The Execution Layer

Archives
Log in
Subscribe
July 22, 2026

Welcome!

Welcome — agentic security, decoded

One brief, every week — new attacks, defensive patterns, and tooling for securing AI agents.
Tom Mooney.

// Welcome

One brief.
Every week.

You’re in. AI agents stopped talking and started acting — and this newsletter is where you learn to secure them. Every week, in one brief:

01 New threats & incidents — what actually happened, and why it matters
02 Defensive patterns & checklists you can lift into a design review
03 Tooling and research worth your time — no filler

Start with the evidence: we ran a structured prompt-injection benchmark against the 20 most-used models — 12 of 20 were breached undefended, and a few sentences of defensive prompt cut attack success to zero.

Read the benchmark →

The book — launching July 15, 2026

Agentic AI Security — the playbook, in print.

Join the waitlist and get the first three chapters (draft) free as a PDF today.

Join the waitlist →

Glad you’re here. If something in your environment keeps you up at night, hit reply — I read every response.

Tom Mooney

— Tom

Author of Agentic AI Security

Tom Mooney — Securing the execution layer

Only what you asked for. No spam. Unsubscribe anytime.

Don't miss what's next. Subscribe to The Execution Layer:
← Newer The agent is now on both sides of the attack
Powered by Buttondown, the easiest way to start and grow your newsletter.