This Emacs package provides direct integration with ChatGPT, enabling users to send text regions or entire buffers for processing and receive formatted responses. It targets Emacs users seeking to leverage large language models for tasks like code generation, debugging, and documentation directly within their editor.

How It Works

The package utilizes the llm-workflow-engine library, offering a flexible backend for interacting with OpenAI's models. This approach allows for customization of prompts, models, and system messages. Responses are rendered with syntax highlighting for code, enhancing readability and usability.

Quick Start & Requirements

• Installation: Install via Straight, Quelpa, Spacemacs, or Doom Emacs.
• Prerequisites: OpenAI API key (set as OPENAI_API_KEY environment variable), Python 3.x, pip, and llm-workflow-engine (installed via pip install git+https://github.com/llm-workflow-engine/llm-workflow-engine).
• Usage: Bind chatgpt-query to a key (e.g., C-c q). Select text to query specific regions.
• Configuration: Customize chat behavior and code query shortcuts via Emacs Lisp variables.
• Links: llm-workflow-engine, OpenAI API Keys

Highlighted Details

• Seamless integration into Emacs workflows.
• Syntax highlighting for returned code snippets.
• Customizable prompt shortcuts for common tasks (e.g., debugging, documentation).
• Built on the extensible llm-workflow-engine.

Maintenance & Community

• The project is actively used daily by the maintainer and considered stable.
• Issues can be submitted via GitHub.

Licensing & Compatibility

• License: Not explicitly stated in the README.
• Compatibility: Designed for Emacs. Commercial use depends on the underlying OpenAI API terms and the license of llm-workflow-engine.

Limitations & Caveats

Multiline input is not yet supported. The project relies on external Python dependencies and an OpenAI API key.