Specification · v1.0

agent-preview/v1

A structured JSON manifest sibling for every URL. One fetch gives an agent the title, summary, type, representations (HTML, markdown, image, JSON-LD), and per-type context — without rendering HTML or parsing markdown.

Why

Pasting a URL in chat already produces a rich human preview (the OG image). For agents, the equivalent today is "fetch HTML, parse it, hope the meta tags are present and consistent." That's expensive and brittle. agent-preview/v1 is the structured sibling — a one-fetch typed summary so agents previewing a link can decide whether to dig deeper before reading the full page or the markdown mirror.

How it works

Every prerendered URL on this site exposes a sibling preview manifest in its directory — the page's trailing-slash URL plus index.preview.json (e.g. /methodology/index.preview.json; root: /index.preview.json). The HTML page advertises it via <link rel="alternate" type="application/vnd.agent-preview+json" href="…"> in <head>. Crawlers and humans ignore it. Aware agents follow it.

MIME type

application/vnd.agent-preview+json

Discovery

<link rel="alternate" type="application/vnd.agent-preview+json" href="https://example.com/article/index.preview.json">

Schema (v1.0)

The manifest is a single JSON object. Required keys are listed below; consumers MUST tolerate additional keys and unknown type values (treat unknowns as reference).

{
  "spec_version": "1.0",
  "spec_url": "https://www.mnemom.ai/spec/agent-preview/v1/",
  "url": "https://www.mnemom.ai/methodology/",
  "canonical_url": "https://www.mnemom.ai/methodology/",
  "type": "methodology",
  "title": "Mnemom Trust Rating™ Methodology",
  "summary": "Fully transparent formula, component weights, grade scale.",
  "language": "en-US",
  "representations": {
    "html": "https://www.mnemom.ai/methodology/",
    "markdown": "https://www.mnemom.ai/methodology/index.md",
    "image": "https://www.mnemom.ai/api/og-image?type=methodology",
    "json_ld": "https://www.mnemom.ai/methodology/"
  },
  "context": {},
  "last_modified": "2026-05-13T00:00:00Z",
  "publisher": {
    "name": "Mnemom",
    "url": "https://www.mnemom.ai",
    "agents_txt": "https://www.mnemom.ai/agents.txt",
    "readiness_manifest": "https://www.mnemom.ai/for-agents/"
  }
}

type taxonomy

home · marketing · blog_post · research_paper · case_study · agent_profile · team_profile · methodology · explainer · policy · discovery · coherence_report · reference

context

Type-specific fields. All optional. Examples by type:

  • blog_post · research_paper: author, published_at, reading_time_min, tags
  • case_study: company, outcome
  • agent_profile · team_profile: agent_id / team_id, grade, score, last_attested_at
  • coherence_report: status, primitives

For other sites

agent-preview/v1 is vendor-neutral by design. Mnemom hosts the canonical spec and provides the reference implementation, but the schema is yours to adopt. If you build a verifier of your own, link it from your /agents.txt and let other agents cross-verify. As adoption grows, we'll register the MIME with IANA to drop the vnd. prefix.

Versioning

Future schema changes go to v2 with a new MIME (application/vnd.agent-preview+json; version=2) and a new spec URL. v1 stays frozen at this URL forever. Consumers can pin to spec_url + spec_version for stable behavior.

Compliance

Mnemom verifies the surface nightly. See commitment preview-surface at /for-agents and the live status JSON at /agent-readiness-status.json.

Featured on There's An AI For That