ContenoxContenox
Open source · Free forever

Describe a goal.
Get it done.

Persistent AI plans. Real shell execution. Pipe any data in, get results back. One binary — no cloud, no daemon, no babysitting.

🏠 Fully local with Ollama🔌 OpenAI / Gemini / vLLM📋 Plans survive reboots🔍 Review before every step
contenox
▋

Install

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

Or build from source

The Cookbook

Verified recipes for real engineering tasks. Copy a command, run it, ship faster.

Generate release notes

$ git log --oneline | contenox run release-notes
## v0.9.1 — Hook policies, SQLite fixes, CLI cleanup
Full recipe

AI-powered commit messages

$ git diff --staged | contenox run commit-msg
feat(hooks): add chain-scoped security policies
Full recipe

Scaffold a project from a goal

$ contenox plan new "scaffold a Go HTTP server with tests" && contenox plan next --auto
Step 1/5: Create project structure... ✓
Full recipe
Browse all recipes

How it works

Four steps. No surprises.

1

Describe

Write a goal in plain English — or pipe stdin directly.

2

Plan

Contenox turns it into a persistent, step-by-step plan saved to SQLite.

3

Execute

Each step runs real shell commands on your machine, one at a time.

4

Review

Approve sensitive steps, run fully-auto in CI, or watch live in the TUI.

Connect anything

Remote hooks let Contenox call external APIs as tools during execution. Open source. Bring your own — or use ours.

Any HTTP API

Point Contenox at any OpenAPI service and the model gains those tools automatically.

Hosted hooks coming soon

Coming soon

Managed RAG, Playwright, and secure code sandboxes — bring your own or use ours.

Read the hooks docs