TypingMind (→ feedback.typingmind.com)

When importing a large number of chats from Claude (JSON export), allow TypingMind to: Recognize the original project/agent for each chat (based on metadata or folder in Claude export) Automatically assign imported chats to corresponding folders (projects) or agents in TypingMind Help users maintain their original organization from Claude with minimal manual effort This would save time for users migrating from Claude and make it much easier to get started in TypingMind.

Can we have conversations where we can choose multiple bots to talk to - similar to Silly Tavern? https://sillytavernai.com/ For example, then we could have brainstorming sessions with different bots with different skills, or learn languages in a room with multiple bots for more immersion. This will also open up the program to more people.

Sometimes I need to refer to an old chat, but its not super easy to find it, or even remember that it exists. It would be really cool if I could chat with the AI in typingmind about the rest of my chats within typingmind. Basically RAG for my typingmind chat library.

Freedom to work in a pipeline like Flowise ( https://github.com/FlowiseAI/Flowise ), Langflow ( https://github.com/langflow-ai/langflow ) and Dify ( https://github.com/langgenius/dify ), those apps are great (not at the level of Typing Mind in my personal opinion) but, the feature to drag and drop block to ensemble the workflow, the inclusion of LangChain and LlamaIndex instead of a simple chat box gives amazing flexibility and much more powerful apps and chatbots can be created.

It would be extremely useful if we could selectively regenerate model responses when using multiple models in parallel. Often, one or two models give a satisfactory response on the first try, while the others need to be regenerated. Currently, we have to regenerate for all models, losing the good responses we would have preferred to keep and driving up the overall cost of the chat by having to needlessly regenerate some models' replies. Obviously we would not be able to edit the prompt when selectively regenerating, but this would allow us to fine-tune responses on a model-by-model basis and save on overall tokens per chat. Please consider this.

My question to the TM Help Bot: "Is it possible to fork a chat in TypingMind from the forking point onwards instead of backwards?" Its answer: "In TypingMind, when you use the Fork Chat feature, it creates a new chat that copies the conversation up to (and including) the forking point—meaning all messages from the chat’s start up to your selected message are included in the new chat. It is not currently possible to fork a chat and have the new chat contain only the messages from the forking point and onwards (“forwards-only”). The fork action always creates a copy that starts from the beginning up to your selected message, not just the branch after that message. If you need more advanced branching (like copying only the forwards/consequent messages), please consider submitting a feature request here: https://www.typingmind.com/ideas " So, here is my suggestion to add the possibility of forking "from here" in addition to the existing "until here".

I want to delete all chats outside a folder. This would be easy if when I click "Select multiple chats" each folder, in addition to each file, would also have a checkbox, which de/selects all the files in the folder. Or give me a simpler way to delete a lot of chats.

If Theo can do it, there's no reason it can't be applied on Typindmind. Currently, only Sonnet 3.7 Sonnet and Gemini 2.5 Flash support it in TM. I would love to see it for O3, O4 mini and Gemini 2.5 Pro here. I know OpenAI restricts some models unless you've verified your organization but even with that, I don't get the reasoning text.