Site IntegrityDemonstrated real incidentv1.22.1

In plain English

This page is part of the technical reference. It keeps the expert detail but starts with a plain-language summary for first-time readers.

  • Why this matters: AI risk can come from the whole arrangement, not one obvious model.
  • What to look for: data, memory, routes, adapters, tools, evaluators, updates, and rollback paths.
  • Technical version below: the expert terminology remains available and is linked through the glossary.

ModelBreeder Architecture Memory Expansion v1.21.9

Evidence levelDemonstrated real incidentTechnical label: Demonstrated

This release preserved the six newly uploaded architecture reports, summarized them, distributed their pointers into short-term and long-term .uai memory, and expanded public content with source-bounded pages.

Added public routes

Memory distribution

The raw reports remain in /docs/source-reports/raw-markdown. Public-safe summaries live in /docs/source-report-summaries. Pointer records were added under .uai/memory/source-reports, .uai/report-pointer-index.uai, .uai/source-report-corpus.uai, .uai/short-term-memory.uai, and .uai/long-term-memory.uai.

Boundary

The new material is treated as report-derived architectural A conclusion or output produced from data. Open glossary definition. External project links are useful source leads, not live verification or endorsement. The release does not add build tools, npm, third-party JavaScript, CDN dependencies, or a hosted AI runtime.