The open-source AI agent ecosystem is genuinely impressive. AutoGPT, CrewAI, LangGraph, and similar frameworks enable powerful multi-agent architectures. They also require Docker, Python environments, dependency management, and comfort with a terminal. This is not a barrier for developers — but it is a near-total barrier for everyone else. The people who could benefit most from AI automation are often not the ones who can install and maintain these stacks.
The Docker Tax
Docker adds approximately 1-2 hours of setup time for a non-developer, a 4GB+ disk footprint, ongoing maintenance when containers break, and a cognitive model (containers, images, volumes, ports) that is orthogonal to the actual task the user wants to accomplish. The AI agent frameworks that run on Docker are powerful — but "powerful and only accessible to developers" is a very different value proposition from "accessible to everyone."
The Skales Approach
Skales is a native desktop app. Download a .exe or .dmg file, double-click it, answer three setup questions, and you have a fully functional AI agent running on your machine. No Docker, no terminal, no Python, no dependency resolution. The entire setup takes under two minutes. The agent capabilities — email, browser automation, file management, calendar, custom skills, autonomous scheduling — are all accessible through a chat interface that works like any messaging app.
The trade-off is that Skales is less customisable at the infrastructure level than a Docker-based framework. If you need to run 50 AI agents in parallel on a server, Docker-based solutions are the right tool. If you are an individual or small team who wants AI automation without an engineering degree, Skales is the right tool. Full comparison here.