X Strict Reply Filter

Hide likely spam replies on X/Twitter status pages using local structural scoring.

X Strict Reply Filter hides likely spam or NSFW-style replies on X/Twitter status pages. It is designed for reply sections that attract repeated low-value bait, including adult-service prompts, local-resource funnels, compact emoji/code replies, emoji-broken English filler, and profile-name bait paired with generic content. The filter runs entirely in your browser. It does not upload, send, sell, or store reply content. How it works: Scans replies below the main post on X/Twitter status pages. Extracts visible reply text, including emoji image alt text when it affects obfuscated patterns. Normalizes common Unicode, emoji, and spelling obfuscation. Applies a local score model built around structural spam signals. Hides replies only when the score reaches a high-confidence threshold. The filter avoids broad keyword-only blocking. Normal replies with ordinary language or trailing emoji are intended to stay visible. Default behavior: The main post is not filtered. Matched replies are hidden directly. A small floating counter shows how many replies were hidden on the current thread. There is no popup, account, remote configuration, analytics, or cloud service.