Managed Agents

The Agents page is the central registry for every AI agent discovered across all integrated platforms. It provides an Agents Overview panel and an interactive Agent Network view, giving you a real-time snapshot of fleet size, risk distribution, platform coverage, and ownership gaps.

Select any agent from Registry > Managed Agents to open its detail page, which brings together the agent's risk posture, connected resources, credentials, associated users, access relationships, activity history, and AI usage in a single view.

Agents Overview

The Agents Overview panel provides a quick summary of your agent fleet's health through two sub-panels:

KPI Stat Cards

Four cards displaying key fleet health indicators including platform count, fleet size, criticality distribution, and ownership coverage.


Card Label	Description
Integrations · distinct types	Number of distinct AI platform integrations from which agents have been discovered (for example, n8n, MS Copilot Studio, Claude Code, and so on).
Total Agents · N flagged	Total number of agents in the registry and the count of agents that carry an active security flag.
Critical / High · % of active agents	Combined count and percentage of agents rated Critical or High risk among all active agents. This metric surfaces the scope of immediate security exposure.
Unassigned · % owned	Number of agents with no assigned owner and the percentage of agents that have a designated owner. Unowned agents represent a governance gap that should be remediated.

Risk Split Chart

A visual breakdown of agents across risk levels, with one-click filter buttons to instantly narrow the Agent Registry grid by severity.

Agent Network View

The Agent Network section displays an interactive visual map of agents grouped by platform. Use the Network and List toggle buttons to switch between views.

Network View: Displays an interactive node graph where each platform integration appears as a cluster node showing its agent count and risk summary (e.g., "N Risky | M Agents"). Hover or click a platform node to expand and view individual agent nodes connected to it.
Canvas Controls: Use the toolbar to search agents by name, filter by status (Active/Inactive) or platform, view the live agent count, toggle the colour-coded legend, and adjust zoom level.

Agent Registry (List View)

Click the List toggle to switch to the Agent Registry a searchable, sortable grid of all registered agents. An Ask AI button in the section header opens a conversational AI assistant scoped to agent data.

Note: The Agent Registry grid default sort order is by Risk Level.


Column	Description
Agent Name	Display name of the agent. When the agent belongs to a workflow, the name is rendered as "`Workflow Name - Agent Name`". Long names are truncated with a tooltip showing the full value on hover. Clicking the row opens the agent detail page.
Platform	The integration type from which the agent was discovered (for example, n8n, MS Copilot Studio, Claude Code). Each platform is shown with its logo icon followed by the platform display name.
Risk Level	Computed risk rating for the agent based on active posture issues and access configurations. Rendered as a coloured severity badge. Hovering the badge opens a Risk Breakdown popover showing the contributing factors.
Owner	The user or service account assigned as the agent owner. Displayed as a truncated string with a full tooltip on hover. Shows a dash (—) when no owner has been assigned.
Status	Operational state of the agent. Active (green pill with check icon) indicates the agent is reachable and running. Inactive (grey pill with minus icon) indicates the agent is not responding or has been deregistered.
First Detected On	Date when the agent was first registered in Agent Identity Security, rendered as `MMM D, YYYY` (for example, "Jun 3, 2026"). Hovering the cell shows the full date and time in a tooltip.