Spotify unveils beta CLI tool enabling AI agents to generate private podcasts

Spotify has released a beta command-line interface tool designed to allow AI coding agents to generate and upload personal podcasts directly to a user's private library. The initiative marks a strategic shift for the streaming platform, aiming to consolidate AI-created audio content within the environment where users already consume their media. By integrating with agents such as OpenAI's Codex and Anthropic's Claude Code, the feature allows for the creation of personalised content like daily briefings and research summaries.

The functionality is specifically engineered for AI coding agents, with explicit support noted for OpenClaw, Claude Code, and OpenAI Codex. Users can instruct these agents to create content by providing a description and appending a simple prompt suffix to trigger the upload process. This integration streamlines the workflow, ensuring that the creation and consumption of audio happen within a single ecosystem.

Access to the feature requires navigating to Spotify's GitHub repository and following the provided setup instructions. Once the software is installed, a browser-based authentication step is necessary to link the user's Spotify account with the CLI. This authentication process ensures that the generated audio is saved securely within the user's specific library rather than being made publicly available on the broader platform.

Crucially, the generated audio remains strictly private and accessible only to the individual user within their personal Spotify library. While Spotify cites user demand as the motivation for the release, reports suggest the tool may have originated from an engineer's personal use case before being formalised. In any case, if you want to try generating your own podcasts, head to Spotify's GitHub page and follow the provided instructions.

After setup is complete and the user has entered their login credentials, they can describe the podcast they want to hear and ask the agent they are using to save it to Spotify. From there, either click the provided link or find the podcast in their Spotify library. Any audio generated this way will only be accessible to the individual user, maintaining a clear boundary between public broadcasting and private utility.

The tool is currently described as a beta command-line interface tool, implying it may be experimental or subject to change. The scope of "AI agents" supported is currently limited to specific coding agents rather than a broad range of generative AI models. Furthermore, the integration relies on the specific capabilities of third-party AI agents to generate audio and execute the upload command, meaning the quality and reliability depend on the agent's performance.