Quickstart

1. Install

macOS / Linux (one line):

curl -fsSL https://contenox.com/install.sh | sh

Or download the binary directly from GitHub Releases.

The whole path — install, setup, first prompt — in one take:

Install demo: install.sh, contenox setup, and a first answer

2. Initialize a workspace

Run this once in the project directory you want Contenox to work in:

contenox init

This creates the workspace marker and writes the default chain and HITL policy presets.

3. Connect a model

For the local path, install Ollama, pull a model, and point Contenox at it:

ollama pull qwen3:8b
contenox setup          # pick Ollama, done
contenox doctor

See the Ollama guide for details, including Ollama Cloud.

Run your first prompt:

contenox "hello, what can you do?"

contenox backend list showing local and hosted providers, then a first chat on a local model

Chat is always session-backed — history persists across invocations automatically. Pass -e to compose your message in $EDITOR instead of on the command line:

contenox chat -e

4. Optional editor use

Contenox can also run inside editor or desktop clients that speak ACP. The same chains, model config, tools, and HITL policy are used either way:

Cloud providers

Contenox needs at least one model to work. Pick the option that fits:

Option	What you need
Ollama	Ollama installed locally, or an Ollama Cloud key
Google Gemini	A free Gemini API key (no GPU)
OpenAI	An OpenAI API key
Anthropic	An Anthropic API key (Claude)
AWS Bedrock	An AWS account with Bedrock model access

If you’re not sure, start with Ollama for a fully local setup, or Gemini for a free hosted key.

Next steps

The terminal UI — run contenox new for chat, plan, and shell in one persistent session
Your first chain — author your own agent in five edits
Core concepts — how chains, tasks, and tools fit together
MCP integration — connect external tools