Skip to content
LLMSRank
Docs menu

Entity identity

Entity identity and vocabulary

Also called entity disambiguation or brand entity.

In short: Add an ## Identity section with “Legal name:”, “Also known as:” or an identifier such as your ticker, and a ## Glossary of your terms (or a link to your glossary page).

  • Category: Agent Readiness · weight 5 of 100 within the category
  • Agent Readiness is worth 30 in technical, 25 in generic and commerce, 20 in regulated (points of 100, by scoring profile)
  • Audit ID A8 (as shown on reports) · rubric v2.1

What it checks

Whether the file states who you are: your legal or brand name, other names or identifiers, and the meaning of your terms.

Why it matters

Agents confuse similarly named organizations and misread jargon. Stating who you are, and what your terms mean, keeps answers about the right entity — the groundwork of AI readiness, because every other signal is wasted on an answer about someone else.

When it applies

Always, for any llms.txt that exists.

An audit that doesn't apply, or that we couldn't verify, is removed from your score and its weight shared among the Agent Readiness audits that do apply. It is never scored as a zero.

How it's scored

0.4 when the file states your identity: a labelled legal or brand name (“Legal name:”, “Brand:”, “Company:”, “Owner:”), “operated by …” or “a subsidiary of …”, or a legal-entity name that shares a word with your H1 (“Acme Payments, Inc.”) · + 0.3 for alternate names or identifiers: an alias label (“Also known as:”, “Formerly:”, “Aliases:”), “formerly …” or “also known as …” right after your name, a stock ticker (“NYSE: ACME”), LEI, Wikidata ID, D-U-N-S or registration number, or “the official website of …” · + 0.3 for a glossary: a ## Glossary, Terminology, Vocabulary or Definitions heading, a “Vocabulary:” line, or a link to a /glossary page. 0 when the file states none of these.

How to fix

  1. State your legal name and brand name as labelled lines (“Legal name: …”, “Brand: …”).
  2. Add former names and common variants (“Also known as: …”), and identifiers where you have them: ticker, LEI, registration number, Wikidata ID.
  3. Add a ## Glossary of the terms you use in a specific way, or link your glossary page.

Example

## Identity

- Legal name: Acme Payments, Inc.
- Also known as: Acme, AcmePay
- Stock ticker: NASDAQ: ACME

## Glossary

- Payout: a transfer of your Acme balance to your bank account

Frequently asked questions

What entity identity details should llms.txt include?
Your legal name, brand names and common variants (including misspellings and former names), identifiers where you have them (a stock ticker, company registration number, LEI or Wikidata ID), and a short glossary of terms you use in a specific way. These let an agent tell you apart from similarly named organizations. The consistency checker lists each mismatch in your entity name, URL and sameAs with the exact line to fix.

Sources: LLMSRank — Entity identity audit · schema.org — Organization (legalName, alternateName, sameAs, leiCode)

How does the scanner score my identity section?
From the file itself, with fixed patterns and no model. 0.4 when the file states your identity (a labelled legal or brand name such as Legal name: Acme Payments, Inc., “operated by …”, or a legal-entity name that shares a word with your H1), + 0.3 for alternate names or identifiers (Also known as:, “formerly …” after your name, a ticker such as NYSE: ACME, an LEI, Wikidata ID or registration number), + 0.3 for a glossary (a ## Glossary or ## Terminology section, or a link to your glossary page). Link lines don’t count: a company named in a page description could be anyone. A version or date stamp no longer counts here; it belongs to Freshness signals.

Sources: LLMSRank — Entity identity audit · LLMSRank — How scoring works

Why add a Wikidata ID to llms.txt?
A Wikidata item ID (a Q followed by a number, such as Q42) names one entity unambiguously, and schema.org’s sameAs uses the same kind of reference to pin identity. Giving it lets an agent connect your file to the right entity instead of a namesake.

Sources: Wikidata — Introduction · schema.org — Organization

Does the entity identity check apply to every site?
Yes, every llms.txt is scored on it, glossary included. The rubric once planned to drop the glossary part when a site has no jargon, but there is no reliable way to detect jargon, so the glossary always counts: a file without one can reach 0.7.

Sources: LLMSRank — Entity identity audit · LLMSRank — How scoring works

In our research, how do sites score on entity identity?
Few files state who they are. In our 187-site corpus (captured 2026-10-10, scored with rubric v2.1), 50 files (26.7%) score above 0 and none earns full marks; the mean score is 0.119. 30 files state a legal or brand name, 17 give identifiers such as a stock ticker, 3 give former names and 15 have or link a glossary. Exactly one file pairs an identity statement with a glossary, which is why full marks remain unclaimed. The best score, 0.7, goes to 13 files that state a legal or brand name together with identifiers or former names.

Sources: LLMSRank — How scoring works · LLMSRank — Entity identity audit