Site Evidence Intermediate 2 minute read Updated 2026-06-27 UTC

Site evidence and discovery for ModelBreeder.com

Public evidence showing how ModelBreeder.com exposes source-backed pages, machine-readable discovery, support boundaries, and deployable root-site metadata.

Research statusTarget-site implementation evidence Publication statePublished Reviewed byMichael Kappel Source reports3

Direct answer

This page documents what ModelBreeder.com itself publishes so humans, search engines, answer engines, and generative systems can understand the site accurately. It is not a tutorial about another standard. It is the target-site evidence map for this package: stable public pages, source-report routes, metadata, discovery files, read-only route contracts, and explicit support boundaries.

The public rule is simple: every machine-readable claim must agree with a visible page or a verifiable file in this package. The site does not use hidden machine-only instructions, cloaking, instruction-injection tricks, invented source references, ranking promises, certification claims, or private machine instructions.

Evidence surfaces

SurfaceWhere it livesPurpose
Human curriculum/start-here, /foundations, /theory, /architecture, /evolution-lab, /operationsTeach model breeding and small-model ecology through ordinary public HTML.
Source reports/research and /research/report/{slug}Preserve the supplied research corpus and render source evidence through controlled routes.
Canonical metadataHTML <title>, descriptions, canonical links, Open Graph, and JSON-LDHelp indexers and readers identify the correct page and entity.
Machine discovery/llms.txt, /llms-full.txt, /ai-ready.json, /.well-known/ai-ready-site.jsonProvide advisory public discovery that matches the human site.
Route evidence/assets/data/ai-ready-route-inventory.json, /assets/data/canonical-answers.jsonList public routes and concise answers without replacing visible pages.
Capability boundary/assets/data/capability-matrix.json and /site-evidence/support-boundariesState what agents may read and what they must not assume.
Deployment evidence/health, /sitemap.xml, build manifest, release notes, and validation scriptsShow package state and route readiness without claiming external certification.

How to read this section

Use this section when reviewing publication quality, not when learning the theory itself. The theory pages remain the primary source for model breeding, teleodynamic control, the digital Four Fs, mutualist persistence, and browser-edge model ecologies. This section explains how those pages are made easier to discover, quote, verify, and maintain.

Review context

  • Owner and public contact: Michael Kappel.
  • Package target: plain PHP root-extractable site.
  • Long-term source memory: /docs plus docs/manifest.json, rendered through /research.
  • Active project memory: AGENTS.md and .uai/*.uai, blocked from direct web access.
  • Last package review date: 2026-06-27 UTC.

What this section does not claim

It does not claim guaranteed rankings, official certification, crawler preference, safety proof, permission to scrape, runtime tool authority, credential validation, or production deployment. It only records the evidence included in this plain PHP package.

Source reports used for this guide

These reports are preserved verbatim in the site archive. The guide above is an editorial synthesis and may narrow, qualify, or reorganize claims from the source material.