Aider

Aider is a command-line tool that allows you to chat with GPT-4 about your code. It can make changes, improvements, and bug fixes to your code. Each change is automatically committed to git with a sensible commit message.

Features

• Chat with GPT about your code by specifying a set of source files to discuss.
• Request changes, improvements, and bug fixes to your code.
• Aider will apply the edits suggested by GPT.
• Aider will automatically commit each change to a git repo with a sensible commit message, providing safety, edit history and easy undo with normal git tools.
• Aider can make coordinated code changes across multiple source files.
• Live, colorized, human friendly output.
• Readline style chat input history, with autocompletion of tokens found in source files being discussed (via prompt_toolkit)
• Use Control-C to safely interrupt GPT if it isn't providing a useful response.
• Provide a multiline chat message by entering { alone on a line. End the message with } alone on a line.
• Aider will notice if asked to discuss files with uncommitted changes and offer to commit them before proceeding.

Installation

1. Install the package: pip install git+https://github.com/paul-gauthier/aider.git
2. Set up your OpenAI API key as an environment variable OPENAI_API_KEY or by including it in a .env file.

Usage

Run the Aider tool by executing the following command:

aider <file1> <file2> ...

Replace <file1>, <file2>, etc., with the paths to the source code files you want to work on.

You can also use additional command-line options to customize the behavior of the tool. For more information, run aider --help. Many defaults can be set with .env or environment variables, see the help output.

Limitations

Aider basically requires GPT-4 for the main chat functions. You can invoke it with aider -3 to try using gpt-3.5-turbo, but it will almost certainly fail to function correctly. GPT-3.5 is unable to follow directions and generate code changes in a stable, parsable format.

Aider also can only edit code that can fit in the context window. For GPT-4 that is 8k tokens. If you have access to gpt-4-32k, I would be curious to hear you experiences using it with aider.