Documentation

Everything you need to build, deploy, and customize your own AI with Chvor.

2 min read

Documentation

click to navigate

Start here — get Chvor running in under 5 minutes:

curl -fsSL https://raw.githubusercontent.com/luka-zivkovic/chvor/main/scripts/install.sh | bash

Or with npm: npm install -g @chvor/cli && chvor onboard

Then head to Installation for detailed setup, or First Run to understand the brain canvas.

Getting Started

The essentials to install, configure, and run Chvor.

Installation — Three methods: from source, Docker, or the desktop app
First Run — Your first conversation and the brain canvas
Configuration — LLM providers, channels, and settings

Core Concepts

How the platform works, from the brain canvas to memory.

Brain Canvas — Real-time visualization of your AI’s reasoning
Skills — Behavioral definitions that shape how your AI acts
Tools & MCP — Connect any tool via the Model Context Protocol
Channels — Web, Telegram, Discord, Slack — same AI everywhere
Memory — Vector embeddings, semantic recall, encrypted storage
Schedules — Cron-based task automation
Directives — Persistent rules that shape behavior

Guides

Step-by-step walkthroughs for common tasks.

Creating Custom Skills — Write YAML skill files
Connecting MCP Tools — Add filesystem, search, databases
Channel Setup — Telegram, Discord, Slack integration
Deployment — Self-host on a VPS with nginx and SSL
Docker — docker-compose, volumes, environment variables
Desktop App — Native installer for Windows and macOS

Reference

Complete specifications and API documentation.

Environment Variables — All configuration options
Skill Schema — Full YAML specification
API & WebSocket — REST endpoints and real-time protocol
CLI & Scripts — Development commands and workflows

Registry

Community skills and tools.

Registry Overview — Browse and install community packages
Publishing — Share your skills and tools