By Vero SecurityBuilt and funded by Vero Security, the AI Security Engineering platform.Read the charter

Hidden instruction tampering

Attackers hide instructions in rule files with invisible Unicode or bury them in content an agent will read. Scan instruction files for hidden characters, and treat any unexplained change as a possible compromise.

Read in this order

  1. ArticleNew Vulnerability in GitHub Copilot and Cursor: How Hackers Can Weaponize Code Agents (Pillar Security)Start here: a working attack with hidden text in rule files.
  2. OfficialGitHub now provides a warning about hidden Unicode text (GitHub Changelog)The platform warning that surfaces hidden Unicode in reviews.
  3. ArticleASCII Smuggler Tool: Crafting Invisible Text and Decoding Hidden Codes (Embrace The Red)How to encode and decode invisible text, to test your scanner.

Save your progress

We email you a sign-in link. No password.

By signing in you agree to the terms. The privacy policy explains how we use your details.

Join the AI security engineering community

Free. Roadmap updates, chapters near you, and first word when credentials open.

We’ll email you community updates. Unsubscribe anytime. By sending this you agree to the terms. The privacy policy explains how we use your details.