Integrated Azure AI Foundry data

After the synchronization, the assets are available in the specified domain.

Warning Do not move the assets to another domain. Doing so may lead to errors during future synchronizations.

By default, the assets are shown in a plain list, but you can enable a multi-path hierarchy to show it in a tree structure.

Newly ingested assets are assigned the first status listed in your Operating Model statuses, and existing assets keep their assigned status.

Note For more information about AI Model Version asset statuses, go to Configure the lifecycle stages of AI Governance assets.

Synchronized metadata per asset type

This table shows the metadata for the Azure AI Foundry integration asset types. If you do not see any of the listed synchronized metadata, you can add characteristics to the layout on the asset type page.

Azure AI Foundry

Asset type Synchronized metadata Public ID
AI Base Model Creation Date in Source CreationDateInSource
Retirement Date in Source RetirementDateInSource
Azure AI Foundry Model Version Description from source system DescriptionFromSourceSystem
Version Version
Supported Input Modalities SupportedInputModalities
Supported Output Modalities SupportedOutputModalities
Supported Model Customizations SupportedModelCustomizations
Any custom metrics defined via the configuration. If you use custom metrics, ensure that you add them to the assignment and layout on the asset type page.  
AI Model Deployment Initiating User in Source InitiatingUserInSource
Creation Date in Source CreationDateInSource
Modification Date in Source ModificationDateInSource
Implemented Content Filtering ImplementedContentFiltering
Compute Configuration ComputeConfiguration
Azure AI Foundry Agent Description from source system DescriptionFromSourceSystem
Instructions Instructions
Document creation date DocumentCreationDate
Creation Date in Source CreationDateInSource

Any custom metrics defined via the configuration. If you use custom metrics, ensure that you add them to the assignment and layout on the asset type page.

 
AI Agent uses / is used by AI Model Version AIAgentUsesAIModel
AI Agent has version / is version of AI Agent Version AgentHasVersion
AI Agent has agent config stored in / contains config files for Storage Container AIAgentContainsFileContainer
AI Agent Version Version Version
AI Agent has version / is version of AI Agent Version AgentHasVersion
AI Agent Version can call / can be called by AI Agent Tool AgentVersionCallsTool
AI Agent Tool AI Agent Version can call / can be called by AI Agent Tool AgentVersionCallsTool
Azure AI Foundry Project Project ID ProjectId
Location Location
URL Url
AI Endpoint Access Method AccessMethod
Access Instructions AccessInstructions
Traffic Split TrafficSplit
AI Monitor Data Drift Detection Enabled DataDriftDetection
Prediction Drift Detection Enabled PredictionDriftDetection
Schedule Schedule
Alert Configuration AlertConfiguration
URL Url
File Description Description
URL Url
Document size DocumentSize
Storage Container    

Microsoft Copilot Studio

Asset type Synchronized metadata Public ID
AI Agent Description from source system DescriptionFromSourceSystem
Creation Date in Source CreationDateInSource
Modification Date in Source ModificationDateInSource
Initiating User in Source InitiatingUserInSource
Instructions Instructions
Tool Usage ToolUsage
Source Tags SourceTags

Status in Source

Status is either Published or Draft.

StatusInSource
Technology Asset groups / is grouped by Technology Asset TechnologyAssetGroupsTechnologyAsset
AI Agent has version / is version of AI Agent Version AgentHasVersion
AI Agent Version Description from source system DescriptionFromSourceSystem
Version Version
Owner in Source OwnerInSource
Creation Date in Source CreationDateInSource
Modification Date in Source ModificationDateInSource
Initiating User in Source InitiatingUserInSource
Source Tags SourceTags

Status in Source

Status is either Published or Draft.

StatusInSource
AI Agent has version / is version of AI Agent Version AgentHasVersion
AI Agent Version can call / can be called by AI Agent Tool AgentVersionCallsTool
AI Agent Tool AI Agent Version can call / can be called by AI Agent Tool AgentVersionCallsTool
File URL Url

Microsoft Agent Builder

  • Modification Date in Source is the date when the agent was last published. It's empty if the agent was never published.
  • Tool Usage is Web Search and only set if web search is enabled for the agent's knowledge. A single shared Web Search AI Agent Tool asset is created only if at least one agent has web search enabled.
  • Each agent has one AI Agent Version asset, named "<agent name> (v1)". The integration doesn't ingest version history.
Asset type Synchronized metadata Public ID
AI Agent Creation Date in Source CreationDateInSource
Modification Date in Source ModificationDateInSource
Initiating User in Source InitiatingUserInSource
Tool Usage ToolUsage

Status in Source

Status is either Published or Draft.

StatusInSource
Technology Asset groups / is grouped by Technology Asset TechnologyAssetGroupsTechnologyAsset
AI Agent has version / is version of AI Agent Version AgentHasVersion
AI Agent Version Owner in Source OwnerInSource
Creation Date in Source CreationDateInSource
Modification Date in Source ModificationDateInSource
Initiating User in Source InitiatingUserInSource
AI Agent has version / is version of AI Agent Version AgentHasVersion
AI Agent Version can call / can be called by AI Agent Tool AgentVersionCallsTool
AI Agent Tool AI Agent Version can call / can be called by AI Agent Tool AgentVersionCallsTool

Operational trust metrics

The Azure AI Foundry integration also reports operational trust and token consumption metrics, sourced from Azure's continuous evaluation feature (pass/fail) and Azure Monitor (token counts).

Asset type Operational trust metrics Token consumption metrics
AI Agent Version Daily pass/fail results from Azure's continuous evaluation judges, via a linked AI Monitor asset Daily input and output token counts
  • Like attributes, these metrics are stored internally. However, unlike attributes, these aren't accessible through the Collibra Platform APIs.
  • These metrics power the Agents dashboard, the Quality tab on AI Agent assets, and the Operational Health theme on the AI Trust Score.
  • Azure's continuous evaluation feature (the source of pass/fail results) only works for agents in Foundry projects because its evaluation API is only available on the Foundry project endpoint. These agents receive an AI Monitor asset via their AI Agent Deployment complex relation. Agents in classic Azure ML workspaces are still cataloged but do not receive an AI Monitor asset. This depends on the resource type, not on the agent API (Assistants or Agents) that the agent uses.
  • Agents are skipped until their AI Agent Version asset exists, so you must run the synchronization before the metrics can appear.
  • On the first run, metrics are collected for up to the last 61 days. After that, each run collects only the days since the last collected day.
  • The metrics are collected every hour on their own monitoring schedule, separate from the regular synchronization schedule. Values are stored as one total per day. Metrics can lag by about two days, so recent days can appear empty until Azure has the data. Once a day's metrics are collected, its values aren't updated. Evaluation runs or token counts that Azure reports late aren't picked up for that day.
  • No additional Azure permissions are required to ingest operational trust or token consumption metrics for the Azure AI Foundry integration.

To explore these metrics, in Settings, under AI Command Center, switch on Operational trust for AI agents.

For more information, go to Operational trust for AI agents and AI monitoring for AI agents.