LLMWikis Public Surface

LlmWikis.org is a practical public handbook for building and maintaining LLM Wikis: source-governed knowledge systems where raw evidence is compiled into durable, AI-readable markdown pages.

Metadata

Field	Value
Source site	llmwikis.org
Source URL	https://llmwikis.org/
Canonical AIWikis URL	https://aiwikis.org/llmwikis/files/uai-public-surface-uai-32044978/
Source reference	.uai/public-surface.uai
File type	uai
Content category	uai-system
Last fetched	2026-06-22T01:56:21.9510185Z
Last changed	2026-06-13T22:24:46.1373887Z
Content hash	sha256:3204497887df20f7d843c85607664daa52af0ab4b971a13fa1504c484ca5b7cd
Import status	unchanged
Raw source layer	data/sources/llmwikis/uai-public-surface-uai-3204497887df.uai
Normalized source layer	data/normalized/llmwikis/uai-public-surface-uai-3204497887df.txt

Current File Content

Review Boundary

Long-run agent and Agent-Staged Page language in the bounded preview is retained as source-file evidence only. Any promoted staged synthesis requires source traces and owner review before it can become canonical public guidance.

Structure Preview

• LlmWikis Public Surface
• Positioning
• Primary Route Families
• AIWikis Multisite Rule
• Current Support Claims
• Hot/Cold Memory Boundary
• Discovery Boundary
• Verification Boundary
• 2026-06-13 Desktop Navigation Surface
• 2026-06-13 Long-Run Agent Wiki Surface

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: 14629
• Preview characters: 11910

---
uaix: "1.0"
type: custom
title: "LlmWikis Public Surface"
created: "2026-04-26"
updated: "2026-06-13"
author: "LlmWikis maintainers"
version: 36
---

# LlmWikis Public Surface

## Positioning

LlmWikis.org is a practical public handbook for building and maintaining LLM Wikis: source-governed knowledge systems where raw evidence is compiled into durable, AI-readable markdown pages.

UAIX.org remains canonical for UAI-1, AI Memory, Project Handoff, Agent File Handoff, schemas, registries, validator behavior, roadmap, governance, and public support boundaries. LlmWikis UAIX/UAI material is non-normative case-study and implementation guidance.

## Primary Route Families

- Core handbook: `/`, `/start-here/`, `/what-is-an-llm-wiki/`, `/why-llm-wikis/`, `/how-to-build-an-llm-wiki/`, `/llm-wiki-structure/`, `/starter-template/`, `/tools/`, `/tools/llm-wiki-setup-wizard/`, `/content-types/`, `/trust-model/`, `/metadata-standard/`, `/for-ai-agents/`, `/llm-wiki-vs-rag/`, `/llm-wiki-vs-ai-memory/`, `/checklist/`, `/maturity-model/`, `/security-and-privacy/`.
- Architecture and operations: `/architecture/`, `/architecture/three-layer-architecture/`, `/architecture/rag-vs-incremental-compilation/`, `/architecture/memory-lifecycle/`, `/operations/`, `/operations/ingest/`, `/operations/two-step-ingest-pipeline/`, `/operations/query/`, `/operations/lint/`.
- Navigation and schema: `/navigation/`, `/navigation/index-and-log/`, `/navigation/graph-navigation/`, `/schema-engineering/`, `/schema-engineering/agents-md/`, `/schema-engineering/durable-llm-wikis/`, `/schema-engineering/page-schema-standard/`.
- Reference standards: `/reference/`, `/reference/specification/`, `/reference/structure/`, `/reference/metadata/`, `/reference/navigation/`, `/reference/content-types/`, `/reference/trust-model/`, `/reference/source-policy/`, `/reference/security-privacy/`, `/reference/agent-rules/`, `/reference/knowledge-graph/`.
- Examples and reports: `/examples/`, `/examples/personal-llm-wiki/`, `/examples/team-llm-wiki/`, `/reports/`, `/reports/using-llm-wiki-with-uai/`.
- Non-normative protocol/case-study routes: `/protocols-case-studies/`, `/standards/`, `/guides/`, `/guide/long-run-agent-wikis/`, `/guides/canonical-ai-memory/`, `/guides/ai-dreaming-memory-management/`, `/guides/knowledge-graphs-for-llm-wikis/`, `/guides/using-llm-wiki-with-codex/`, `/guides/llm-wiki-agentic-orchestration/`, `/guides/llm-wiki-ai-memory-project-handoff/`, `/guides/agent-handoff-patterns/`, `/guides/validate-a-uai-1-message/`, `/comparisons/`, `/glossary/`, `/implementations/`, `/uaix-bridge/`.
- Governance/support pages: `/explore/`, `/mission/`, `/source-policy/`, `/source-map/`, `/related-links/`, `/page-templates/`, `/community/`, `/contribute/`, `/governance/`, `/governance/review-gated-publication-model/`, `/roadmap/`.
- Template/scaffold routes: `/model-registry/`, `/model-registry/model-profile-template/`, `/model-registry/compare-models/`, `/evaluation-hub/`, `/evaluation-hub/benchmark-page-template/`, `/evaluation-hub/benchmark-contamination/`, `/security/`, `/security/owasp-llm-top-10/`.

Root discovery files are `llms.txt`, `robots.txt`, and `sitemap.xml`.

## AIWikis Multisite Rule

When LlmWikis guidance discusses AIWikis.org as a shared, multi-project LLM Wiki, the public namespace rule is hard: every `https://aiwikis.org/{site}/` route is its own AI Wikisite. Site-owned files, concepts, reports, source guides, and long-memory retrieval pages belong under that site's namespace, for example `/llmwikis/files/` or `/uaix/files/`. Global company/workspace guidance belongs under `/org/`, and `/site/` is the site directory. Root memory indexes such as `/files/`, `/concepts/`, `/reports/`, `/browse/`, and `/browser/` are invalid because they cause agents to traverse sibling-site memory during targeted work.

## Current Support Claims

Current:

- Launch-ready WordPress package using `llmwikis-knowledge-theme` and `llmwikis-core`.
- Managed public routes with per-route descriptions, canonical links, social metadata, JSON-LD, and sitemap/`llms.txt` coverage.
- Public source map at `/source-map/` that groups high-value routes by job, names LlmWikis/UAIX/AIWikis authority boundaries, explains discovery files, and points readers to the smallest useful next route.
- Public LLM Wiki specification at `/reference/specification/` that defines practical non-normative contracts, conformance profiles, required artifact set, minimal page record, search/retrieval requirements, validation gates, and links to metadata, navigation, repair, setup wizard, and UAIX boundary context.
- Styled public search results using the WordPress search template, with task shortcuts into specification, metadata, repair, source policy, AI Memory, and setup-wizard routes, visible search-state panels, result cards, and no-result recovery links.
- Dynamic LLM Wiki Starter Bundle at `/downloads/llm-wiki-starter-bundle-v2.8.0.zip`, generated from the same registry displayed on `/starter-template/`, including `llm-wiki/agent/ORCHESTRATION_RUNBOOK.md` for agentic orchestration operating rules, `llm-wiki/agent/TASK_PACKET_TEMPLATE.md` for one-run task packets, and `llm-wiki/agent/SUPPORT_ESCALATION_CHECKLIST.md` for blocked-run support escalation.
- Shared setup wizard at `/tools/llm-wiki-setup-wizard/` for humans and visiting AI agents. It is a step-by-step browser planner for new LLM Wikis, existing documentation migrations, Agent File Handoff intake, selectable Project Handoff alignment, combined File Handoff plus LLM Wiki memory without requiring Project Handoff, Canonical AI Memory layer routing, AI Dreaming Memory review-only consolidation, Knowledge Graph setup, Long-Run Agent Wiki Strategy planning, and skill/capability catalog setup. It asks about project stage, collaboration model, single-site versus multisite workspace routing, required `.uai/coding-standards.uai` for code-capable setup, per-repository Git preflight, mutable runtime artifact policy, raw/wiki/index/log paths, entity page and episodic log patterns, transfer evidence logs, archive targets, source collection, update timing, AI Dreaming Memory mode/input scope/run path/candidate path/contradiction path/review policy/metrics path, trust labels, review gates, linting, source policy, coding standards, local File Handoff buckets, Project Handoff follow-up fields only when that option is selected, UAIX AI Memory package file alignment, optional `LLM_WIKI_MEMORY_PLAN.md` boundaries, support escalation, graph strategy, graph storage model, stable IDs, claim/source-span policy, review-state policy, graph validation, export boundary, retrieval/abstention policy, Long-Run Agent Wiki mode fields for Folded Run State, Active Subgoal, Completed Subgoal Summary, Agent-Staged Page, Process-Quality Checks, Retrieval Budget Checks, and skill/capability boundaries. It generates a browser-only Markdown setup packet plus structured setup model JSON with setup-readiness checks, first files, first actions before ingest, direct fragment/query setup paths, local browser draft restore, and an embedded visitor-AI digest. Homepage keeps the structured knowledge-flow graph in the top hero using a rendered image, while the wizard visual entry cards sit below the hero flow and on the wizard page. The AI-agent card preserves `https://llmwikis.org/tools/llm-wiki-setup-wizard/`.
- Handbook guidance for `raw/`, `wiki/`, `AGENTS.md`/`CLAUDE.md`, `index.md`, `log.md`, source policy, trust labels, metadata, lifecycle states, ingest/query/lint, graph navigation, Knowledge Graph design, examples, security/privacy, and hot/cold memory separation.
- Public guidance for using Codex with LLM Wiki, review-gated publication, Canonical AI Memory layers, and the relationship between durable LLM Wiki memory, compact UAI AI Memory, and UAI Project Handoff.
- Public guidance for AI Dreaming Memory as offline, review-only proposal memory over visible source-backed wiki memory, with raw run artifacts, candidate queues, contradiction drafts, metrics, protected surfaces, and reviewed promotion gates.
- Public guidance for Long-Run Agent Wiki Strategy at `/guide/long-run-agent-wikis/`, using only neutral LlmWikis terms: Context-Folded Agent Run, Folded Run State, Macro Context, Micro Context, Active Subgoal, Completed Subgoal Summary, State Review, Active Wiki Navigation, Index-First Retrieval, Evidence-Scoped Retrieval, Agent-Staged Page, Process-Quality Checks, and Retrieval Budget Checks. It is handbook/profile guidance and browser-only wizard packet generation, not a hosted service, hidden memory system, model-training feature, public MCP/write API, autonomous publisher, certification, endorsement, or conformance claim.
- Public guidance for multisite work now includes an operational Git preflight: check each repository named by `workspace.uai` separately for merge state, unmerged paths, ahead/behind status, and tracked generated/runtime artifacts before Sync, pull, merge, commit, or push. WordPress Studio SQLite databases are explicitly treated as mutable local runtime artifacts, not durable source truth.
- Public guidance for using LLM Wikis with agentic agents and orchestration layers, with a compact Agentic Orchestration Mode on `/for-ai-agents/` and deeper how-to guidance at `/guides/llm-wiki-agentic-orchestration/`. The guidance is framed as governed source memory, staged proposal flow, support escalation, review evidence, and update boundaries around existing runtimes rather than a LlmWikis-owned runtime or support service.
- Public comparison guidance for LLM Wiki vs RAG and LLM Wiki vs AI Memory.
- Repair guidance for existing wikis includes a standards scorecard, input/platform map, migration/remediation outputs, and a clear automation boundary for human-approved versus human-decided changes.
- Non-normative UAI/UAIX explainers with canonical UAIX links.
- Local `AGENTS.md`, `readme.human`, `.uai` handoff bundle, and Agent File Handoff intake pattern. Current public guidance requires a complete intake outcome for safe relevant files: reviewed disposition, actual site/system work, hot-memory outcome, LLM Wiki/AIWikis preservation outcome when configured, targeted checks, blockers, and source-site removal or keep-active reason only after those records exist.
- Permanent `410 Gone` retirement for the former `/content-quality/` route family.
- Targeted verification policy in `.uai/test-plan.uai`.

Not current:

- Open public editing.
- Live benchmark dashboards or real-time model telemetry.
- Automated arXiv, repository, or web ingestion.
- Public MCP server.
- Semantic, vector, conversational, or AI-generated search results beyond normal WordPress search.
- Hosted graph database, public graph API, public SPARQL endpoint, automatic graph sync, or public GraphRAG service.
- Official UAIX `.uai` generator, hosted validator, certification, endorsement, SDK, or CLI support.
- Public skill-folder starter bundle, capability catalog, capability registry, capability linter/validator, official adapter matrix, or portable permission profile for modular agent capabilities.
- Hosted AI Dreaming Memory service, hidden model memory, model-weight training, autonomous repository writer, automatic LLM Wiki sync, automatic publication, or dream-run promotion without review.
- Public multilingual support.
- Automatic repository writes, hosted import validation, automatic LLM Wiki sync, automatic publication, public MCP, public write APIs, open editing, and official UAIX generator/SDK/CLI/certification/endorsement/conformance claims from the setup wizard.
- Memberships, grants, or sponsor programs.

## Hot/Cold Memory Boundary

Public guidance now teaches that LLM Wikis are the cold/deep layer and AI Memory or Project Handoff packets are the hot/portable layer.

Source-site practice follows the same rule:

Why This File Exists

This is a UAI AI Memory handoff file from llmwikis.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: LlmWikis Public Surface; Positioning; Primary Route Families; AIWikis Multisite Rule; Current Support Claims; Hot/Cold Memory Boundary; Discovery Boundary; Verification Boundary. 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

• Source overview
• Site file index
• Site report index
• UAI system index
• Source provenance
• Site directory
• Organization reports

Provenance And History

• Current observation: 2026-06-22T01:56:21.9510185Z
• Source origin: current-source-workspace
• Retrieval method: local-source-workspace
• Duplicate group: sfg-259 (primary)
• Historical hash records are stored in data/hashes/source-file-history.jsonl.

Machine-Readable Metadata

{
    "title":  "LLMWikis Public Surface",
    "source_site":  "llmwikis.org",
    "source_url":  "https://llmwikis.org/",
    "canonical_url":  "https://aiwikis.org/llmwikis/files/uai-public-surface-uai-32044978/",
    "source_reference":  ".uai/public-surface.uai",
    "file_type":  "uai",
    "content_category":  "uai-system",
    "content_hash":  "sha256:3204497887df20f7d843c85607664daa52af0ab4b971a13fa1504c484ca5b7cd",
    "last_fetched":  "2026-06-22T01:56:21.9510185Z",
    "last_changed":  "2026-06-13T22:24:46.1373887Z",
    "import_status":  "unchanged",
    "duplicate_group_id":  "sfg-259",
    "duplicate_role":  "primary",
    "related_files":  [

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