› Why AON

Memory and RAG are table stakes now. What AON adds:

🔓 Obliterated and fully yours

An uncensored 27B model on a single consumer GPU (RTX 4090). Your chats, memory and documents never leave the box.

🧬 It extends itself

Paste a link and AON works out what it is: a model to pull, an MCP server to connect, a skill to import, or docs to learn from.

🔌 Plugs into your AI apps

aon mcp makes AON an MCP server for Claude Desktop, Claude Code and Cursor.

👥 One GPU, whole crew

vLLM serves several people at once, behind email-PIN sign-in with every user approved by hand.

🛡️ Honest by design

Shell commands and file writes wait for approval; "done" claims without a tool run aren't trusted; secrets never go to memory.

🗺️ Grounded, not guessed

Places and routes come from OpenStreetMap routing on a 3D globe; distances are computed, not invented.

› AON vs cloud AI chat

	🛰️ AON	☁️ Typical cloud AI chat
Runs on your own hardware	✅	❌
Your chats and files stay with you	✅	❌
Provider content filtering	❌ none	✅
Cost	your GPU's power	subscription or per token
Long-term memory + your documents (RAG)	✅	✅
Runs commands on your server (with approval)	✅	❌
Installs models, MCP servers and skills from a link	✅	❌
Usable from Claude Desktop / Cursor (MCP)	✅	❌

› MCP server

Use AON inside Claude Desktop, Claude Code, Cursor and any other Model Context Protocol client. Your MCP client starts aon mcp for you.

Tool	What it does
`ask_aon`	Hand AON a task; it answers with its own model, memory, documents and tools
`search_memory`	Search AON's long-term memory about you
`remember`	Save a fact to AON's memory
`search_documents`	Search your document library
`list_documents`	List the library
`add_document`	Add a document to the library

Shell commands and file writes are always refused over MCP: nobody is there to approve them.

Claude Desktop and Cursor config:

{
  "mcpServers": {
    "aon": { "command": "aon", "args": ["mcp"] }
  }
}

Claude Code:

claude mcp add aon -- aon mcp

› Demo

› Quick start

# 1. install
pipx install aon-agent

# 2. create ~/.aon with a starter config
aon init

# 3. web UI on http://127.0.0.1:8700
aon serve

# or: AON as an MCP server
aon mcp

You need Ollama with an embedding model and a chat model that supports tools:

ollama pull nomic-embed-text
ollama pull qwen3:14b

For several users at once, put the chat model on vLLM. Full docs: README on GitHub.