When cloud AI becomes a risk to your business
Why relying on cloud AI providers can actually ruin your business and what are the alternatives
PwnYourAgent Blog
Plain-language security writing for founders and teams building AI products.
Hardening an AI Agent: Least Privilege in Practice
A practical experiment in agent hardening: using least privilege, tool restrictions and security policies to limit an AI agent's access to Linux.
Indirect Prompt Injections in log files
OWASP LLM01:2025 Prompt Injection
When AI agents see your data, who watches them?
An AI agent can touch your sensitive data. Learn what to record, so you can trace every step after the fact.