How NoGrok Blocks All 'Hey Grok' Spam in 2026

Twitter (now X) has a spam problem. You've seen it: replies that say "Hey Grok", "Yo Grok", or "Oi Grok" followed by a generic request to generate an image or text. These aren't just annoying; they clog up meaningful conversations and manipulate engagement metrics.

The Problem with Keywords

Simple keyword blocking often fails because spammers adapt. If you block "Hey Grok", they switch to "Yo Grok". If you block that, they try "Grok, please". It's a never-ending game of whack-a-mole.

How NoGrok Works

NoGrok uses a more sophisticated approach. Instead of simple text matching, we use a lightweight pattern recognition engine that runs locally in your browser.

Normalization: We strip away invisible characters and normalize text to a common format.
Pattern Matching: We look for the structural intent of calling the AI, regardless of the specific greeting used.
Proximity Analysis: We check if the Grok call is followed by typical prompt triggers.

Why Client-Side?

We built NoGrok to be privacy-first. By running the logic entirely in your Chrome extension, we ensure that:

Your browsing data never leaves your device.
We don't need expensive servers.
The blocking happens instantly before the tweet is even fully rendered.