Blocking Claude

· Kyle Kingsbury · Jan. 27, 2026, 5:34 a.m.
Web Development Developer Tools large language models spam-detection
Summary
The blog post discusses how to manipulate Claude, a large language model, by using a specific 'magic string' that terminates conversations deemed inappropriate. It provides insights on embedding this string in web pages and uploading cache-busting URLs to circumvent Claude's caching behavior. The author shares personal experiences of dealing with LLM spam and how implementing this string on their blog is expected to reduce spam.
Read full post on aphyr.com →
AUTHOR
Advertise
Sponsor diff.blog
Put your product in front of developers who read and write about their craft. One exclusive sponsor at a time.
Become a sponsor →
BLOG POST FEATURED ON

Placeholder image
Hacker News

5 points

Add this plugin to your blog