Giving agents access to a sandboxed environment with a shell and a filesystem has been the latest hype when it comes to agentic harnesses. Recent examples of this include: Turso’s AgentFS Anthropic’s Agent SDK, which brings Claude Code’s harness to non-coding domains Vercel rebuilding their text-to-SQL agent on top of a sandbox Anthropic’s Agent Skills for filesystem-based progressive disclosure The argument for why this is good goes something like this: The big labs are doing heavy RL for coding tasks in these kinds of environments. Aligning more closely with such a harness brings free gains from the coding domain to other problem spaces.