Classify YouTube songs with Azure OpenAI, review the moods, then sync private managed playlists.
Effective date: May 15, 2026
VibeShelf uses YouTube API Services to let a signed-in user organize their own YouTube playlists. The app is also subject to the Google Privacy Policy.
After you sign in with Google, the app may access playlist IDs, playlist titles, playlist descriptions, playlist item counts, playlist privacy status, playlist item IDs, video IDs, video titles, video descriptions, owner channel titles, source playlist and position metadata, and OAuth token data.
The app uses this information for playlist selection, song classification, preview generation, user-confirmed private playlist sync, classification caching, and session continuity. The app only uses YouTube data to provide the playlist organization features requested by the signed-in user.
The app shares data with YouTube and Google APIs for authorization and playlist operations. It sends the provided video metadata to the configured Azure OpenAI endpoint for mood classification. The app does not sell YouTube data, share it with advertising networks, or make it available to unrelated third parties.
The app stores the Google OAuth token payload in an encrypted HttpOnly browser cookie. Preview and apply state is carried in encrypted browser-submitted state. Local runs and classification cache entries may be stored in SQLite. On Vercel deployments, the deployment filesystem is ephemeral, so durable local run history and durable local classification cache are not kept by the deployment.
The app uses a session cookie, a Google token cookie, and encrypted app state for authentication and workflow continuity. The app does not use advertising cookies.
You can revoke this app's access to your Google data at the Google security settings page. You can also use the app's Disconnect YouTube action to clear the local Google token cookie in this app.
Privacy questions or complaints can be sent to ayush@scorptech.co.