AI/TLDR

Save to Spotify

Spotify's official CLI and agent skill for publishing generated audio into your own private shows

Agent Skills & PluginsOpen source
Updated
7 May 2026
Language
Go
License
Apache-2.0
$curl -fsSL https://saveto.spotify.com/install.sh | bash

What's new

7 May 2026

Spotify launched Personal Podcasts in beta alongside this CLI, letting audio generated by agents such as Claude Code, Codex or OpenClaw be saved to a personal Spotify library.

Overview

Save to Spotify is Spotify's own command-line tool for putting audio you made into your Spotify library, built for agents and automation. The workflow it is designed around is: an agent assembles the content, a text-to-speech engine narrates it, and this CLI publishes the result as an episode in one of your private shows, where it sits alongside everything else you listen to.

It deliberately does one job. The CLI saves audio; it does not generate speech. You bring the voice engine — Kokoro is the default because it is free, local, needs no API key and has no limits, with OpenAI TTS and ElevenLabs supported if you already have a key. `save-to-spotify tts status` reports what is available and `save-to-spotify tts add` registers any other engine. Supported upload formats are .mp3, .m4a, .wav and .ogg.

The object model is small: shows are containers, episodes are the files inside them, and a timeline can add chapter markers and in-player companion content to an episode. Beyond the binary, the installer also drops an agent skill into `.agents/skills/`, so Claude Code, Cursor, Codex and any other agent that reads that directory can invoke it directly — with published paths for the skills.sh installer, ClawHub/OpenClaw and the Claude Code plugin marketplace. Written in Go, Apache-2.0, with signed release archives; the Windows binaries are not code-signed yet, which the README calls out.

What it does

  • Uploads .mp3, .m4a, .wav and .ogg files as Spotify episodes, with title, summary, cover image and language
  • Shows, episodes and timeline commands — create and organise containers, check readiness, add chapter markers and companion content
  • Ships as an agent skill installable into Claude Code, Cursor, Codex, OpenClaw and any agent reading .agents/skills/
  • Pluggable TTS: local Kokoro by default, OpenAI TTS or ElevenLabs if you have a key, plus `tts add` for anything else
  • Browser and headless (`--no-browser`) OAuth login, with automatic token refresh and DPoP-bound tokens by default
  • Distributed via curl-bash, skills.sh, ClawHub, the Claude Code plugin marketplace, or a checksummed release archive
  • Single Go binary you can also build from source

Getting started

Install the binary, log in once, then upload. The installer also registers the agent skill unless you pass --no-skills.

Install the CLI

bashbash
curl -fsSL https://saveto.spotify.com/install.sh | bash

# binary only, skip the agent skill
curl -fsSL https://saveto.spotify.com/install.sh | bash -s -- --no-skills

Or install just the agent skill

bashbash
npx skills add spotify/save-to-spotify
# or, in OpenClaw
openclaw skills install @spotify/save-to-spotify

Authenticate

Tokens are stored in ~/.config/save-to-spotify/token.json and refreshed automatically. Use --no-browser over SSH.

bashbash
save-to-spotify auth login
save-to-spotify auth login --no-browser   # headless
save-to-spotify auth status

Check which voice engines are available

bashbash
save-to-spotify tts status

Upload an episode

Without --show-id or --new-show the CLI uses your most recent show, creating one if none exists.

bashbash
save-to-spotify upload ./briefing.mp3 \
  --title "Morning Briefing" \
  --summary "Weather and calendar for today" \
  --image ./cover.jpg

Organise and check readiness

bashbash
save-to-spotify shows create --title "Lecture Notes" --summary "Audio summaries of CS 101 lectures"
save-to-spotify episodes --show-id spotify:show:abc123
save-to-spotify episodes status <episode-id> --wait

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

When to use it

  • Have an agent build a daily briefing from your calendar and repos and publish it where you already listen
  • Turn an article, PDF or meeting transcript into a private audio episode
  • Keep generated audio organised in separate shows per topic instead of a folder of loose files
  • Publish from CI or a headless server using the --no-browser login flow

How Save to Spotify compares

Save to Spotify alongside other open-source agent skills & plugins tools AI/TLDR tracks, ranked by GitHub stars.

ToolStarsWhat it does
Superpowers★ 284kA composable skills plugin that installs a spec-first, TDD, subagent-driven development methodology into Claude Code, Codex, Cursor, Gemini CLI and other coding harnesses.
Skills for Real Engineers★ 258kMatt Pocock's everyday agent skills for coding agents, covering alignment grilling, planning, code review and research — small, composable and meant to be edited.
ECC★ 255kAn installable plugin that adds 68 agents, 286 skills, hooks, rules, memory and an agent-config security scanner to Claude Code, Codex and other coding harnesses.
Addy’s Agent Skills★ 93.1kA pack of production engineering skills for AI coding agents, with nine lifecycle slash commands — /spec, /plan, /build, /test, /review, /ship — installable into 70+ agents.
Humanizer★ 45.8kAn agent skill that rewrites AI-sounding prose against a list of 25 named writing patterns, grouped into five sections and applied without changing the claims in the text.
Claude for Financial Services★ 34.8kAnthropic reference agents, skills and MCP connectors for investment banking, equity research, private equity and wealth management, as Cowork plugins or Managed Agents.
Anthropic-Cybersecurity-Skills★ 32.5kA community-built library of 800+ agentskills.io-format cybersecurity skills for AI agents, each mapped to MITRE ATT&CK, NIST CSF 2.0, ATLAS, D3FEND, NIST AI RMF and MITRE F3.
Save to Spotify★ 395Spotify's official CLI and agent skill for publishing generated audio into your own private shows