Launching today
Remedy
AI feed filter: type what you don't want to see
1 follower
AI feed filter: type what you don't want to see
1 follower
Remedy is an AI feed filter for Chrome on desktop. Type what you don't want to see in plain words, like "no rage bait" or "nothing about the finale", and it removes matching posts, comments, images and videos on X, YouTube, Reddit, Instagram, Facebook, Threads, TikTok and any site you add. Unlike muted words, it filters by meaning, not keywords. Removed posts just disappear, and a counter on the toolbar icon shows how many. Posts are checked by AI models with zero data retention.
Free Options
Launch Team
Atlassian LoomTalk it through in Loom. Ship with AI
Promoted
Hi Product Hunt! I'm Arman, and I built Remedy.
Muted words match spellings, not meaning. Mute a show's name and the reaction posts, the screenshots and the "can't believe she picked him" still get through. Remedy lets you say what you're done seeing, in plain words.
You write a rule the way you'd say it ("no rage bait", "nothing about the finale", "no golf, my dad won't stop"), and as you scroll X, YouTube, Reddit, Instagram, Facebook, Threads, TikTok or any news site you add, an AI checks each post, comment, image and video frame against it. Matches just disappear, and the toolbar icon counts them.
A few honest notes:
• It's AI, so it will sometimes miss something or remove something harmless. The popup lists what it took out of the page you're on, and you can reword a rule or pause Remedy for an hour anytime.
• Posts are checked by AI models with zero data retention. Remedy keeps a fingerprint of each post and its verdict, never the post itself.
• Desktop Chrome only, for now.
• Lite ($6.99/mo) checks text. Plus ($14.99/mo) adds images and video, and it's free for 5 days (card required, cancel anytime from the extension).
What would your first rule be? Post it below and I'll tell you how Remedy handles it.