tl;dr — RedditDownloader (RMD) was a solid tool for archiving Reddit media. In July 2023, Reddit killed Pushshift and locked down direct API access. The maintainer archived the project and explicitly invited forks. I replaced the dead data layer with the Sylvia API (sylvia-api.com) and got subreddit scanning working again. The fork is at github.com/c1nn3r/RedditDownloader.
The problem
RMD was well built. The repo has about 1,200 stars. It shipped a real filter engine, a local web UI, image-hash duplicate detection, a sqlite state database, and resume support so a large archive run can stop and start without re-downloading.
It scanned many places for media: subreddits, user posts, saved lists, upvoted lists, and multireddits. It extracted links from post URLs, from self-post text, and from comments. It handled most of the hosts you meet on Reddit: i.redd.it, v.redd.it, preview.redd.it, imgur, gfycat, and tumblr, with yt-dlp in the corner for video. It ran headless or as a server with a web UI. For the data-hoarding crowd — the people who believe the internet is ephemeral and the stuff they like will vanish — it was the tool.
Then Reddit changed the rules. In July 2023, Reddit destroyed Pushshift, which RMD used for historical data. Reddit also restricted direct API access and made OAuth app registration slower and stricter. The maintainer, ShadowMoose, wrote a blunt shutdown note. He said that the changes made it impossible for RMD to operate. He said that he no longer wanted to build on Reddit's platform at all. Then he archived the repo. But he left the door open:
