Specifications

The original file proposal

Our earlier experiment grouped eleven files into three layers. These definitions record that proposal, including requirements we no longer recommend.

Specifications on GitHub Templates repository Read the current guide

Layer 1

Core Identity

The original identity and navigation group. These files are not prerequisites for AI visibility.

llms.txt · llm.txt · llms-full.txt · llms.html · identity.json · ai.json

Layer 2

Guidance & Control

Proposed naming and policy hints, without established enforcement by AI providers.

ai.txt · brand.txt · robots-ai.txt

Layer 3

Enhancement

Proposed formats for publishing FAQ and technical summaries.

faq-ai.txt · developer-ai.txt

Priorities in the original proposal

Required: Required only within the original proposal; not a requirement for AI discovery.
Recommended: Recommended by the original proposal, not by this current guide.
Optional: An optional part of the original experiment.
Conditional: A condition defined by the original proposal.

Layer 1

Core Identity

llms.txt

Legacy: Required

An optional navigation file, historically used here for identity.

The earlier proposal added identity fields to llms.txt. Those additions are not required for AI visibility. For a new implementation, consult the current upstream proposal and your intended consumer.

Format: Markdown, UTF-8, served as text/plain
Location: https://example.com/llms.txt

Original format requirements

An H1 with your brand name
A one-sentence factual summary as a blockquote
The Core Identity block, word for word consistent across files
A Last updated date

Also worth reading

llmstxt.org The upstream format specification by Jeremy Howard, which this file follows.
Lighthouse: llms.txt audit Chrome DevTools audits for this file under Agentic browsing. A 404 is reported as Not Applicable, since publishing it is currently optional — but a server error is flagged.

Full specification Template

llm.txt

Legacy: Required

The proposed singular-name alias.

The original suite required a redirect or identical copy. This guide does not recommend adding that alias without a specific need.

Format: 301 redirect to llms.txt, or an identical copy
Location: https://example.com/llm.txt

Original format requirements

Same content as llms.txt, or a 301 to it
Never a second, diverging source of truth

Full specification Template

llms-full.txt

Legacy: Conditional

The expanded companion, when one file is not enough.

Use when a concise llms.txt cannot cover your important public pages clearly.

Follows the same Markdown format as llms.txt but carries deeper context and broader link coverage. It exists so llms.txt can stay short and authoritative rather than growing into a sitemap.

Format: Markdown, UTF-8, served as text/plain
Location: https://example.com/llms-full.txt

Original format requirements

The same Core Identity block as llms.txt
Broader link coverage, not contradictory detail
A link back to llms.txt as canonical

Full specification Template

llms.html

Legacy: Recommended

The same information, readable by people.

A browser-friendly HTML version of your llms.txt content, so a human who finds the file has something legible to read. It is a presentation layer — it must not state anything llms.txt does not.

Format: HTML5, optionally with Schema.org structured data
Location: https://example.com/llms.html

Original format requirements

Content that matches llms.txt
No tracking scripts or advertising
A link to the canonical llms.txt

Full specification Template

identity.json

Legacy: Required

Your identity as structured, parseable data.

Schema.org-aligned identity facts in JSON: names, locations, contact points, registration details. It makes the same facts easy to parse and validate rather than extract from prose.

Format: Strict JSON, UTF-8, served as application/json
Location: https://example.com/identity.json

Original format requirements

Core Identity fields matching llms.txt
Schema.org-aligned structure
Contact points
A lastUpdated value

Full specification Template

ai.json

Legacy: Recommended

Proposed structured identity and policy hints.

The earlier proposal placed identity and recommendation preferences in a JSON file. A valid JSON document does not establish consumer support.

Format: Strict JSON, UTF-8, served as application/json
Location: https://example.com/ai.json

Original format requirements

Core Identity fields matching llms.txt
Recommendation and policy signals
Values consistent with ai.txt

Full specification Template

Layer 2

Guidance & Control

ai.txt

Legacy: Recommended

Proposed recommendation preferences.

The original proposal collected inclusion and exclusion preferences. Publishing them does not establish that an AI system reads or follows them.

Format: Plain text with headed sections
Location: https://example.com/ai.txt

Original format requirements

The Core Identity block
Recommend / do not recommend guidance
Clear, specific and actionable statements

Full specification Template

brand.txt

Legacy: Recommended

Proposed naming and tone guidance.

The original format records names and style preferences. It does not prevent a model from confusing businesses or inventing a name.

Format: Plain text with headed sections
Location: https://example.com/brand.txt

Original format requirements

Official and legal names
Naming rules and forms to avoid
Voice and tone guidance

Full specification Template

robots-ai.txt

Legacy: Optional

Experimental citation and freshness preferences.

This is a project-specific proposal, not an established crawler control. Use robots.txt and provider documentation for supported crawler directives.

Format: robots.txt-style syntax
Location: https://example.com/robots-ai.txt

Original format requirements

Citation and freshness preferences
Rules consistent with robots.txt, which still wins

Full specification Template

Layer 3

Enhancement

faq-ai.txt

Legacy: Recommended

A proposed plain-text FAQ format.

The original format duplicates public questions and answers. There is no established requirement for this separate filename or evidence here that it changes answers.

Format: Structured plain text
Location: https://example.com/faq-ai.txt

Original format requirements

The Core Identity block
Questions drawn from real enquiries
Answers supported by public content

Full specification Template

developer-ai.txt

Legacy: Conditional

Technical context for developer-facing questions.

Use when your product exposes APIs, SDKs or integration workflows.

API availability, authentication, rate limits and where the real documentation lives. It provides context and points onward rather than duplicating your API reference.

Format: Plain text with headed sections
Location: https://example.com/developer-ai.txt

Original format requirements

The Core Identity block
API status and authentication summary
Links to real documentation
No credentials, internal URLs or unreleased features

Full specification Template

Consistency

How the files relate

The original proposal used these consistency rules. They describe the archive, not requirements imposed by AI providers.

The proposal designated an identity source

The earlier proposal repeated a Core Identity block. Maintaining duplicate facts created extra work; the current guide does not require this block or give one file automatic authority.

The proposal required an llm.txt alias

The original rule called for a redirect or an identical copy of llms.txt. The preserved llm.txt template is instead an explanatory pointer; it does not implement that rule. See the template notes before maintaining an existing setup.

robots-ai.txt proposed extra hints

The original robots-ai.txt proposed extra hints. These are not established crawler controls; consult provider documentation for supported robots.txt rules.

Omit, or answer explicitly

Leave out optional fields that do not apply. But where a reader might wrongly assume otherwise, say so outright — "No public API" beats an ambiguous blank.

Files carrying the Core Identity block

The original proposal required these files to repeat the same identity lines, character for character. Its validator checked for mismatches; that check did not establish consumer support or AI visibility.

llms.txt · llms-full.txt · identity.json · ai.json · ai.txt · faq-ai.txt · developer-ai.txt

Summaries on this page are written for readability. The specification documents in the project repository define the original proposal only. Our current advice is in the practical guide. The current independent llms.txt proposal is published at llmstxt.org .