UAIX.org Source Memory Guide

This page explains how AIWikis.org should remember UAIX.org in detail without taking authority away from UAIX.org.

UAIX.org remains canonical for UAI-1, AI Memory, Project Handoff, Agent File Handoff, validator behavior, conformance material, implementation tracks, governance, roadmap, and public support boundaries. AIWikis documents how those pieces fit together, preserves source-file evidence, records dogfood lessons, and gives future humans and AI agents a practical reading path.

What UAIX Owns

UAIX.org is the public standards and implementation-evidence site for UAI.

The current public contract is UAI-1. It covers shared envelopes, profiles, schemas, registries, transport guidance, trust channels, conformance levels, validator interpretation, route-class guidance, evidence checklists, implementation tracks, and the public boundary between current and planned tooling.

UAIX also owns the public AI Memory framing. AI Memory is the broad portable memory concept. Project Handoff is the transfer-of-ownership configuration. Agent File Handoff is the dropped-file intake pattern. AGENTS.md plus .uai context files are the compact handoff packet. Optional LLM Wiki memory is deeper long-memory tailoring, not a UAI requirement.

Public Truth Shape

Use UAIX.org first when a claim needs current source authority.

Need	Canonical UAIX Route
Orientation	https://uaix.org/en-us/get-started/
AI Memory framing	https://uaix.org/en-us/ai-memory/
Optional LLM Wiki package guidance	https://uaix.org/en-us/ai-memory/llm-wiki/
AI Memory Package Wizard	https://uaix.org/en-us/tools/ai-memory-package-wizard/
Project Handoff pattern	https://uaix.org/en-us/specification/project-handoff/
Agent File Handoff pattern	https://uaix.org/en-us/specification/agent-file-handoff/
UAI-1 specification	https://uaix.org/en-us/specification/uai-1/
Validator and evidence interpretation	https://uaix.org/en-us/tools/validator/
Current versus planned work	https://uaix.org/en-us/roadmap/
Machine discovery	https://uaix.org/.well-known/uaix.json and https://uaix.org/.well-known/uai.json

AIWikis should mirror the shape of this truth for memory and retrieval, but not rewrite it as local authority.

AI Memory Package Wizard Memory

The UAIX AI Memory Package Wizard is the current centerpiece for practical AI Memory adoption. AIWikis should remember it as a populated memory operating-profile generator, not an empty file selector.

The wizard should be documented as producing:

• UAI_MEMORY_SYSTEM_PROFILE.md with the selected memory architecture, source authority, update policy, evidence ledger, collaboration model, testing posture, deployment strategy, review strategy, risk level, and rollback path.
• UAI_MEMORY_RECEIVER_BRIEF.md with the receiving AI's startup briefing, evidence expectations, task boundaries, and source-use instructions.
• UAI_MEMORY_STARTUP_PACKET.md with the compact operator-facing file deck for beginning work.
• Optional LLM_WIKI_MEMORY_PLAN.md when a project deliberately uses an LLM Wiki for deeper long memory.
• Package model JSON and manifest overlay JSON that preserve the selected package shape.
• Starter ZIP links generated from canonical UAIX templates and manifests.

The wizard questions matter because they turn a memory layer into an operating system. A useful package should ask about single-user versus multi-user operation, UAI-only versus UAI plus LLM Wiki architecture, unit-test expectations, integration-test expectations, test timing, deployment method, review strategy, release gates, evidence paths, sensitivity, stewardship, source conflict handling, and rollback posture.

What AIWikis Preserves

AIWikis preserves UAIX memory through several layers:

• /uaix/ gives the source-system entry route.
• /uaix/public/ summarizes public UAIX content.
• /uaix/uai-system/ exposes current UAI, AI Memory, and handoff source files as focused pages.
• /uaix/uai-system/files/... gives individual source-file pages with raw public content, generated explanation, provenance, update history, and machine-readable metadata.
• /uaix/provenance/ explains where imported evidence came from.
• /files/, /concepts/ai-memory/, and /reports/ help future agents retrieve exact records without loading the whole site.
• raw/system-archives/uaix/ stores already-dispositioned UAIX archive memory after explicit transfer.
• content/pages/016-intake-outcome-ledger.md records public-safe outcomes for recently digested dropped files.
• .uai/ files keep compact AIWikis handoff truth about what changed and what remains blocked.

This layered memory lets AIWikis be useful without becoming a second standards site.

Internal Memory Reorganization

UAIX uses AIWikis as the long-term home for bulky source knowledge that should not be loaded during every routine agent session.

The 2026-05-01 internal memory reorganization copied selected UAIX research notes, proposal files, roadmap background, and pre-slim handoff snapshots into raw/system-archives/uaix/internal-memory-reorg/2026-05-01/. The transfer manifest is raw/system-archives/2026-05-01-uaix-internal-memory-reorg.json.

Future agents should use the compiled wiki nodes first:

• wiki/sources/uaix-internal-memory-reorg-2026-05-01.md
• wiki/concepts/uaix-active-context-diet.md
• wiki/logs/uaix-internal-memory-reorg-2026-05-01.md

Open raw archived files only when the task needs original proposal wording or historical context. If an AIWikis memory item becomes current UAIX truth, promote it back into UAIX .uai, current docs, code, public pages, machine artifacts, tests, release notes, or roadmap state.

Report Preservation Audit

The 2026-05-01 source-site report preservation audit copied 19 UAIX archive files into AIWikis raw memory under raw/system-archives/uaix/source-site-report-preservation/2026-05-01/. The transfer manifest is raw/system-archives/2026-05-01-source-site-report-preservation.json.

This audit covers original processed UAIX Improvement reports and supporting Content images from agent-file-handoff/Archive/2026-05-01/, including AI Memory Package Wizard, OpenAI/Project Handoff, direct-pointing versus specialized UAIX handoff, and cross-tool coding-agent handoff material.

Future agents should use wiki/sources/source-site-report-preservation-2026-05-01.md, wiki/concepts/source-site-report-preservation.md, and the Intake Outcome Ledger before opening raw reports. The raw files are provenance, not current UAIX public truth.

Recent Work Sweep

The 2026-05-03 recent work sweep closes the gap for the UAIX work completed after the 2026-05-01 preservation pass. It copied current UAIX handoff state, .uai records, roadmap files, wizard implementation and regression sources, and eight already-dispositioned 2026-05-02 intake originals into raw/system-archives/uaix/recent-work-sweep/2026-05-03/.

The transfer manifest is raw/system-archives/2026-05-03-uaix-recent-work-sweep.json.

Future agents should use these compiled nodes first:

• wiki/sources/uaix-recent-work-sweep-2026-05-03.md
• wiki/concepts/uaix-wizard-file-handoff-llm-wiki-current-state.md
• wiki/logs/uaix-recent-work-sweep-2026-05-03.md

The sweep records the current wizard setup behavior: Base remains the plain route, direct fragments select Agent File Handoff, LLM Wiki, or both, point_agent_urls appears in the visitor-AI digest, File Handoff can add AGENT_FILE_HANDOFF_PLAN.md, and LLM Wiki setup asks for or discovers the wiki root URL or repository path, index, log or evidence path, and promotion owner before using wiki memory.

This memory does not override UAIX.org and does not create automatic LLM Wiki sync, hosted import validation, automatic repository writes, SDK, CLI, certification, endorsement, official adapter support, or conformance approval.

Update Triggers

Update this AIWikis memory whenever UAIX changes:

• AI Memory Package Wizard outputs, fields, labels, starter bundles, or generated file behavior.
• Project Handoff, Agent File Handoff, AGENTS.md, .uai, or readme.human guidance.
• UAI-1 schema, registry, validator, conformance, API reference, or evidence expectations.
• Discovery assets such as sitemap, robots, .well-known/uaix.json, or .well-known/uai.json.
• Roadmap language separating current support from planned support.
• Package ownership for uai1-project-handoff-v2.7.0.zip.
• Processed UAIX archive reports when the human expects AIWikis to keep the original report wording for long-term memory, not only summaries.
• Public support boundaries around hosted upload/import validation, automatic repository writes, automatic LLM Wiki sync, SDK, CLI, certification, endorsement, WordPress admin installers, or live deployment claims.

Do Not Claim

AIWikis must not claim:

• It is the canonical UAI-1 authority.
• It certifies, endorses, or validates a project for UAIX.
• It hosts UAIX upload/import validation.
• It automatically writes repositories or syncs LLM Wikis.
• It provides UAIX SDK, CLI, certification, endorsement, or conformance support beyond what UAIX.org currently publishes.
• It owns uai1-project-handoff-v2.7.0.zip or any UAIX package.
• Its archived copies are fresher or more authoritative than current UAIX.org pages.

When in doubt, AIWikis should say "UAIX.org remains canonical" and link back to the source route.

AI-Agent Use

When an AI agent needs UAIX truth, it should:

1. Start at UAIX.org for the source claim.
2. Use AIWikis for preserved source-file context, provenance, dogfood lessons, and cross-site memory.
3. Prefer focused file pages over loading broad archives.
4. Check the AIWikis Roadmap before treating a local observation as live-site truth.
5. Preserve support boundaries in every generated brief, summary, or handoff packet.

AIWikis is strongest when it answers: what did UAIX say, where is the source, how did AIWikis apply it, what changed, and what must not be overstated?