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:

2. Connect a model

contenox setup is the entry point. For the local path, install Ollama and pull a model first:

ollama pull qwen3:8b
contenox setup          # pick Ollama, then pick your model from the list

When Ollama is running, the wizard reads the models you have actually pulled and offers them as a numbered menu — press Enter to take the suggested one.

See the Ollama guide for details, including Ollama Cloud.

Then confirm you can chat:

contenox doctor

Its first line is the verdict:

Ready: yes — chat now with `contenox new` or `contenox "your prompt"`.

If it says Ready: no, the line under it names the one command that fixes it.

3. Initialize a workspace

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

contenox init

This creates the project-local .contenox/workspace.id marker; the default chains and HITL policy presets live globally in ~/.contenox (a workspace-local file with the same name overrides its global counterpart). See Your first chain for the full layout.

4. Start working

The terminal UI is the main surface — chat, plan, and shell in one persistent session:

contenox new

The transcript flows into your terminal’s own scrollback, / opens commands, ! runs a shell line, @ attaches a file, and gated tool calls are answered inline with one keystroke. Press ? on an empty composer for the full key list.

new always starts fresh. To carry on where you left off, reopen the last session with its transcript replayed:

contenox resume

For one-shot and scripted use, pass the prompt directly:

contenox "hello, what can you do?"
echo "summarise README.md" | contenox

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

5. 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:

• Use from Zed
• Use from JetBrains
• Use from AionUi
• Use from OpenClaw

Cloud providers

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

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

Next steps

• Your first chain — author your own agent in five edits
• Core concepts — how chains, tasks, and tools fit together
• How contenox compares — what it shares with the coding agents, and the three things that are built differently
• MCP integration — connect external tools
• Workspace index & search — ask the repo a question, get file:line citations back
• Pairing a machine with a relay — reach a running session from your phone: one typed key, optional always, free for you and three teammates (one machine each)
• AI sovereignty & the EU AI Act — hosting, state, and oversight controls you own