Discover your agent network with the map

  • Release version: Zurich
  • Updated May 2, 2026
  • 2 minutes to read

    • Use the agent map to visualize your entire agentic ecosystem, including ServiceNow and third-party AI assets.

    About this task

    The agent map shows a holistic view of the relationships between your AI providers, AI models, MCP servers, AI agents, agentic workflows, and tools. You can use the map to review these relationships and get details about the AI assets in your enterprise.

    Below the map, all of your managed agentic AI assets are listed.


    Agent map with managed agentic assets list below it.

    Before you begin

    Role required: AI steward [sn_ai_governance_ai_steward]

    Configure Veza access intelligence to show risk score and other information for each AI asset in the map. For more information, see Configure Veza access intelligence in the agent map.

    Procedure

    1. Navigate to AI Control Tower > Govern > Security.
    2. On the Overview tab, select the agent map to go to the agent map details.
    3. On the agent map, select agentic assets to view details such as type and domain.
      • To find an asset in the map, enter a partial or entire name in the search box in the map.
      • To identify asset symbols, select Legend.
      • To zoom and refresh the map, use the controls in the map.
      Table 1. Map node - Access intelligence tab
      Node Details
      Agent and AI Model
      Agent risk
      The severity of risk as defined by the Veza tenant for your instance. Risks can be classified Critical, High, Medium, and Low. For more information, see Risk level tiers in Veza documentation.
      Risk score
      A score representing the risk of the agent or model. Risk scores in Veza are calculated for each AI asset based on the number of associated risks and their risk levels. The scoring system weighs both the severity of risks and their cumulative impact using an algorithm designed to provide comparable scores across all AI assets in the agent map. For more information, see How risk scoring works in Veza documentation.
      Nonhuman identity
      The user of the AI agent or model.
      Human owner
      The creator of the AI agent or model.
      Table 2. Map node - Details tab
      Node Details
      Agentic workflow
      Description
      A brief description of the workflow.
      Type
      This value is always Agentic Workflow.
      Domain
      The domain scope of the workflow.
      Execution mode
      How the workflow is executed (autonomous or manual).
      Created by
      The creator of the workflow.
      Agent
      Description
      A brief description of the AI agent.
      Type
      This value is always Agent.
      Domain
      The domain scope of the AI agent.
      Created by
      The creator of the AI agent.
      Source system
      The external AI system.
      Provider
      The AI provider.
      Note:
      Only agents with access issues display the Access issues details.
      Access issues
      The resources and operations causing the access issues.
      Resource
      The resource the agent is having issues accessing.
      Operation
      The operation the agent attempted.
      Count
      The number of times of the agent encountered the issue.
      AI model
      Type
      This value is always AI Model.
      MCP
      Type
      This value is always MCP.
      Provider
      Type
      This value is always Provider.
      Tool
      Description
      The description of the tool.
      Type
      This value is always Tool.
      Domain
      The domain scope of the tool.
      Execution mode
      The execution mode of the tool (supervised or autonomous).
      Tool type
      The type of the tool. For example, action.
      Target record
      The record the tool modifies.
      Created by
      The creator of the tool.
      Note:
      Select Open tool record to view additional tool details.