🎵 Spotify CLI (PKCE Auth) — Discontinued

A command-line tool to authenticate with the Spotify Web API using the PKCE flow (no client secret required), fetch saved tracks, and (formerly) analyze audio features to create playlists based on user mood.

💡 Original Idea

The project was designed to:

• Analyze the audio features (like energy, danceability, valence, etc.) of a user's saved tracks
• Detect a rough estimate of the user’s current mood or musical taste
• Automatically generate a playlist tailored to that mood

🔧 Features

• Authenticates via PKCE flow (no need to store client secret)
• Stores and refreshes access/refresh tokens securely
• Fetches:
  • User profile data
  • Saved songs
  • (Previously) audio features for tracks

⚠️ Status: Discontinued

This project has been discontinued due to the deprecation of Spotify’s audio-features endpoint, which was essential for mood-based playlist generation.

The code still works for basic authentication and fetching user data, and may be useful as a reference for:

• Implementing PKCE with requests
• Manually handling access/refresh tokens
• Interacting with REST APIs via Python

📁 Structure

• auth(): Starts PKCE login flow
• storeTokens() / getTokens(): Handles saving/loading tokens
• refreshAccessToken(): Refreshes expired access token
• getAllSavedTracks(): Fetches all liked/saved tracks
• getAudioFeaturesWithIds(): (⚠️ Deprecated)

📌 Notes

• No external Spotify SDK used — built purely with requests and pkce
• Tokens are stored at: ~/.config/spotify-cli/tokens.txt
• Custom redirect URI: https://krishs-site.netlify.app/test

📚 Learnings

This project helped me understand:

• OAuth2 PKCE authentication flow
• Manual API calls without SDKs
• Token storage and refresh logic
• Spotify Web API structure

📦 Dependencies

• requests
• pkce
• webbrowser
• urllib.parse
• pathlib

🔚 Final Thoughts

While the mood-based playlist generator idea couldn’t be fully realized due to API limitations, this repo stands as a useful reference for future CLI and API-based experiments.

---

Made by Krishanth — 2025