Feature Requests

I have very long technical chats and finding relevant sections again is difficult without a time stamp to refer to. A checkbox to turn then on/off would be excellent.

There are MANY things to praise about BoltAI, but there is one thing that is a constant annoyance: the fact that it does not behave like a normal app. (If there are configuration changes I can make that will make it behave normally, they are not clearly labelled: I looked through every setting I could find and nothing suggested that it would create the normal behavior that I would prefer. Here's what I mean: when I open it, it should ALWAYS be immediately visible, NOT hidden in the menu bar when I open it, it should ALWAYS be the top-most app when I switch away from it, it should NEVER move to the end of the command-tab menu but instead be second (just like any other normal app) when I do not use it for a time, it should NEVER hide itself from the command-tab menu It is okay if quitting the app (optionally) leaves a menubar app running in the background, but the main app should following the user's directions immediately. If I open it, it should remain open (and in my command-tab menu and Dock) until I quit it, at which time it should cease to be in my command-tab menu or Dock. I do NOT want to need to memorize a special global keystroke to "Activate [the] Main Window"—I just want to be able to navigate to it the way I navigate to any other apps. An app which demands that it be treated specially creates a bad user experience. Maybe in the future I might adopt a special keystroke, but that should NOT be forced on me or any other users. PLEASE MAKE YOUR OTHERWISE EXCELLENT APP ALSO EXCELLENT IN THIS RESPECT AS WELL. THANK YOU!

Trying to sniff interaction with a custom endpoint and i have Charles proxy setup. Would like option in advanced options to disable the ssl cert checking/enforcing. I'm totally ok showing a broken SSL icon or something.

Problem Thinking models are getting slower. As a user I am no longer actively waiting for the model to respond. I may start a parallel request or two in another folder. Then forget which chats are still active, and which are unread. Suggestion: Add an icon to the conversation to indicate request in progress request complete (unread) Would also be nice to have a "Recents" dynamic folder that would display all chats I initiated from a prompt manually (i.e. assistants and inline chats should be filtered out by default).

Currently BoltAI stores all the data in a SQLite database locally. Data synchronization with this SQLite db is quite challenging and so I was thinking about migrating to CoreData (native sync with iCloud) This can be quite time consuming as it can mess up the schema/data. Vote on this one to help me priotize this feature

Please add support for a RAG inside projects that we can query inside them. We use another tool for this and it's vital for our workflow to be able to query our knowledge bases without sending the entire file each time.

Hi, Could you please add support for the new Elevenlabs conversational API that they launched yesterday? Thanks!

As a user, I want to be able to have a conversation between agents. Logic: I ask Agent 1 to create an output. I ask Agent 2 to review output by Agent 1. Etc. Implementation idea: Allow summoning @assistants with '@' sign inside a prompt. Example: In a chat, I ask @ask_digital_strategist to write text of a landing page. Then, in the same chat I @ask_ICP_profile1 (my ideal customer persona) to share feedback how he likes the landing page text. Would he order my service? Then, I again @ask_digital_strategist to update his text based on provided feedback. It would be nice to be able to mention other @assistant everywhere (inside chats, prompts, assitants, commands).

It would be really really great if there was an option to for the chat bot to execute python code. Also the output should be able to contain images. Additionally it would be great if it could use files from your system inside its programs. Or maybe just specific files that you select before running the command. It would basically be similar to the code interpreter from GPT plus. Adding this feature would basically make this application a no brainer and extremely powerful. Of course this is not easy to implement but if this was added it would add massive value to the app. Of course not specifically for coders but it would just allow anyone to programatically process files, visualise data, create charts or do some moderately complex calculations with ease.