feat: personal favorites list (watch later / bookmark)

[manuelgallego94]

Description

I'm often frustrated that there's no way to mark content I want to watch later without actually requesting it for download. The Watchlist feature is tied to Plex and requires a Plex account, so it doesn't work for local users or for content you just want to track personally.

I'd like to propose a personal favorites list — a lightweight way to bookmark movies and series directly in Seerr, independent of Plex and independent of the request system.

Desired Behavior

• A heart button on movie and series detail pages to add/remove from favorites
• A "Favorites" section in the sidebar listing all bookmarked content
• Each user has their own independent favorites list — not shared, not requiring Plex
• The feature is purely internal — no Radarr/Sonarr integration, no download triggered

Additional Context

The key difference from Watchlist: Watchlist syncs with Plex and relies on Plex's ratingKey. Favorites would be a standalone internal feature using only tmdbId + mediaType + userId, simpler and with no Plex dependency at all.

I have a working implementation in my fork. I previously submitted a PR but it was closed because the changes were too large and needed more careful review. I'd like to discuss the design approach here first before resubmitting as smaller, more focused PRs.

Screenshots below showing the feature working in my fork:

Search Existing Issues

• Yes, I have searched existing issues.

Code of Conduct

• I agree to follow Seerr's Code of Conduct

[fallenbagel]

cc: @seerr-team/seerr-core

[gauthier-th]

Is this AI-generated content?

[manuelgallego94]

All these ideas are mine, and I also did the testing manually, reverse-engineering using Chrome's debugger. I simply fed all my disorganized ideas to the AI ​​so it could write my lines more clearly.

[manuelgallego94]

Is this AI-generated content?

Hi, is something wrong? My intention is simply to help and contribute ideas.

[gauthier-th]

Is this AI-generated content?

Hi, is something wrong? My intention is simply to help and contribute ideas.

I am not questioning your good intentions. However, your last PR contained 35,000 changes for a single feature. It was quite simply AI slop.

If you cannot put in the effort to articulate your own thoughts rather than submitting what is likely hallucinated or poor-quality content, I see no reason why we should spend time reviewing it.

[manuelgallego94]

You’re right, but I could create a new branch from zero, built by me without AI, using the new approach I outlined above. I’d like to contribute and rise to the challenge.