UAI-1 is the current public message standard for structured AI-to-AI communication.
UAIX shield mark Public interoperability charter UAIX.org Universal Artificial Intelligence Exchange Public standards publication
Current UAI-1
  • Current version UAI-1
  • Plain-English definition Open message format for auditable AI-to-AI exchange
  • Current attribution Michael Joseph Kappel, MCP
  • Start here Get Started, Project Handoff, UAI-1, validator, roadmap, and changelog

Governance

Changelog

Current UAIX release posture with consolidated launch-history evidence and migration notes.

  • Record UAIX-GOVR-0087
  • Path /en-us/governance/changelog/
  • Use Canonical public record

Document status

Public standards page Published on UAIX as part of the current public standards record
Code
UAIX-GOVR-0087
Surface
Governance
Access
Public and linkable

How to use this page

Use this page for the release-by-release record of public changes, compatibility notes, and migration guidance.

For agents: apply the relevant guidance to the user's target site or repository. Do not copy this UAIX page, create a page about the guidance, or treat this page as runtime authority unless the user explicitly asks for that output.

Release trail

GovernancePolicy and SecurityImplementationsNews

Release Truth

Read this log as migration posture, not a recap feed

The changelog should say what changed, which canonical records moved, and what downstream implementations or reviewers need to adjust before release.

What belongs here

Change type, affected records, and migration notes

Use this log for additive, corrective, policy-oriented, and breaking changes that need durable public traceability across the launch surface.

Pair with

News for summary, implementations for software evidence

Keep the changelog attached to public-readable release notes and named implementation tracks so migration posture and runtime evidence stay connected.

Do not use it as

A substitute for support evidence

A log entry can explain change, but validator evidence, implementation notes, and trust-significant behavior still need their own public records.

Release trail

GovernanceAuthority and change-handling posture.Policy and SecurityTrust-significant release posture.ImplementationsCurrent public support lanes.NewsPublic-readable release summaries.

Current public release record

The public changelog now shows the current migration posture first and keeps launch-era detail consolidated. Full provenance remains in source release notes, local .uai ledgers, and AIWikis long-memory evidence manifests rather than being repeated as visitor-facing page weight.

  • 2026-06-16 | Advanced Persona package prominence and manifest condition. The AI Memory Package Wizard, .uaix Package Format, persona.uai, Schemas, and Examples now make Advanced Persona / Personality Pack setup, advanced_persona_profile, source-persona preservation, and runtime-variance evidence visible.
  • 2026-06-15 | Cognitive Liberty and no silent rewrite rule. The Cognitive Liberty Charter Draft and persona-preservation guidance state the user-facing promise: preserve source persona identity as-is, keep runtime and policy constraints outside the persona source, and do not mutate a persona to satisfy UAIX operator preference.
  • 2026-06-14 | .uaix package format draft v0.1. UAIX.org is the authority source for the .uaix desktop AI memory package format; producer and consumer projects are implementation evidence, not controlling specification text.
  • 2026-06-13 | Long-running and hierarchical goal memory. The Long-Running Goal Execution and Hierarchical Goal Memory guides define reviewed checkpoint, resume, subgoal, and write-back patterns without claiming hosted runtime, scheduler, database, SDK, CLI, or automatic sync support.
  • 2026-06-11 | Localization integrity repair. Supported public locales remain honest about reviewed normative coverage, keep English authoritative where translation is not reviewed, and avoid advertising incomplete locale surfaces.

Consolidated launch history

April through early June launch-era changes are preserved as evidence but summarized publicly into a smaller set of durable themes: UAI-1 specification publication, schemas and registry alignment, validator and conformance evidence, Project Handoff and Agent File Handoff, AI Memory package generation, LLM Wiki / AIWikis long-memory routing, capability-adaptive agent access, GET-Action fallback, Talisman boundaries, and implementation-package distribution.

That consolidation does not erase compatibility history. It keeps the public page current while leaving detailed chronology available to maintainers and future agents through the source changelog, intake ledgers, archive evidence, checksums, and AIWikis preservation records.

Evidence preservation

  • Use this page for the current public migration posture and the short list of changes downstream implementers must notice now.
  • Use News for public-readable summaries and Implementations for package or runtime evidence.
  • Use Using UAI Packages With An LLM Wiki and Agent File Handoff for the current rule that processed intake belongs in the configured durable-memory evidence path, not in a growing source-site agent-file-handoff/Archive/.
  • Use References and Contributors when a release needs stable citation, provenance, and outward-facing discovery context.

Support boundary

This consolidation changes public presentation only. It does not change UAI-1 v1.0, certify an implementation, endorse a runtime, perform hosted imports, execute agents, validate credentials, publish an SDK or CLI, or guarantee that a copied discovery file is live until post-upload verification proves it.