Daily Defender News
Aboutmerill.net
Microsoft Defender for Endpoint
Endpoint protection

Local Agent Discovery Overview

9 October 2026Feature update+14 / −15 lines

In brief

The documentation now states that Defender lists a local AI agent after observing activity on an onboarded device; installation alone does not guarantee discovery. It also clarifies that the inventory is not a complete software inventory and references the support matrix for coverage details.

What Defender admins need to know

Treat the agent inventory as activity-based security context, not a complete list of installed agents. Use Software inventory for installed software details and consult the support matrix for coverage.

Summaries are generated from the documentation change itself.

Documentation change

View full page diffView on Microsoft Learn

The comparison below shows only the changed extract. Use the full-page view for complete context.

Local AI agents run with user-level permissions and can access files, tools, and services on the devices where they operate. Without visibility into which agents are running and what they can reach, security teams can't assess exposure, enforce governance, or respond to agent-related incidents.

Microsoft Defender automatically discovers supportedcan discover local AI agents and their configured MCP server configurationsservers after it observes agent activity on onboarded devices, then surfaces them in the Microsoft Defender portal. This gives security teams a centralized view of AI agent presence acrossagents used in the organization.

Screenshot showing the local AI agents inventory in the Microsoft Defender portal with discovered agents listed.

This article explainsLearn how local AI agent discovery works, lists supportedreview coverage for agents and MCP server configurations, and describes how to view discovered agents in the Microsoft Defender portal.

Tip

Local AI agent discovery on endpoints

Defender automatically detects supportedlists a local AI agent after it observes agent activity on an onboarded device. Installing an agent doesn't by itself guarantee that the agent appears in the local AI agent inventory. An installed agent might not appear until Defender observes agent activity.

Local AI agent discovery isn't a complete inventory of installed software. It provides security context for agents that Defender observes on endpoints, including the device and MCP server configurationsaccount associated with the activity. To review known software installed on onboarded devices. devices, use Software inventory in Microsoft Defender Vulnerability Management.

When Defender identifies a supported local AI agent, the agent is displayed as a discoverable asset in the Microsoft Defender portal with visibility into:

  • Local AI agent inventory: A centralized view of discovered local AI agents with device and user associations and discovery metadata.
  • Exposure map: Visual relationships between local AI agents, devices, identities, and the resources those identities can access, to help assess potential impact.

Defender defines an agent as a combination of a user, a device, and an agent type. For example, if Claude Code runs in 15 different project folders on the same device for the same user, it appears as a single agent entry in the inventory.

Defender discovers supported local AI agents on Windows and macOS endpoints. This includes agents that run from the command line, desktop apps, agentic IDEs, VS Code extensions, and Claw-based local agent implementations. When supported, Microsoft Defender can also discoversdiscover local and remote MCP server configurations associated with these agents, including localagents.

For coverage by agent, operating system, and remote MCP server configurations.configuration, see Microsoft Defender for Endpoint AI agent support matrix.

Examples of supported local AI agents include:

  • CLI agents: Claude Code, Codex CLI, Gemini CLI, GitHub Copilot CLI, Junie CLI, Kiro CLI, OpenCode, Warp, Antigravity CLI
  • Desktop apps: ChatGPT Desktop, Claude Desktop, Codex Desktop, GitHub Copilot app, Goose Desktop, Hermes Agent, Microsoft Copilot app, Ollama Desktop, Perplexity Desktop, Poe Desktop
  • Agentic IDEs: Cursor, Devin Desktop (formerly Windsurf), Kiro IDE, Antigravity IDE
  • VS Code extensions: Claude Code, Cline, Codex, Gemini Code Assist, GitHub Copilot, Roo Code
  • Claw-based agents: OpenClaw, Clawpilot, QClaw, Claw/Nanobot, ZeroClaw

For supported agents, Defender associates discovered MCP server configurations with the agent. The available configuration details can include the server name, type, endpoint, and the command used to start a local MCP server. An agent can have multiple configured MCP servers. To query these details, see Review the MCP servers and tools that local AI agents use.

To learn how to discover and view local AI agents, see Discover local AI agents.

Broader AI security capabilities

Microsoft Defender's discovery capabilities are part of a comprehensive AI security approach. Microsoft Defender provides other capabilities acrossin your organization's AI ecosystem:

  • Discover cloud and platform agents: Find agents built with Microsoft Copilot Studio, Microsoft Foundry, Amazon Web Services (AWS) Bedrock, and Google Cloud Platform (GCP) Vertex AI.
  • Detect and investigate threats: Correlate alerts and investigate suspicious agent behavior acrossin your security infrastructure.

For details on these capabilities and how to apply them, see Protect AI assets from emerging threats and vulnerabilities using Microsoft Defender. \ No newline at end of file