What happens when you give an AI agent root-level access to your entire computer? Jeriko, a new tool from Etheon Inc., is betting the answer is "productivity" rather than "chaos."
Jeriko installs as a single binary daemon on macOS, Linux, or WSL (Windows Subsystem for Linux). Once running, it takes natural language commands and translates them into OS-level actions: managing files, controlling browsers, sending emails, scheduling calendar events, running terminal commands, and automating workflows. The pitch is one CLI to rule your whole machine.
The feature list is ambitious. Jeriko claims 23 capabilities including browser automation with real Chrome cookies and anti-detection, parallel research using up to 20 sub-agents, 24/7 automation via cron jobs and file watchers, and integrations with services like Stripe, GitHub, Gmail, Google Drive, and Shopify. It can also build web apps from 40+ templates with a live dev server, generate images via DALL-E 3, and transcribe audio through Whisper. You can even control it through Telegram or WhatsApp.
The "no cloud lock-in" angle is interesting. Jeriko runs locally and supports multiple AI backends: OpenAI, Claude, Ollama (for fully local models), or custom providers. Your data stays on your machine, which matters for anyone handling sensitive business information.
But ambitious feature lists from brand-new tools deserve skepticism. Browser automation that claims anti-detection and CAPTCHA handling is the kind of thing that sounds great on a landing page and breaks constantly in practice. And giving an AI agent control over your files, processes, and clipboard requires a level of trust that Jeriko has not yet earned through a track record.
No pricing is listed. Installation is a one-line curl command. If you are comfortable running an AI daemon with broad system access on a development machine, Jeriko is at least worth watching as the "AI operating system" category takes shape alongside tools like Replit Agent and similar code-focused assistants. Just maybe do not point it at your production environment on day one.