-Extended" unblocked to allow Gemini training, or add "Disallow: /" to opt out. Either way, Google Search and AI Overviews are unaffected — those crawl through Googlebot.

Why AI cares: This is a training opt-in decision, not a visibility one. Unblocking Google-Extended will not add you to AI Overviews, and blocking it will not remove you.

Allow large image previews

low priorityQuick win

Set max-image-preview: large to allow AI systems to use your images.

How to implement ▸
Add to meta robots: max-image-preview:large. This is the default for most sites unless explicitly restricted.

Why AI cares: Multimodal AI models may use your images to understand your product. Restricting previews reduces your visual footprint.

Frequently Asked Questions

Which AI crawlers should I allow in robots.txt?+

Split them by what they do. The agents that decide whether you can be cited are OAI-SearchBot and ChatGPT-User (ChatGPT), Claude-SearchBot and Claude-User (Claude), and PerplexityBot and Perplexity-User (Perplexity) — allow these unless you want to be absent from those answer engines. The training crawlers are GPTBot, ClaudeBot, Google-Extended (Gemini and Vertex AI, not Search), Applebot-Extended, Amazonbot, meta-externalagent, CCBot, and Bytespider — blocking any of these is a legitimate choice that costs no citation visibility.

Will blocking AI crawlers prevent AI models from using my content?+

For future training data, yes. But AI models already trained on your content (before your block) will still reference it. For real-time web search features (Perplexity, ChatGPT browsing, Gemini), blocking crawlers removes you from live citations going forward.

Should I block AI crawlers to protect my content?+

For most B2B companies, blocking AI crawlers is counterproductive. Being cited by AI models drives awareness and qualified traffic. The exception is if your content is a core IP asset and AI reproduction would directly harm your business model — in that case, consult legal counsel.

Find out your AI Crawler Access score — free

ansly audits your site across all 7 AEO categories including AI Crawler Access. Get your score in under 60 seconds.

Audit my site free →

Other AEO Categories

Structured Data
23% of AEO score · 9 checks
AI Agent Readiness
23% of AEO score · 9 checks
Content Extractability
16% of AEO score · 8 checks
Semantic HTML
13% of AEO score · 10 checks
LLMs.txt & LLM Content
8% of AEO score · 6 checks
SEO Fundamentals
6 checks
anslyansly

Ansly DGOS — the operating system for digital growth. Start free with the AEO Audit tool.

@tryansly

Product

  • Audit
  • Modules
  • How It Works
  • Pricing
  • Blog
  • Changelog

Company

  • About
  • Privacy Policy
  • Terms of Service
  • Contact Us
© 2026 ansly. All rights reserved.
PrivacyTermsContact