Skip to content
Get Started Free

AI Memory Package Wizard

[uaix_ai_memory_wizard_entry_paths]

What this page is for

The UAI AI Memory Package Wizard turns the supported starter configurations from AI Memory into a six-step package planning flow. The first step now starts with a setup path: Base, Agent File Handoff, LLM Wiki, or both. The plain wizard URL defaults to Base; fragment URLs preselect the optional File Handoff and LLM Wiki paths for humans and visiting AI agents.

Current support boundary

  • Current: choose a setup path and supported preset, add package metadata, move through page-level validation, restore a local browser draft, generate a populated operating profile, generate a receiver startup brief, generate a one-file startup packet, optionally add .uai/AGENT_FILE_HANDOFF_PLAN.md, optionally answer whether the project uses an LLM Wiki inside the receiver step, choose single-site .uai handoff or multisite workspace.uai coordination, set workspace target policy and site registry notes, set advanced audience, sensitivity, deck-scope, review gates, collaboration model, UAI-only or UAI plus Wiki architecture, unit and integration test posture, test timing, deployment posture, code review strategy, release gate, source-authority policy, memory-update policy, optional AI Dreaming Memory consolidation policy, evidence ledger path, conflict-resolution policy, change-risk level, and rollback strategy, inspect a readiness checklist on the dedicated review/export step, copy individual files, download a package model JSON, download a manifest overlay JSON, download .uai/UAI_MEMORY_SYSTEM_PROFILE.md, download .uai/UAI_MEMORY_RECEIVER_BRIEF.md, download .uai/UAI_MEMORY_STARTUP_PACKET.md, download optional .uai/AGENT_FILE_HANDOFF_PLAN.md and .uai/LLM_WIKI_MEMORY_PLAN.md files, use .uai/DEPLOYMENT_MEMORY_AND_TEST_REPORT.md for release-bound memory/test reporting, and fetch the canonical starter ZIP for the selected preset.
  • Current ZIP boundary: ZIP files are generated from canonical starter templates. Wizard metadata, system profiles, receiver briefs, startup packets, overlays, the deployment memory/test report template, and generated plan files travel beside them under the selected target’s .uai/ folder so the page does not pretend to be a general hosted repository writer.
  • System profile boundary: .uai/UAI_MEMORY_SYSTEM_PROFILE.md is a generated local operating file. It defines collaboration, workspace coordination, memory architecture, source authority, memory update rules, reviewed AI Dreaming Memory consolidation, testing, deployment, code review, evidence ledger, conflict handling, risk escalation, rollback, placeholder-population rules, harness boundary policy, production deployment memory sorting for release-bound builds only, durable memory/test reporting, and before/after hot-memory size reporting; it still requires the receiver to inspect the actual codebase, CI, deployment docs, owners, workspace coordinator, runtime harness, and environment before acting.
  • Receiver boundary: .uai/UAI_MEMORY_RECEIVER_BRIEF.md is a local startup instruction file for the next human, team, or AI. It names read order, workspace target resolution, first-response expectations, support boundaries, selected setup mode, targeted-check expectations, the release-only memory-sorting rule, and the after-deployment hot-memory size report; it is not an importer, repository writer, automatic sync system, certification, endorsement, or UAI-1 conformance result.
  • Startup packet boundary: .uai/UAI_MEMORY_STARTUP_PACKET.md is a local operator packet that embeds generated instructions, selected setup mode, selected-file indexes, workspace routing guidance, optional File Handoff plan, optional LLM Wiki plan, optional AI Dreaming Memory review policy, agentic harness boundary guidance, and production deployment memory sorting only when the build is release-bound. It is still not a hosted importer, automatic repository writer, automatic wiki sync, SDK, CLI, certification, endorsement, or UAI-1 conformance result.
  • Workspace boundary: single-site projects can use the normal AGENTS.md, .uai/readme.human, and local .uai handoff path. Multisite or multisolution workspaces should keep workspace.uai as the only UAI memory file outside local .uai/ folders, and every local AGENTS.md should point to both that coordinator and its own local .uai/ folder. Verify the coordinator path and selected site, solution, or project root before local memory loads, resolve the explicit human target, and load only the selected target’s hot memory unless cross-target work is explicit. A stale absolute workspace path should be repaired or reported; it must not make the agent fall back to the current shell directory.
  • Root deployment instruction boundary: deployment-sensitive packages should point to a root instruction surface that names shared package versioning, publish folders, upload or copy targets, checksum evidence, rollback expectations, and mixed-stack differences. If WordPress sites use one consistent theme/plugin layout while .NET or other sites deploy differently, that difference belongs in the root instruction before site-local package guidance loads.
  • Production deployment memory boundary: production deployment builds and release packages sort current deployed truth, versions, changed files, checks, blockers, owners, and next actions into hot memory while routing bulky history, old reports, raw sources, stale plans, and background or rejected rationale into cold memory when a named AIWikis, LLM Wiki, or archive path exists. The after-deployment summary must be durable, not chat-only, and report active short-term memory surface size before and after reorganization, including bytes, lines, estimated tokens, removed redundancy/history, cold evidence path, checks run, checks skipped, and blockers. Ordinary dev builds, local test builds, local package experiments, and smoke checks do not run that pass unless the human explicitly marks them release-bound.
  • AI Dreaming Memory boundary: AI Dreaming Memory means an optional reviewed consolidation or replay pass over visible hot memory, reviewed intake outcomes, evidence ledgers, and named cold-memory archives. It can propose duplicate merges, stale-fact pruning, contradiction notes, next actions, and hot/cold memory deltas, but each output remains a draft until reviewed and promoted into Project Handoff or another named truth surface. It does not create hidden model memory, model-weight updates, unsupervised private-data processing, automatic repository writes, automatic LLM Wiki sync, hosted dream services, SDKs, CLIs, certification, endorsement, or UAI-1 conformance evidence.
  • Agent File Handoff boundary: Agent File Handoff setup generates local dropped-file intake guidance only. It asks the receiving AI to check active agent-file-handoff/Content/ and agent-file-handoff/Improvement/ buckets, review every pending file, perform accepted work for safe relevant files, record dispositions and completed work during chat pickup, and move processed source files to Archive/ unless the human explicitly keeps a file active. Folder creation and memory routing alone are not completion; it does not create CI pickup, watchers, hosted upload validation, automatic repository writes, or publication authority.
  • LLM Wiki boundary: LLM Wiki is not required by UAI specs or standards. When a project also uses an LLM Wiki long-memory layer, use Canonical AI Memory to keep raw source, reviewed wiki, optional graph projection, compact AI Memory, Project Handoff, and runtime execution distinct; the wizard can add a .uai/LLM_WIKI_MEMORY_PLAN.md planning file. LLM Wiki setup asks the receiver to ask for or discover the wiki root URL or repository path, index, log or evidence path, promotion owner, and whether the setup is site-local, source-site plus shared archive, or shared LLM Wiki with source routing before using wiki memory. The combined File Handoff plus LLM Wiki path tells the receiver to complete the reviewed-work loop: intake files, actual site/system work, hot handoff memory, long-term wiki/archive evidence, archive movement, and a report of what went where. Use Using UAI Packages With An LLM Wiki for the practical package/wiki memory operating path, and use LLMWikis.org-style guidance for the deeper wiki layer rather than treating UAIX.org as the wiki host. Learn more at LLMWikis.org.
  • Outside current support: hosted upload or import validation, WordPress admin installation, automatic unsupervised repository or site writes, automatic LLM Wiki sync, hosted dream services, SDKs, CLIs, certification, endorsement, and broad package-manager support. This page creates local files and starter ZIP links for people to review; it does not install, import, sync, certify, or write into another system on its own.

Wizard

[uaix_ai_memory_package_wizard]

How this connects to Project Handoff

Use Project Handoff when the chosen package needs to become a repository-local AGENTS.md, .uai/readme.human, and .uai transfer bundle. The UAIX Agents Protocol page is separate: it documents UAIX’s experimental WordPress package used for dogfood and test-site evaluation. It is not part of the wizard flow, and it does not mean this page can import ZIPs, install plugins, or write into a WordPress site.

Before widening a support claim

Check the Roadmap and Changelog before describing any future admin installer, upload/import validator, source-control or site writer, automatic LLM Wiki sync, hosted dreaming service, SDK, CLI, certification, endorsement, or broad package-manager behavior as current UAIX support.