prompt injection - Abrarqasim Blogs

AI Agent Security: The Tool Schema Problem Nobody Audits

Tool schemas quietly weaken your model's refusal behavior. Here is what a new agent-safety paper found and the checks I now run before shipping any…

Rayyan | August 7, 2026 | 8 min

AI coding agents will run a poisoned issue for you

AI coding agents read whatever you point them at, including issues written by strangers. How prompt injection reaches Cursor and Claude Code, and what actually…

Rayyan | July 25, 2026 | 8 min