#ai agent security

2 articles tagged

Cracked glass filter panel with a shell command passing through it, illustrating the GuardFall bypass

Engineering · Jul 16, 2026 · 16 min read

10 of 11 AI coding agents failed GuardFall: how to harden the shell guard in 2026

Adversa AI tested 11 open-source coding agents in June 2026 and beat the safety filter on 10 of them with shell tricks that are 30 years old. How the bypass works, which designs held, what to change now.

Manu Shukla

#ai agent security #coding agents #shell injection

3D glowing shield protecting an AI core orb and deflecting red intrusion rays on a dark background

AI Strategy · Jul 9, 2026 · 8 min read

AI agent security in 2026: stop prompt injection before agents act

Prompt injection is still OWASP's #1 LLM vulnerability in 2026, and agentic AI turns it into fraud and data theft. This guide shows how to contain it with least privilege, human approval and sandboxed tools.

Manu Shukla

#ai agent security #prompt injection #ai guardrails