The CISO's Guide to Agentic Security and Operations
How security leaders run an agent estate that is already doing the business's work, from the first inventory to a program that scales.
Most security leaders no longer need convincing that agents are running in their organization. Agents have moved out of pilots and into work the business relies on every day, and the question now is a practical one: how do you run security for an agent estate that grows faster than any review cycle, spans endpoint, cloud, code and the browser, and whose ownership is still being argued over?
This second edition answers it from the CISO’s seat. It draws on our work with organizations running agents at real scale, on conversations with more than 60 of them about who owns agent security and operations, and on what we have found inside their estates.
Included:
- The estate you are already running, and the four routes agents arrive by: platforms the business bought, developers’ coding agents, workflows built by business teams, and tools employees adopted themselves.
- Why agents break the security playbook, and what we are finding inside real agent estates.
- Seeing every agent, wherever it runs, and controlling what agents can reach.
- Enforcement that keeps the work running, rolled out in rings rather than blocks that get routed around.
- Risk and value from one record, so the evidence of a threat and the evidence of spend come from the same trace.
- The ownership problem, and a plan for what to do in the next two quarters.
