Feedback

I would love the ability to edit chats like in chatgpt and playground. I find that I sometimes don't like the answers i'm getting and have to start the entire chat in another window. it would be great to just edit the chat.

I have had a couple of issues either inadvertently deleting text or moving a scene summary and dropping it in the wrong place or doing a copy paste that I changed my mind on etc. where a quick undo feature would have helped.

Problem: User wants to see past versions of a scene summary or chapter. User clicks past version, but it's difficult to tell what exactly was changed. Proposal: Add highlight bg to changed text. Optionally, collapse identical lines/paragraphs (e.g. VSCode git diff).

You need a center text option in text alignment in the formatting menu. For example, for centering quotes.

It's a super minor improvement but would help people like me with a million different Codex categories that are starting to get a little jumbled and making it harder to organize everything.

It'd be cool if I could make each of my novel dashboards have a different background color - see screenshots - could offer up a few options based on the novel's cover (like canva does)

WHy do I need to re-add all the details if the information is all within the text? The point of AI should be able to scrub through the text and gather that data.

Please add support for message editing, response retry, and conversation branching within the chat interface. These features are standard in many modern LLM front-ends and greatly improve iteration speed, experimentation, and overall usability. When interacting with the LLM, users often want to: * Correct or refine a previous prompt without rewriting the entire conversation. * Retry the last model response to explore alternatives or fix generation errors. * Fork a conversation into a new branch to explore different directions without losing the original context. Currently, the interface does not provide edit/retry/branching tools, which forces users to copy/paste or restart flows manually. This disrupts workflow and makes iterative prompting much less efficient.

It clutters up the screen. If the prompt is empty, there's no context, no nothing, then please don't save empty threads. Just save them as drafts if I started writing a prompt. But not if the prompt text box is empty.

When highlighting text in the main text editor, then creating a Snippet from it (with the +Snippet button), it would be amazing if that text would stay highlighted after navigating back to the main text editor from the newly created snippet window. 99% of the time when I'm creating a snippet from text in the main window, I then want to delete the exact same text (since I now have it stored in a snippet for kitbashing) and it's kind of a pain to have to re-highlight it again just to delete it, especially if it's a longer passage.