Hacker News
points by bloppe 2 days ago | 0 comments
I recently got really weird with my personal computer after discovering browser DevTools MCP. I basically inverted the relationship between browser and desktop environment, booting directly to a browser in kiosk mode (with Ghostty quick terminal as well, so I always have easy terminal access) and running a local Weston server that shows up as a bookmarked web page on localhost, streamed over vnc, so I can still use legacy gui apps pretty normally if I need to, but I've basically refocused my whole gui environment on the web. I run code-server on localhost instead of VSCode and prefer web apps (often running locally) over traditional GUIs in general. I still use the terminal very heavily as well, and that's where I manage my pi agents.

Why? Because the web is an incredibly rich, open, text-based platform with massive representation in training sets, and with tools like DevTools MCP, your agent can do almost anything with your browser: run JS in the page, use the debugger, see network activity; it's extremely powerful.

My co-worker does something similar with emacs, another incredibly rich text-based "operating system" like a browser with comprehensive hookability, which he basically uses instead of a desktop environment.

[–] nathanfig 2 days ago
I'm trying to understand this. You do this so that your agents can do web based tasks for you, or so that they can do desktop tasks via a web browser? If the latter, for what kinds of tasks do you find terminal insufficient?
[–] bloppe 2 days ago
It's the former. I just find that the browser provides an incredibly reach ecosystem for the agents to operate in. Of course, you can also set up a devtools MCP connection while still using a traditional desktop environment, I just thought it was cool.

I mentioned the Weston fallback to illustrate that switching to a web-first GUI environment can be incremental.