Agent Craft London 2026

If we want agents to do things for us, they need permission to do things for us. This is the tightrope we walk when building and deploying agents today. What are the principles and practices that are emerging? And what are the security basics you should keep no matter what?

Our panel of industry experts will share what's working for them and the simple steps you can take to make sure you stay safe and sane.

Globally recognised innovator in open source, cybersecurity and artificial intelligence, with a distinguished career dedicated to building and fortifying complex software ecosystems.

As the creator of Sigstore, the industry transforming technology used today by Google, Nvidia, GitHub, and much of the open source world, Luke is now focused on bringing the same innovation to the AI Agent space.

Luke's more recent project nono.sh provides runtime safety infrastructure for AI agents. Kernel-enforced isolation, supply-chain security, immutable auditing, atomic rollbacks, credential management, and more.

Our panel of industry experts will share what's working for them and the simple steps you can take to make sure you stay safe and sane.

Globally recognised innovator in open source, cybersecurity and artificial intelligence, with a distinguished career dedicated to building and fortifying complex software ecosystems.

Luke Hinds

Sessions

Panel: Safe and Sane - Securing AI Agents

About Luke Hinds

Luke Hinds

Sessions

Panel: Safe and Sane - Securing AI Agents

About Luke Hinds