Are you prompting the model or is the model prompting you?
rewrAIt offers a unique 'text-editor' style interface for turn-based conversations with large language models that lets you revise any part of an LLM conversation- system, user, or AI messages, at any point in time. Add context, modify responses, remove information or change providers to your liking.
The entire interface is an active text editor with three main labels: @system:, @user: and @ai. Messages are sent to the llm provider when the enter key is pressed as the caret is in an active @user: section. AI responses are streamed back with an @ai: label. For newlines in @user: sections use shift + enter.
At any point you can edit any text seen on screen, regardless of whether it's been written by the user or generated by the LLM. Send @user: clear to clear the conversation without resetting the system prompt, and use the reset button to fully clear.
The save and load buttons both export the conversation into a .txt file, or load .txt files back into the interface.
The config button allows you to specify your provider and model of choice. Current providers supported are: openai, anthropic, and gemini. Ensure you have your specific provider key in the environment or .env file of your project.
- Clone the repo
$ git clone https://github.com/ALucek/rewrAIt.git
$ cd rewrAIt- Create a
.envfile for provider API keys
OPENAI_API_KEY=""
ANTHROPIC_API_KEY=""
GEMINI_API_KEY=""- Deploy with Docker
$ docker compose up --buildOnce the server is running, open your browser to the printed URL and start your conversation.
Contributions welcome, please feel free to submit a Pull Request.
Todo:
- Local model support
- Additional providers (HF, DeepSeek, etc)
MIT License - See LICENSE
