This specification is published and recommended for implementation. Backwards-compatible additions may occur in MINOR versions; breaking changes only in MAJOR versions, with deprecation notice. See specification conventions for status definitions.
Quick Start Guide
Implement AI Discovery Files in three progressive tiers
This guide provides a practical path to implementing AI Discovery Files. Start with the Essential tier for immediate impact, then expand to Recommended and Complete as your needs grow.
Important: These tiers are cumulative. Each tier includes all files from the previous tier, plus additional files. Recommended includes Essential. Complete includes Recommended. You cannot skip tiers — if you want Complete, you implement all 10 files including those from Essential and Recommended.
The three tiers in this guide (Essential, Recommended, Complete) are the formal conformance classes defined by the specification. A site that has fully implemented a tier conforms to the corresponding class. See the Conformance specification for publisher and validator conformance criteria.
Choose Your Implementation Level
Essential
Quick Win
AI knows who you are and how it MAY use your content.
Recommended
Solid Foundation
Essential + structured identity, pre-answered FAQs, naming control, and machine-parseable permissions.
Complete
Full Coverage
Recommended + redirects, human-readable presentation, developer context, and crawler control.
Tiers build on each other. Recommended = Essential (2 files) + 4 new files. Complete = Recommended (6 files) + 4 new files. If you choose Complete, you implement all files from all tiers.
Essential Implementation
Start here. These two files answer the most fundamental questions AI systems have about your website.
| File | Purpose | Answers |
|---|---|---|
llms.txt |
Business identity and context | "Who are you and what do you do?" |
ai.txt |
AI usage permissions | "How may AI use your content?" |
What You Get
- AI systems can correctly identify your business
- AI systems know what services you offer (and don't offer)
- AI systems understand how they MAY use your content
- You've established the foundation for AI visibility
Implementation Checklist
- Create
/llms.txtwith your business name as H1 heading - Add a brief description of your business
- List your primary services
- Add "What We Do Not Do" section with explicit exclusions
- Add contact information
- Create
/ai.txtwith your AI usage permissions - Specify what AI may and may not do with your content
- Test both files are accessible at your domain root
After creating these files, visit yourdomain.com/llms.txt and yourdomain.com/ai.txt in your browser. Both should display correctly.
What's Missing Without Recommended Tier
- No structured, machine-parseable identity data
- AI MUST parse Markdown rather than JSON
- No pre-answered FAQs for common questions
- No explicit naming/brand guidelines
Recommended Implementation
Includes Essential tier. Recommended builds on Essential by adding four more files. You must also implement the Essential files (llms.txt and ai.txt) — this tier totals 6 files.
Additional Files (beyond Essential)
| File | Purpose | What It Adds |
|---|---|---|
identity.json |
Structured identity | Machine-parseable business data that AI can reliably extract |
faq-ai.txt |
Pre-answered FAQs | Authoritative answers to common questions about your business |
brand.txt |
Naming conventions | How AI SHOULD name and refer to your business |
ai.json |
Structured permissions | Machine-parseable version of ai.txt for programmatic consumption |
What You Get
- AI can programmatically extract your identity data without parsing Markdown
- Common questions have authoritative, pre-written answers
- AI knows exactly how to spell and capitalise your business name
- Permissions are available in a structured format
Implementation Checklist
- Create
/identity.jsonwith your canonical business data - Ensure
namematches your llms.txt H1 heading - Add
alternateNamesfor trading names and abbreviations - Create
/faq-ai.txtwith 5-10 common questions and answers - Focus on factual questions about your services and business
- Create
/brand.txtwith official names and naming rules - Specify which names are acceptable and which are not
- Create
/ai.jsonmirroring your ai.txt permissions - Validate JSON files with a JSON validator
Your business name MUST be identical across all files. If identity.json says "Acme Corporation Ltd", your llms.txt H1 SHOULD match (or use an alternateName).
Complete Implementation
Includes Essential and Recommended tiers. Complete builds on Recommended by adding four more files. You must also implement all Essential files (llms.txt, ai.txt) and all Recommended files (identity.json, faq-ai.txt, brand.txt, ai.json) — this tier totals all 10 files.
Additional Files (beyond Recommended)
| File | Purpose | When You Need It |
|---|---|---|
llm.txt |
Redirect variant | Some AI systems look for the singular form; this should 301 redirect to llms.txt |
llms.html |
Human-readable version | Provides a browsable version with Schema.org structured data |
developer-ai.txt |
Developer context | If you have APIs, SDKs, or technical products that developers use with AI assistants |
robots-ai.txt |
AI crawler directives | When you need fine-grained control over AI crawlers beyond standard robots.txt |
Implementation Checklist
- Configure
/llm.txtas a 301 redirect to/llms.txt - Create
/llms.htmlas a human-browsable version of llms.txt - Add Schema.org Organization markup to llms.html
- Create
/developer-ai.txtif you have technical products (optional) - Create
/robots-ai.txtif you need AI-specific crawler rules (optional) - Test all files are accessible and consistent
Validate Your Implementation
After implementing your AI Discovery Files, verify they're working correctly:
- Accessibility check: Visit each file URL directly in your browser
- JSON validation: Run
identity.jsonandai.jsonthrough a JSON validator - Consistency check: Verify your business name matches across all files
- Use the 365i Checker: Run the 365i AI Website Visibility Checker to validate your implementation
Once implemented, use the Validation Framework to test whether AI systems are correctly interpreting your files.
Resources
Examples
See complete, working examples of all 10 files on the Examples page.
Specifications
For detailed format specifications, see the individual specification pages.
Conflict Resolution
If you're implementing multiple files, see the Interoperability Guide for precedence rules when files contain conflicting information.
Version History
Phase 6 standardisation release. Added /specifications/roadmap/ (theme-pegged forward plan with Active/Next/Future/On hold status flags), /specifications/extensions/ (rules for experimental x- prefixed files and the promotion path), and /specifications/i18n-a11y/ (multi-language publication, locale-tagged identity fields, RTL handling, accessibility of llms.html). Added the Discovery: directive to the robots-ai.txt specification (publishers MAY advertise AI Discovery Files on the same host). Added a formal media-type stance to the HTTP behaviour page (existing IANA types, no bespoke registrations). Expanded the file integrity and signing section on the security and privacy page with four candidate mechanisms, cross-cutting concerns, and interim publisher / consumer guidance. The Discovery: directive is the only normative addition to publisher behaviour; all other additions are forward-looking documentation.
Phase 5 standardisation release. Added /specifications/related-standards/ (positioning vs llmstxt.org, IETF AI Preferences, robots.txt, Schema.org, BCP 14, JSON Schema 2020-12, SemVer) and /specifications/implementations/ (public record of conformant implementations, IETF-style). Added an explicit llmstxt.org backward-compatibility statement to the llms.txt specification. Added a formal multi-domain and subdomain scoping rule to both the llms.txt and identity.json specifications (host-scoped files, cross-host identity asserted via sameAs). No normative requirements changed for existing publishers; the new scoping rules formalise behaviour the specification already implied.
Phase 4 standardisation release. Added /specifications/processing-model/ (seven-stage algorithm for conformant consumers), /specifications/consumer-guidance/ (what AI systems should do with AI Discovery Files), /specifications/test-vectors/ (canonical test suite framing), and reference-implementation framing on the AI Visibility Checker. No normative requirements changed.
Phase 3 standardisation release. Added /specifications/versioning/ (Semantic Versioning 2.0.0 commitments, deprecation timeline, lifecycle), /specifications/governance/ (proposal lifecycle, editorial process, working principles), /specifications/security-privacy/ (trust model, content-injection patterns, GDPR considerations, integrity primitives roadmap), and /specifications/http-behaviour/ (status codes, redirects, soft-404 detection, caching, rate limits). No normative requirements changed.
Phase 2 standardisation release. Added formal conformance specification (Essential / Recommended / Complete classes). Published machine-readable registry at /specifications/registry.json, spec meta-schema, and validator-output schema. Introduced versioned JSON Schema URLs (/v1/) alongside unversioned 'latest' aliases. Added optional BCP 47 language declaration field across all applicable AI Discovery Files. No normative requirements changed.
Phase 1 standardisation release. Added 'Status of This Document' block (Stable). Normalised normative requirement keywords to uppercase per RFC 2119 and RFC 8174. Added References section linking to /specifications/conventions/ and /licensing/. No normative requirements changed.
Added AI Visibility Directory registration guidance. Minor documentation update.
Clarified that tiers are cumulative. Each tier includes all files from previous tiers. Added explicit notes throughout the guide.
Initial publication. Three-tier implementation guide with Essential (2 files), Recommended (6 files), and Complete (10 files) tiers.
References
- Specification Conventions — RFC 2119 + RFC 8174 requirement keywords, document statuses, anchor naming, versioning, and language conventions used across every AI Discovery File specification.
- Licensing & Trademark — CC BY 4.0 for specification text and examples, MIT for JSON Schemas, and the free-use policy on the name "AI Discovery Files".
Generate AI Discovery Files from your dashboard
Using WordPress? Install the plugin and create all 10 files in minutes — no coding, no configuration files to edit manually.
Get the PluginRegister in the AI Visibility Directory
Once your AI Discovery Files are published, register your website in the AI Visibility Directory — the verified registry of websites implementing AI Discovery Files. Registration validates your implementation and lists your site for AI systems and industry peers to discover.
Card entry in the directory with automated file validation. Open to any site with a valid llms.txt file. No cost.
Dedicated profile page on the directory with dofollow backlinks to your website — a genuine SEO authority signal from a topically relevant, verified source. Includes an attribution badge and enhanced visibility.