Even the prompt history is notoriously weak given how little Claude Code and some of the others display to give developers confidence in the process.

There needs to me more design rep indeed.

▲

tdeck 2 days ago | parent [-]

Claude Code displays plenty in my opinion, if you make it ask you for approval before each code change. You can read the code as it's being built up and understand if it's going in a bad direction before it does that and then piles on more and more slop.

The trouble is people don't want to bother reviewing the changes.

	▲	LoganDark 2 days ago \| parent [-]
		Claude Code used to stream the thinking process in verbose mode. Now that has been replaced with "transcript mode" which doesn't actually give much more information and also doesn't stream anything. They also recently removed (in certain situations) the counter of how many tokens the model's generated in its response in progress, so the only way to tell if it's stuck is to wait 10 minutes and then retry. Sure, I can read the diffs as they're generated (and I do). But proper transparency goes further than that, and it's being stripped away.