Exploring agent-friendly runtime diagnostics for Vite #21783
MapleCity1314
started this conversation in
General
Replies: 1 comment 1 reply
-
|
Hey @MapleCity1314, Vite is usually blazingly fast, but issues like this often crop up around dependency pre-bundling or rollup configurations.
Let me know if clearing the Vite cache or tweaking the Rollup config gets you past this hurdle. Hope this points you in the right direction! Let me know how it goes. Happy coding! |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I have been exploring a problem that feels under-served in the current Vite ecosystem:
debugging a running Vite app from an agent loop is a different problem from browser automation, and a different problem from human DevTools usage.
Most tools seem to fall into two buckets:
That leaves a gap:
a Vite-focused runtime diagnostics layer designed for agent workflows from the start
I built a small project around that idea:
vite-browserThe point is not to be another browser wrapper. The point is to expose Vite-aware runtime signals and framework state as structured CLI output:
There are also supporting utilities like logs, network inspection, screenshots, and page eval, but those are secondary. The main idea is making runtime diagnostics queryable in a way that fits an agent loop.
The command model matters too.
Each CLI call is a one-shot request against a long-lived browser daemon. That gives agents a simple command surface while avoiding the cost of rebuilding browser lifecycle management on every step.
In practice, that means an agent can:
What I am interested in is whether this framing resonates:
Project:
https://github.com/MapleCity1314/vite-browser
Beta Was this translation helpful? Give feedback.
All reactions