sunapp-ai/sun-to-spotify

Sun to Spotify turns any topic, question, or article feed into a spoken dialogue podcast, then delivers it straight into your Spotify library so you can listen from any device. You describe what you want in plain language, something like a five minute morning brief on the stock market or a twenty minute summary of a book, and the underlying Sun service handles the research, script writing, multi speaker dialogue, audio mixing, and publishing. This repository is not the audio engine itself. It is a command line tool called sun, plus a skill file that teaches Claude Code how to drive that tool. The skill is what most users are expected to install: you clone it into your Claude Code skills folder, either for one project or for every project, and from then on you can just ask Claude Code for a podcast on a topic and it will call the CLI, wait for the job to finish, and download the result. Setup starts with an install script that detects whichever Python package manager you already have, uv, pipx, or plain pip, and uses it to install the sun-cli package from PyPI, which needs Python 3.10 or newer. After installing, you log in once through a browser window, and the tool caches your session so you do not have to log in again. For automated or headless setups, an environment variable can hold an API token instead. Once logged in, the CLI can start a generation job, check its status, and fetch the finished manifest and audio files, either after the job completes or partway through while it is still running. Every command also supports a JSON output mode for scripting. Publishing the finished podcast into Spotify itself is handled by a separate tool from Spotify called save-to-spotify, which this project points to rather than reimplements. The project is aimed at people who already use Claude Code and want a quick way to turn a question or a piece of writing into something they can listen to, rather than at developers looking to build their own podcast pipeline from scratch.