Agent Operations
Agent Threat Model
Also: deployment threat model
A short written map of what an agent deployment protects, what can influence the agent, what it can do and how you would notice trouble.
In more detail
An agent threat model starts from the deployment (who uses which agents, where, as which identity, with which tools) and lists assets, inputs, actions and trust boundaries. It ranks risks by impact and reversibility and names a control you can verify for each top risk. Unlike a model of ordinary software, it records who can write each text input the agent reads, because that text can steer the agent.