If you're running AI agents on your local machine, you've probably thought about sandboxing them. Most people jump straight to VMs, but there's a simpler way: give the agent its own drive letter.
Let me compare the two approaches.
VMs: The Heavy Approach
VMs work, but they have downsides:
- Slow to start
- Resource-heavy
- Annoying to share files between host and guest
I used to set up a VM every time I ran an AI agent on a new project. It worked, but it was slow and heavy.
Drive Letters: The Lightweight Approach
The drive letter approach is much simpler:
- Create a dedicated folder for the AI agent's work
- Map that folder to a drive letter on Windows
- Point the agent only at that drive
The agent sees only that drive. It can't touch anything outside. OS-level isolation, no VM overhead.
Two Tools, Two Scenarios
I use two tools depending on where the files live:
Local Files: ScsSubst
If the agent works with local files, I use ScsSubst to map the local folder to a drive letter.
It's lightweight, portable, and one-time purchase. Way cheaper than VM licenses.
Cloud / WebDAV Files: ScsDrive
If the data lives in the cloud (WebDAV, NAS, etc.), I use ScsDrive to mount the cloud folder as a drive.
Same isolation, no local safety concerns. The agent works with cloud files but still can't escape the drive.
Give It a Try
Next time you run an AI agent, don't set up a whole VM. Just create a dedicated folder, map it to a drive, and let the agent work inside that drive only.
Both are one-time purchases, way cheaper than VM licenses.
Top comments (0)