“Make it nicer” will lead to unpredictable and controversial changes, likely to clash with the project owner’s vision. The owner would be better off using the LLM as a collaborator, if anything, to decide where to take the design.

However, UX can also be improved by some simple heuristics that could be seen as mechanistic by the standards of modern LLMs.

Performance and responsiveness, for example, contribute to UX even when the UI remains fixed. I’d wager Fable-level LLMs could improve perf on most open-source GUIs by benchmarking typical use cases and progressively applying standard tricks like caching, buffering, batching network calls.

Claude Design can help. Fable is pretty good at UI when you narrow down to a specific page or—better—something like a modal, table, etc.

If someone is a user, they'll have specific complaints. Claude Design is more than capable, which can then be fed directly into Claude Code to implement, test, iterate, and ship.

I agree that it lags behind coding and imo the highest leverage area for Anthropic to invest in. Hopefully they do so; UI is consistently one of the main (severe) differentiators between OSS and paid options.

No but if you have specific asks like “make this resizable” “I should be able to drag and drop to combine” or “add a button on the top right to close” it’s quite good at directed changes