Open standard · Technical specification
EntityMap is a proposed open standard for publishing a structured, entity-first index of website knowledge for AI systems, retrieval pipelines, and language-model-based applications.
Where sitemap.xml tells crawlers what pages exist, entitymap.json tells AI systems what a site knows — which entities it covers, how they relate, and where the evidence is.
The problem
AI retrieval systems today operate at the page level — fetching HTML and extracting passages without structured awareness of entities, publisher identity, or concept relationships. This produces three recurring failures for publishers.
Disambiguation
The same concept under different surface forms is treated as separate signals rather than one entity.
Attribution
Publisher identity is absent from retrieved content and does not survive aggregation into AI answers.
Relationships
Connections between concepts are buried in prose rather than declared as explicit, typed relations.
EntityMap addresses these by giving publishers a standard way to declare their entities, evidence, and relationships in a machine-readable file at a predictable URL. Read the full rationale →
The files
An EntityMap consists of two files published at predictable root-level URLs:
https://yourdomain.com/entitymap.json ← machine-readable primary file https://yourdomain.com/entitymap.html ← crawler and human readable view
A minimal entity entry:
{
"entityId": "e_001",
"@type": "DefinedTerm",
"name": "Companion Planting",
"description": "The practice of growing different plants in proximity
for mutual benefit — pest control, pollination, improved yield.",
"sameAs": "https://www.wikidata.org/wiki/Q905413",
"relations": [
{ "predicate": "IMPROVES", "targetId": "e_002", "targetName": "Crop Yield" }
],
"hasChunks": [
{
"chunkId": "c_001",
"text": "Companion planting pairs plants that benefit each other —
growing basil near tomatoes repels aphids and improves flavour.",
"sourceUrl": "https://acmegardens.com/companion-planting-guide",
"pageTitle": "The Complete Companion Planting Guide",
"publisher": "Acme Gardens",
"retrieved": "2026-03-27T09:14:00Z",
"relevanceScore": 0.95
}
]
}
Full specification → · Minimal valid example → · Implementation guide →
For whom
Publishers
Declare entity ownership, surface-form equivalences, and publisher attribution in a form AI systems can consume directly.
RAG developers
Use a structured, publisher-attributed retrieval source instead of raw HTML chunking.
AI search teams
Consume a predictable, versioned knowledge index from any conforming site.
Tool builders
Build generators, validators, and plugins against an open, stable spec. Register implementations in the community registry.
Implementations
EntityMap follows a familiar web pattern: publish an open vocabulary, demonstrate implementation value, and improve through adoption. GoodRelations showed that independently published vocabularies have shaped the wider structured web. Contributions, predicate proposals, and third-party implementations are welcomed via GitHub.