Connecting Discovery Data Sources
Last Modified: 2026-11-19

Connecting Discovery Data Sources

Agent ID discovers agents, non-human identities, and shadow assets by reading from connected data sources, managed from the Connectors page. Connect at least one source so Discover has data to work with. Each connector is fully independent—for example, Microsoft Entra ID and Microsoft Defender are separate connectors even though both come from Microsoft, and each can be added, edited, disabled, or removed on its own.

The Connectors page gives you the following:

  • Health at a glance. Summary cards show how many connectors are healthy, need attention, or are disconnected.
  • A configurable scan schedule. Choose how often each connector refreshes when you set it up, and change it later.
  • Sync on demand. Trigger an immediate refresh from any connector's menu instead of waiting for the next scheduled scan.
  • Disable without losing data. Disabling a connector stops new data coming in but keeps everything already discovered, so you can turn it back on later. Remove connection is a separate, deliberate action.
  • Visibility into granted access. Open a connector to see its sign-in method (secret or certificate), the exact permissions it was granted, when it last fetched data, and what that fetch found.

The page opens with a Connection summary showing three counts—Healthy, Needs attention, and Disconnected—followed by a Connections table that has the following columns:

  • Connection (name and provider)
  • Type (Identity, Endpoint, Gateway, or Cloud)
  • Provider
  • Scope (the tenant domain, account, or org the connection reads from)
  • Status (Connected, Syncing, Needs attention, or Disabled)
  • Last sync
  • Issues (this column names the specific problem, such as a revoked access token or an IAM permission gap, for any connection needing attention)

You can filter by Status, Type, Provider, or Scope. You can add a connector from Add source on the top right.

Use the Actions menu at the end of a connection's row to View details, Sync now, Test connection, Edit connection, Disable, or Remove connection.

The following sources are available:

SourceCategoryWhat It Contributes
Microsoft Entra IDIdentity ProvidersUsers, groups, service principals, and app registrations
OktaIdentity ProvidersUsers, groups, and OAuth/OIDC application registrations
AWSCloudIAM users, roles, groups, policies, and Bedrock agents
CrowdStrike FalconSecurityAI agents, automation frameworks, and MCP servers observed on managed endpoints
Microsoft DefenderSecurityEndpoint and device posture signals
MCP GatewayGatewayResources and tools registered behind a gateway, feeding Discover-side inventory

Follow this procedure to add a connector.

Procedure 

  1. On the Connectors page, click Add source.
  2. In the Add a source dialog, locate the source you want to add. Filter the list by category (Identity providers, Cloud, Security, Gateway, or Developer), or use the search box. Click Connect on the source's card.
  3. Enter a Connection name (for example Production AWS).
  4. Fill the connection fields required for that source. See the connector-specific procedures for the exact fields each source expects.
  5. If prompted, select an Authentication Method (Secret or Certificate), then enter the corresponding credential.
  6. Click Test connection to confirm Agent ID can reach the source.
  7. Click Save and start discovery.

Results 

The new connector appears in the connections table with a status of Syncing while its first scan or search job runs automatically, then Connected once discovery completes. Identities found by this connector begin appearing in the Agents/NHI registry.

Note:  Credentials are stored encrypted. Rotate them from the source provider's own administration console, then update the connection in Agent ID.

Next Steps: