Microsoft has integrated agentic AI features into the Windows 11 taskbar, enabling users to easily summon an AI agent to perform tasks. To engage an agent, users type a trigger in the Ask Copilot search field, select an agent from the list, and specify the desired outcome. Once a task is initiated, the agent appears on the taskbar like any other app, and hovering over it reveals a compact status window showing progress and any requests for user input. Visual badges indicate when an agent requires input or when a task is complete. Outputs from all agents are routed into the Microsoft 365 Copilot app, allowing users to view results in one place, even if the agent is from a third party.

Microsoft explains that this feature uses Agent connectors and a Model Context Protocol (MCP), which provide a consistent method for agents to access applications, files, and services on the PC. Agents operate within an isolated Agent workspace that enforces policies, logs activity for auditing, and prevents disruption to the active desktop session. However, these AI agents have introduced new security vulnerabilities . Cross-prompt injection attacks are a significant concern, where attackers embed malicious directives in ordinary documents or interface elements, hijacking the agent's original instructions and causing it to perform unintended actions. Hopefully, Microsoft irons out these concerns before AI agents get a wider push. You can see these AI agents live at work below: