← Back to Research
Origin
An Origin is the canonical source of an organisation's identity, knowledge, capability and trust. It is the single thing every other representation derives from.
A durable, addressable, versioned source of truth that publishes a negotiable identity rather than a website. One canonical object, from which humans, software and agents each obtain the projection they can consume.
Why it matters
- The web made the browser the privileged consumer. The agentic internet has many consumers, each with different capabilities.
- If the human page and the machine data drift apart, trust erodes. A single origin keeps every projection consistent by construction.
- Provenance, versioning and canonical addressing are properties of the origin, so every projection inherits them for free.
In practice
- heliacon.com serves HTML to a browser, JSON-LD to a crawler and an MCP manifest to an agent, all rendered from the same origin.yaml and corpus.
- An origin declares its definitions, capabilities and provenance once. Its site, its docs, its llms.txt and its API are all projections of it.
Not this
- A marketing homepage with a separate, outdated API and a bolted on llms.txt.
- Treating the browser render as canonical and generating the machine formats as an afterthought.
Provenance
| Version | 0.1 |
|---|---|
| Status | Draft |
| Author | Pete Dainty |
| Updated | 5 July 2026 |
| Source | Canonical origin. See its provenance record. |