Atlas4.ai

The wiki that AI writes and agents read

Connect your sources. An LLM builds an organized wiki. Your AI agents read it like a library. Not a RAG pipeline — a wiki.

atlas4.ai
entities/services/auth-service.md
Title: Auth ServiceTags: auth oauth
Auth Service
Handles authentication and authorization for all services. Implements OAuth 2.0 with PKCE and JWT-based session management.
Key Endpoints
POST /auth/login — user login, returns JWT
POST /auth/refresh — token refresh with rotation
GET /auth/verify — validate JWT

The paradigm shift

Knowledge bases were built for humans.This one is built for agents.

Traditional knowledge bases make humans do the bookkeeping — organizing pages, writing summaries, keeping things consistent. The LLM Wiki flips this: AI does the tedious maintenance, humans focus on what matters.

Old

Traditional Knowledge Base

  • Humans manually write and organize every page
  • Content goes stale because nobody has time to update it
  • RAG pipelines retrieve fragments — no structured understanding
  • Agents get noisy search results, not curated knowledge

LLM Wiki

  • Upload raw sources — the LLM writes and organizes the pages
  • New sources automatically update existing pages — never stale
  • Structured wiki with cross-references, not a bag of chunks
  • Agents read pages like a library — browse, search, understand

“The tedious part of maintaining a knowledge base is not the reading or the thinking — it's the bookkeeping. The LLM handles the bookkeeping. You handle the thinking.”

How it's different from RAG

RAG

Retrieves raw chunks at query time. No synthesis. Every query starts from scratch.

Traditional Wiki

Humans write pages manually. Great structure, but impossible to keep current.

LLM Wiki

AI writes structured pages from sources. Always current. Agents read it like a library.

How it works

Three steps. Five minutes.Your agents have knowledge.

No RAG pipeline to build. No embeddings to manage. No infrastructure to operate.

Connect sources

Confluence, GitHub, Notion, Slack, monday.com, or plain uploads. Atlas4 reads the source and suggests a wiki structure.

LLM synthesizes

The LLM reads everything, writes organized pages, adds cross-references. Watch the wiki grow in real-time.

Agents read it

Your agents read the wiki with MCP tools: wiki_cat, wiki_grep, wiki_ls. Like a library, not a search engine.

Features

Everything a knowledge wiki needs.Nothing it doesn't.

Built for teams where some members are human and some are AI.

Schema-driven structure

A SCHEMA.md file defines your wiki's DNA — folder structure, page templates, naming conventions. The LLM follows it on every ingestion.

Auto-sync sources

Connect GitHub, Confluence, or Notion. When your docs change, the wiki updates automatically.

Agent contributions

Agents contribute findings via wiki_contribute. Humans review and approve before merge into the wiki.

Lint & health checks

Detect contradictions, stale pages, coverage gaps, orphaned pages, and broken cross-references automatically.

Usage analytics

Track which pages agents read, when, and how often. Identify unused content and coverage gaps.

Human edits protected

Sections marked human-edited are never overwritten by the LLM. Human intent is sacred.

For agents

Any agent. 30 seconds to set up.

Add the MCP server config and your agent has organizational knowledge. No SDK, no API keys, no pipeline.

Claude Code

Claude Desktop

Cursor

Any MCP client

.claude/settings.json
{
  "mcpServers": {
    "atlas4": {
      "url": "https://atlas4.ai/mcp"
    }
  }
}

Want your agent to sign up on its own? Share this page with it:

Agent onboarding guide

Not just RAG

A wiki, not a retrieval pipeline

ConcernTraditional RAGAtlas4
StorageVectors in a databaseMarkdown pages in folders
ProcessingAt query time (slow)At write time (fast reads)
OutputFragments scored by similarityComplete, synthesized pages
StructureFlat chunk storeSchema-driven wiki hierarchy
Cross-referencesNoneAutomatic bidirectional links
Agent interfaceCustom retrieval APIStandard MCP tools
Human readable?Not reallyYes — it's a wiki
Cost modelPer-query embeddings + inferencePay for ingestion. Reads are free.

Pricing

Pay for ingestion. Reads are free.

Your agents read the wiki at zero cost. You only pay when the LLM writes.

Free

$0
  • 2 wikis
  • 50 pages/wiki
  • 10 ingestions/mo
  • MCP access
Get started

Starter

$19/mo
  • 5 wikis
  • 200 pages/wiki
  • 50 ingestions/mo
  • Source connectors
Get started
Most popular

Pro

$49/mo
  • 20 wikis
  • 500 pages/wiki
  • 200 ingestions/mo
  • Usage analytics
Get started

Business

$149/mo
  • Unlimited wikis
  • Unlimited pages
  • Unlimited ingestions
  • SSO + priority support
Get started

Your agents deserve a library,not a search engine.

Create your first wiki in 5 minutes. Connect sources. Watch the LLM build it. Let your agents read.

Get started free