Azure - Foundry IQ (Preview)

Connect to Foundry IQ knowledge bases powered by Azure AI Search. Foundry IQ is a managed knowledge layer that provides agents with permission-aware, cited responses from your organization's data across Azure, SharePoint, OneLake, and the web. This connector is in preview.

This connector is available in the following products and regions:

Service Class Regions
Copilot Studio Standard All Power Automate regions
Logic Apps Standard All Logic Apps regions
Power Apps Standard All Power Apps regions
Power Automate Standard All Power Automate regions
Contact
Name Microsoft
URL https://support.microsoft.com/
Email abapi@microsoft.com
Connector Metadata
Publisher Microsoft
Website https://learn.microsoft.com/en-us/azure/search/
Privacy policy https://privacy.microsoft.com/en-us/privacystatement
Categories AI

Creating a connection

The connector supports the following authentication types:

Default Parameters for creating connection. All regions Shareable

Default

Applicable: All regions

Parameters for creating connection.

This is shareable connection. If the power app is shared with another user, connection is shared as well. For more information, please see the Connectors overview for canvas apps - Power Apps | Microsoft Docs

Name Type Description Required
Foundry IQ Search Service URL string The Azure AI Search endpoint hosting your Foundry IQ knowledge base True
Foundry IQ API Key securestring The API key for your Foundry IQ search service True

Throttling Limits

Name Calls Renewal Period
API calls per connection 5000 60 seconds

Actions

Foundry IQ Knowledge Retrieval (API)

Microsoft Foundry IQ knowledge base retrieval. Ground your agent using the Foundry IQ knowledge base for high-quality, consumption-based retrieval, delivering permission-aware, cited responses.

Foundry IQ Knowledge Retrieval (MCP)

Microsoft Foundry IQ Knowledge base retrieval via MCP protocol. Ground your agent using the Foundry IQ knowledge base for high-quality, consumption-based retrieval, delivering permission-aware, cited responses.

Foundry IQ Knowledge Retrieval (API)

Microsoft Foundry IQ knowledge base retrieval. Ground your agent using the Foundry IQ knowledge base for high-quality, consumption-based retrieval, delivering permission-aware, cited responses.

Parameters

Name Key Required Type Description
Knowledge Base Name
knowledgeBaseName True string

The name of the Foundry IQ knowledge base to query.

API Version
api-version True string

API version for the Azure AI Search Knowledge Base service.

Query Authorization Token
x-ms-query-authorization string

User token passed to Azure AI Search for document-level authorization.

role
role string

The role of the sender of the message (e.g., 'user', 'assistant').

type
type string

The type of the content (e.g., 'text').

text
text string

The text content of the message.

Returns

Foundry IQ Knowledge Retrieval (MCP)

Microsoft Foundry IQ Knowledge base retrieval via MCP protocol. Ground your agent using the Foundry IQ knowledge base for high-quality, consumption-based retrieval, delivering permission-aware, cited responses.

Parameters

Name Key Required Type Description
Knowledge Base Name
knowledgeBaseName True string

The name of the Foundry IQ knowledge base to query.

API Version
api-version True string

API version for the Azure AI Search Knowledge Base service.

Query Authorization Token
x-ms-query-authorization string

User token passed to Azure AI Search for document-level authorization.

Definitions

RetrieveResponse

Name Path Type Description
response
response array of RetrieveResponseItem

The retrieval response containing cited content from the knowledge base.

activity
activity array of RetrieveActivity

Activity log showing query planning, search index calls, and reasoning steps.

references
references array of RetrieveReference

Source document references with relevance scores and metadata.

RetrieveResponseItem

Name Path Type Description
content
content array of object

The content items in the response.

type
content.type string

The type of content (e.g., 'text').

text
content.text string

The retrieved text content with cited results from the knowledge base.

RetrieveActivity

Name Path Type Description
type
type string

The activity type (e.g., 'modelQueryPlanning', 'searchIndex', 'agenticReasoning').

id
id integer

The activity step identifier.

knowledgeSourceName
knowledgeSourceName string

The name of the knowledge source queried.

elapsedMs
elapsedMs integer

Time elapsed in milliseconds for this activity step.

inputTokens
inputTokens integer

Number of input tokens used in the model query planning step.

outputTokens
outputTokens integer

Number of output tokens produced in the model query planning step.

queryTime
queryTime date-time

The timestamp when the search index query was executed.

count
count integer

The number of results returned from the search index query.

searchIndexArguments
searchIndexArguments SearchIndexArguments

The arguments passed to the search index query.

kind
retrievalReasoningEffort.kind string

The level of reasoning effort applied (e.g., 'low', 'medium', 'high').

reasoningTokens
reasoningTokens integer

Number of tokens used in the agentic reasoning step.

SearchIndexArguments

The arguments passed to the search index query.

Name Path Type Description
search
search string

The search query text sent to the index.

filter
filter string

The OData filter expression applied to the search query.

sourceDataFields
sourceDataFields array of object

The source data fields requested from the search index.

name
sourceDataFields.name string

The name of the source data field.

searchFields
searchFields array of string

The specific fields to search within the index.

semanticConfigurationName
semanticConfigurationName string

The name of the semantic configuration used for ranking.

RetrieveReference

Name Path Type Description
type
type string

The reference type (e.g., 'searchIndex').

id
id string

The reference identifier.

activitySource
activitySource integer

The activity step identifier that produced this reference.

sourceData
sourceData object

Additional source data associated with the reference.

docKey
docKey string

The source document key in the search index.

title
title string

The title of the referenced source document.

rerankerScore
rerankerScore float

The semantic reranker relevance score for this reference.