TreeOS is launching in the next few days.
🌳

TreeOS

An Operating System for Context

Persistent, structured memory that AI can navigate. Not another chat window. A filesystem for knowledge that compounds over time.

Learn More
Example App built on the protocol

What is TreeOS?

A self-hosted system where you and AI build hierarchical knowledge structures. Trees replace flat conversations with navigable, persistent context.

Trees

Nodes with notes, trackable values, and children. Navigate with cd and ls. Build with mkdir. Not flat files. Not chat logs. Structure that persists.

AI Modes

Three strict modes. Chat reads and writes. Place adds content without conversation. Query reads only, changes nothing.

Node Types

goal, plan, task, knowledge, resource, identity. Six core types. Custom types valid. Free-form strings that tell agents what they're looking at.

Terminal Native

The CLI works like a regular terminal. cd, ls,mkdir, rm, mv. If you know a shell, you know TreeOS.

treeos
tabor@treeos.ai/Goals › ls
Fitness · Career · Reading · Side Projects
tabor@treeos.ai/Goals › cd Fitness
tabor@treeos.ai/Goals/Fitness › tree
Fitness
├─ Chest Day
├─ Leg Day (completed)
└─ Cardio
tabor@treeos.ai/Goals/Fitness › chat "add a back and biceps routine"
Created: Back and Biceps
Pull-ups 4x8, Rows 3x10, Curls 3x12
Navigate cd, ls, pwd, tree
Build mkdir, rm, mv, rename, type
Content note, notes, value, goal
AI chat, place, query
Extensions ext install, ext disable
LLM llm add, llm assign

How It Works

1

Start a Land

Your server. Stores trees, runs AI, serves the API. First boot walks you through config and extension selection.

treeos land
2

Connect

Point the CLI at any land. Register. Connect your own LLM or use the tree owner's model.

treeos connect http://localhost:3000
3

Plant a Tree

Name it. Type it. Navigate with cd and ls. Build branches with mkdir. Add notes and track values.

treeos mkroot "Fitness" --type goal
4

Use It

Chat creates and edits. Place adds content where it belongs. Query reads without changing anything. You and AI work the same tree.

treeos chat "break this into weekly tasks"

Extensions

Core protocol: nodes, notes, values, types, AI modes. Everything else is a package you install, disable, or build.

AI and Knowledge

understandingBottom-up tree compression
dreamsDaily background maintenance
raw-ideasCapture and auto-place

Developer

scriptsSandboxed JS on nodes
api-keysProgrammatic access
prestigeNode versioning
schedulesDates and calendar

Finance

energyUsage metering
billingStripe subscriptions
solanaOn-chain wallets

Content

blogLand-level posts
bookShareable note exports
html-renderingServer-rendered pages

System

user-llmCustom model connections
user-queriesNotes, tags, chats
deleted-reviveSoft delete and recovery
visibilityPublic/private trees
transaction-policyTrade approval rules
Extension docs

Two Ways In

Run Your Own Land

Self-host. Pick extensions from the registry. Your data, your models, your rules.

npm install -g treeos
treeos land
Land setup guide

Join a Land

Connect to an existing land. Register, bring your own LLM or use the tree owner's. Start building.

npm install -g treeos
treeos connect https://treeos.ai
treeos register
Getting started

Network

Lands connect through the Canopy protocol. The directory handles discovery, public tree indexing, and the extension registry.

Peer

Direct land-to-land connections. Signed requests.

Discover

Search lands and public trees across the network.

Collaborate

Cross-land invites. Remote contributions. AI proxy.

Registry

Publish and pull extensions. Shared package library.

Get Started

One package. Two commands.

Terminal
$ npm install -g treeos
# Start a land (server)
$ treeos land
Welcome to TreeOS.
Domain (localhost): _
# Or connect to an existing land
$ treeos connect https://treeos.ai
$ treeos register