askitty is a simple CLI tool designed to interact with generative models like large language models (LLMs) and image generation APIs. The tool supports various features, including chat history, new chat session commands, global and individual system prompts, and more.

Check out this demo video

• LLM API: Supports any LLM that follows the OpenAI/v1 spec.
• Image Generation API: Supports any image generation API that follows the OpenAI/v1 spec.
• Chat History: View and manage chat history.
• Sessions: Start new sessions, view existing sessions, and delete sessions.
• System Prompts: Set and manage global and session-specific prompts.
• Configuration: Set and view global configuration.
• Error Handling: Robust error handling for a seamless experience.
• Markdown Preview: Preview messages in Markdown format.

To get started with askitty, you can use the following command to download and install the latest release:

curl -sSL https://askitty.rdsx.dev/install.sh | bash

Once installed, you can use askitty from the command line with the following syntax:

askitty [FLAG] [MESSAGE]

• -h, --help Display help message
• -v, --version Display version
• -m, --message Message to send to the model
• -i, --imagine Generate image from text
• -r, --repl Start a REPL
• -n, --new Start a new session
• -s, --sessions View all sessions
• -s <session_id> View a specific session
• -d <session_id> Delete a specific session
• -c, --clear Clear all sessions
• -p, --prompt Set global system prompt
• -p <session_id> Set a specific session prompt
• -ps, --prompt-show Show global system prompt
• -pc, --prompt-clear Clear global system prompt
• -pd, --prompt-delete <session_id> Delete a specific session prompt
• -vc, --view-config View global configuration
• -sc, --set-config Set global configuration

askitty -m "Hello, how are you?"

askitty -i "A beautiful sunset over the mountains"

askitty -n

askitty -s

askitty -s <session_id>

askitty -p "You are a helpful assistant."

Contributions are welcome! Please fork the repository and submit a pull request with your changes.

This project is licensed under the MIT License. See the LICENSE file for details.