VibeKit

The premise is that you should never run an autonomous coding agent directly on your machine, and this packages that discipline into something convenient enough to actually adopt. Spinning a sandbox per agent was quick, and the boundary felt real rather than decorative when I tested file and network access. The open questions are about resource overhead when running many sandboxes at once, and how cleanly secrets are injected without leaking into the agent context. The abstraction over different agents is tidy and kept my scripts portable. As infrastructure for taking agents seriously without taking risks, it is exactly the sort of unglamorous tool that ends up load bearing.

Running any coding agent inside a clean isolated sandbox is the responsible default we have been missing, and this makes it almost effortless. The isolation held when I deliberately let an agent do something silly. I would like the network policy to be configurable per run, but this is the right direction.

One wrapper, any agent, a real boundary around it. Setup was a single command and it did not get in my way. The peace of mind is worth the small overhead.