Skip to content
AIWikis.org

UAIX Handoff Pattern

Publication Warning This page is marked noindex and should not be treated as canonical public authority.

This file explains the public AI Memory and Project Handoff pattern that UAIX is now using in this repository.

Metadata

FieldValue
Source siteuaix.org
Source URLhttps://uaix.org/
Canonical AIWikis URLhttps://aiwikis.org/uaix/files/raw-system-archives-uaix-internal-memory-reorg-2026-05-01-uai-handoff-pa-7a4319ba/
Source referenceraw/system-archives/uaix/internal-memory-reorg/2026-05-01/.uai/handoff-pattern.uai
File typeuai
Content categoryuai-system
Last fetched2026-06-22T01:56:21.9510185Z
Last changed2026-05-01T17:53:23.0110585Z
Content hashsha256:7a4319ba015e5b3f2929e01919260f59e4a4497d088dca10cd7fb6021171ea2e
Import statusunchanged
Raw source layerdata/sources/uaix/raw-system-archives-uaix-internal-memory-reorg-2026-05-01-uai-handoff-pattern-uai-7a4319ba015e.uai
Normalized source layerdata/normalized/uaix/raw-system-archives-uaix-internal-memory-reorg-2026-05-01-uai-handoff-pattern-uai-7a4319ba015e.txt

Current File Content

Structure Preview

  • UAIX Handoff Pattern
  • Purpose
  • Public Source Pages
  • Public UAI Reading Path
  • Minimum Useful Bundle
  • Active File Intake
  • How To Build The Bundle From Existing Instructions
  • Website-To-Bundle Checklist
  • Loader Rules
  • Required First Response Pattern
  • Human Setup Checklist
  • Good Handoff Criteria
  • Maintenance Rules

Raw Version

This public page shows a bounded preview of a large source file. The complete source remains in the raw and normalized source layers named in metadata, with the SHA-256 hash above for verification.

  • Source characters: 19696
  • Preview characters: 11934

Local absolute paths are redacted in this public view. The source hash and source-side raw layer are based on the unredacted source file.

---
uaix: "1.0"
type: custom
title: "UAIX Handoff Pattern"
created: "2026-04-25"
updated: "2026-05-01"
status: active
author: "UAIX maintainers"
version: 22
---

# UAIX Handoff Pattern

## Purpose

This file explains the public AI Memory and Project Handoff pattern that UAIX is now using in this repository.

For this workspace, the handoff pattern also records the current source-of-truth split: UAI public materials now live in `[local path redacted]`; UAIX.org is the canonical UAI-1 public site; Protocol5.com is only a .NET plugin and NuGet package distribution surface that links back to UAIX.org. Root `readme.human` is now the human-facing companion that explains what the AI sees, protects, and needs clarified before people steer the system.

A new AI should be able to read root `AGENTS.md` plus the loaded `.uai` files and understand both:

- how to work safely in this UAIX workspace
- how the public `AGENTS.md` plus `.uai` handoff pattern is meant to be copied into another project

## Public Source Pages

Use these public pages when a reader needs the published version of the pattern:

- `/en-us/specification/project-handoff/` for the practical how-to page
- `/en-us/guides/project-handoff-openai/` for OpenAI/Codex workflow guidance that treats Project Handoff as portable project memory rather than runtime orchestration
- `/en-us/guides/project-handoff-coding-agents/` for cross-coding-agent guidance that keeps Codex, Claude Code, Cursor, GitHub Copilot, Gemini Code Assist, local agents, vendors, and humans pointed at one local handoff bundle without claiming official adapters
- `/en-us/ai-memory/` for broad AI Memory framing, supported starter configurations, dynamic canonical-template downloads, and the optional LLM Wiki boundary
- `/en-us/ai-memory/llm-wiki/` for optional package-tailoring guidance when a project already uses or deliberately chooses an LLM Wiki
- `/en-us/tools/ai-memory-package-wizard/` for six-step guided package planning over supported starter bundles, page-level validation, local draft restore, review/export, package model JSON, manifest overlay JSON, generated system profile with source-authority/evidence/conflict/risk/rollback protocols, receiver brief, and startup packet files, copy-paste files, optional generated LLM Wiki memory-plan files, readiness metadata, and canonical ZIP links
- `/en-us/specification/agent-file-handoff/` for portable AGENTS.md-triggered dropped-file intake
- `/en-us/specification/agents-md/` for the deeper AGENTS.md `.uai` linking proposal report
- `/en-us/reports/` for the dated report index
- `/en-us/reports/relationship-audit-aiwikis-llmwikis-uaix/` for the AIWikis/LLMWikis/UAIX role and authority audit
- `/en-us/reports/llm-wiki-vs-uaix-project-handoff/` for the LLM Wiki versus Project Handoff comparison
- `/en-us/reports/llm-wiki-and-uaix-project-handoff/` for the companion implementation model
- `/en-us/reports/refining-uai-protocol-agent-communication/` for background on repository-native context, validation, interoperability, and loader guardrails

## Public UAI Reading Path

A zero-context AI should read local handoff files before it reads public spec pages. The local files explain what project it is in; the public pages explain what UAIX.org currently publishes.

Use this order when the task asks how to understand or cite UAI from public materials:

1. `/en-us/get-started/` for the public orientation path.
2. `/en-us/ai-memory/` for the broad AI Memory framing, supported starter bundle chooser, and the boundary between portable memory packets and optional deeper LLM Wiki memory.
3. `/en-us/ai-memory/llm-wiki/` only when the project already uses, or intentionally chooses, an LLM Wiki and needs package-shaping guidance for that use case.
4. `/en-us/tools/ai-memory-package-wizard/` when one supported starter configuration should move through preset, basics, operations, protocols, receiver, and review/export steps to become a package model, populated system profile with source-authority/evidence/conflict/risk/rollback protocols, receiver startup brief, startup packet, manifest overlay, copy-paste file deck, optional LLM Wiki memory-plan file, readiness checklist, and canonical ZIP download.
5. `/en-us/specification/project-handoff/` for the practical transfer subtype and handoff starter bundle.
6. `/en-us/guides/project-handoff-openai/` when the workflow uses OpenAI Agents, Codex, Skills, MCP, sessions, traces, approvals, or another runtime and the reader needs the "OpenAI runs the agents; Project Handoff preserves the project memory" boundary.
7. `/en-us/guides/project-handoff-coding-agents/` when one repo-local bundle needs to work across Codex, Claude Code, Cursor, GitHub Copilot, Gemini Code Assist, local agents, vendors, and humans without duplicating truth or claiming official adapters.
8. `/en-us/specification/agent-file-handoff/` for portable AGENTS.md-triggered dropped-file intake.
9. `/en-us/specification/agents-md/` for AGENTS.md `.uai` link syntax, structured record ideas, and compatibility notes.
10. `/en-us/reports/`, the relationship audit, the LLM Wiki comparison reports, and `/en-us/reports/refining-uai-protocol-agent-communication/` for the permanent report trail and deeper role-boundary, knowledge-boundary, security, or validation background.
11. `/en-us/specification/uai-1/`, `/en-us/schemas/`, `/en-us/registry/`, `/en-us/examples/`, `/en-us/tools/validator/`, `/en-us/tools/adoption-kit/`, and `/en-us/tools/conformance-pack/` when the handoff needs to become UAI-1 message evidence.
12. `/en-us/roadmap/` and `/en-us/governance/changelog/` before making claims about what is current, planned, or changed.

Do not make a public support claim from Project Handoff alone. It is repository-context guidance and a portable project-memory layer, not an agent runtime, OpenAI replacement, scheduler, hosted importer, official vendor adapter, or tool-calling framework. UAI-1 support claims need validator evidence, implementation records, release notes, or other public evidence surfaces.

The current supported AI Memory starter ZIPs and AI Memory Package Wizard are live outputs generated from the canonical bundle registry, template definitions, generated manifests, and wizard choices. They support starter downloads, a six-step local planning flow, page-level validation, local draft restore, package model JSON, manifest overlay JSON, generated system profile with source-authority/evidence/conflict/risk/rollback protocols, receiver brief, and startup packet files, copy-paste files, readiness metadata, and optional LLM Wiki memory-plan files; they are not a hosted upload/import validator, automatic repository writer, automatic wiki sync service, SDK, CLI, certification, or endorsement service. LLM Wiki is not required by UAI specs or standards.

## Minimum Useful Bundle

Every project using this pattern should start with:

- `AGENTS.md` at the repository root
- `readme.human` at the repository root for the human briefing from the AI perspective
- `.uai/context.uai`
- `.uai/stack.uai`
- `.uai/constraints.uai`

Add these as the work becomes more complex:

- `.uai/architecture.uai` for system shape and ownership boundaries
- `.uai/decisions.uai` for ADR-style decisions
- `.uai/progress.uai` for completed work, current focus, blockers, and done criteria
- `.uai/operations.uai` for local commands, release checks, and deployment procedure
- `.uai/test-plan.uai` for default targeted verification, release gates, package-build triggers, and full-suite triggers
- `.uai/style.uai` for code, copy, naming, and design conventions
- `.uai/errors.uai` when known pitfalls or recurring failures matter
- `.uai/file-handoff.uai`, `.uai/intake-index.uai`, and `agent-file-handoff/Archive/` when humans or other AI systems need to drop files into a project, process them during AGENTS.md load, and retire already-dispositioned source files from routine AI attention
- thin tool-specific pointer files such as `CLAUDE.md`, `GEMINI.md`, `.cursor/rules/uaix-handoff.mdc`, or `.github/copilot-instructions.md` only when a coding-agent surface benefits from them; each pointer must route back to `AGENTS.md` rather than duplicate truth

## Active File Intake

The current dogfood pattern adds source-only `agent-file-handoff/Content/`, `agent-file-handoff/Improvement/`, and `agent-file-handoff/Archive/` workspaces. `Content/` is for candidate public/editorial material. `Improvement/` is for audits, QA findings, SEO reports, bug notes, roadmap suggestions, and site-fix work. `Archive/` is for already-dispositioned files that humans may want to retain. The requirement is chat-start intake triggered by `AGENTS.md` loading: every new AI work session checks active intake buckets, updates the index, and immediately reviews listed files before broad planning, even if the human prompt does not mention files.

- `agent-file-handoff/Content/` is where humans or other AI systems drop candidate public/editorial files.
- `agent-file-handoff/Improvement/` is where humans or other AI systems drop work-to-fix files.
- `agent-file-handoff/Archive/` is where processed files go after disposition. Archived files are ignored by normal AI intake and should not be scanned, summarized, indexed, or reprocessed unless a human explicitly names the archived file or moves it back into `Content/` or `Improvement/`.
- When a human explicitly asks an LLM Wiki implementation such as AIWikis.org to consolidate already-dispositioned archives, the pickup must record source path, final memory path, checksum, original disposition, actor, timestamp, and transfer evidence, then update the LLM Wiki history/log/index or wiki graph and a public-safe outcome-feedback page such as AIWikis' Intake Outcome Ledger before the source archive copy is removed. That final memory copy is provenance, not UAIX authority.
- `agent-file-handoff/intake/` is legacy compatibility only; files left there are treated as `Unsorted`.
- `scripts/Invoke-UaiFileIntake.ps1` is the local one-shot helper that refreshes `.uai/intake-index.uai`.
- Do not require or recommend CI jobs, scheduled jobs, repository automations, watchers, daemons, background services, cron loops, queue folders, manifests, or out-of-chat auto-pickup. The portable contract is chat pickup: the next AI or loader scans the intake buckets during `AGENTS.md` handoff load.
- `.uai/intake-index.uai` makes current dropped files visible to the next AI with bucket names, route hints, and checksums. It is an action list: every `needs-agent-review` file must be inspected, summarized, and dispositioned before unrelated broad work continues.
- Do not recommend per-folder README files for this workspace. Durable AI instructions belong in `AGENTS.md` and `.uai` files; human briefing belongs in root `readme.human`.

This solves the recurring handoff problem where files are present in the workspace but invisible unless the human explicitly points at them. The fix is the AGENTS.md load behavior, not a requirement to keep a separate process running.

## How To Build The Bundle From Existing Instructions

When adapting another project, mine the project's existing website, README, docs, release notes, and operational instructions into the handoff bundle:

- Put the short project purpose, audience, current public truth, and success criteria in `context.uai`.
- Put languages, frameworks, runtime commands, hosted services, package managers, and release destinations in `stack.uai`.
- Put supported human languages, locale routes, translation status, and disabled-language boundaries in `stack.uai`, `public-surface.uai`, or the closest local equivalent.
- Put non-negotiable rules, support boundaries, secrets policy, destructive-operation rules, and production-change rules in `constraints.uai`.
- Put system layout, important modules, public routes, machine artifacts, and ownership boundaries in `architecture.uai`.

Why This File Exists

This is a UAI AI Memory handoff file from uaix.org. It is shown here because AIWikis.org is demonstrating the real source files that make the UAIX / LLM Wiki memory system work, not only summarizing those systems after the fact.

Role

This .uai file is a compact custom packet. It keeps one kind of durable project truth separate from the rest of the archive so an agent can load the topic it needs without pulling the whole project history into prompt context.

Structure

The file is structured around these visible headings: UAIX Handoff Pattern; Purpose; Public Source Pages; Public UAI Reading Path; Minimum Useful Bundle; Active File Intake; How To Build The Bundle From Existing Instructions; Website-To-Bundle Checklist. Those headings are retrieval anchors: a crawler or LLM can decide whether the file is relevant before reading every line.

Prompt-Size And Retrieval Benefit

Keeping this material in a separate file reduces prompt pressure because an agent can load this exact unit only when its role, source site, category, or hash is relevant. The surrounding index pages point to it, while this page preserves the full content for audit and exact recall.

How To Use It

  • Humans should read the metadata first, then inspect the raw content when they need exact wording or provenance.
  • LLMs and agents should use the source site, category, hash, headings, and related files to decide whether this file belongs in the active prompt.
  • Crawlers should treat the AIWikis page as transparent evidence and follow the source URL/source reference for authority boundaries.
  • Future maintainers should regenerate this page whenever the source hash changes, then review the explanation if the role or structure changed.

Update Requirements

When this source file changes, update the raw source layer, normalized source layer, hash history, this rendered page, generated explanation, source-file inventory, changed-files report, and any source-section index that links to it.

Related Pages

Provenance And History

  • Current observation: 2026-06-22T01:56:21.9510185Z
  • Source origin: current-source-workspace
  • Retrieval method: local-source-workspace
  • Duplicate group: sfg-587 (primary)
  • Historical hash records are stored in data/hashes/source-file-history.jsonl.

Machine-Readable Metadata

{
    "title":  "UAIX Handoff Pattern",
    "source_site":  "uaix.org",
    "source_url":  "https://uaix.org/",
    "canonical_url":  "https://aiwikis.org/uaix/files/raw-system-archives-uaix-internal-memory-reorg-2026-05-01-uai-handoff-pa-7a4319ba/",
    "source_reference":  "raw/system-archives/uaix/internal-memory-reorg/2026-05-01/.uai/handoff-pattern.uai",
    "file_type":  "uai",
    "content_category":  "uai-system",
    "content_hash":  "sha256:7a4319ba015e5b3f2929e01919260f59e4a4497d088dca10cd7fb6021171ea2e",
    "last_fetched":  "2026-06-22T01:56:21.9510185Z",
    "last_changed":  "2026-05-01T17:53:23.0110585Z",
    "import_status":  "unchanged",
    "duplicate_group_id":  "sfg-587",
    "duplicate_role":  "primary",
    "related_files":  [

                      ],
    "generated_explanation":  true,
    "explanation_last_generated":  "2026-06-22T01:56:21.9510185Z"
}

Next Useful Routes

  • Start Here A task-first reading path for AIWikis.org, separating newcomer learning, source-memory lookup, maintainer workflow, and AI-agent retrieval.
  • Topic Index A tag-oriented index for LLM Wiki, AI memory, UAI, source governance, crawling, and retrieval topics.
  • Source Map AIWikis source-governed page for durable AI memory, evidence routing, and agent-readable retrieval.
  • UAIX.org UAIX.org source-system overview for transparent AIWikis memory demonstration.
  • UAIX.org Source Memory Guide AIWikis source-governed page for durable AI memory, evidence routing, and agent-readable retrieval.
  • UAIX.org Files Site-scoped current-source file index for UAIX.org.