Why? Multi-tenant environments. First, we need to understand a few differences between environments:
So
Most people physically separate their tenancy, such as Claude Code, from their personal vs. work laptops. So in most cases, it's not a big deal.
But when you need multi-tenancy, it becomes super stressful. For example, say you have two different toolkits:
Most MCP auth states or code harnesses don't support profiles, so you can only log in to one.
So therefore... a natural evolution was to have both:
to physically isolate tenancies.
Now we've solved the multiple-profile issue, but the client's problems persist. Now let's get back to the environments:
All MCP auth or toolkit auth info should always be saved in the Agent Runtime Environment IMHO. However, a surprising number of harnesses tie them to the LLM server (such as Codex Apps or Claude.ai Plugins) or put them in the end-user UI (Claude Desktop or Codex Desktop).
Now the problem is:
The only way to reliably isolate different auth information is thus:
Then
are both isolated VPS, and
This way, you can provide different toolkits, creating multiple dev environments.
Pieter Levels is a Dutch entrepreneur known for his innovative work in the realm of digital nomadism. He's the creator of several successful online businesses and platforms, which he built while traveling the world. One of his best-known projects is Nomad List, a website that rates cities around the world based on criteria such as cost of living, internet speed, safety, and weather, to help digital nomads decide where to live and work. He also created Remote OK, a job board for remote jobs, and Hoodmaps, a crowdsourced map to navigate cities based on local insights. His work focuses on building platforms that enable and facilitate remote and nomadic work. He is also recognized for his minimalist approach to building startups, often working alone or with small teams, and his transparency about his revenue and methods, which he often shares online.