Foundry IQ

azureSearch indexazure-foundry-iq

Permission-aware knowledge base that answers agent queries by agentic retrieval over connected enterprise data sources.

Provider
azure
Category
Data
Properties · 2 set by this component
propertytypevaluesrequiredkind defaultthis component
citations · component propertyboolean—notrueinherited
geoScheme · advancedenumnone | s2 | h3 | geohash | quadtreenononeinherited
indexingenumsync | async | batchnoasync▸ batch
knowledgeSourceType · component propertyenumazure-blob | sharepoint | onelake | web | search-indexno—inherited
managedboolean—nofalse▸ true
permissionEnforcement · component propertyboolean—notrueinherited
rebuildCadenceHours · advancedinteger—no—inherited
replicasinteger—no—inherited
retrievalReasoningEffort · component propertyenumminimal | low | mediumno—inherited
semanticReranking · component propertyboolean—notrueinherited
shardingStrategy · advancedenumnone | hash | geo | time | prefix-rangenononeinherited
structureenuminverted | trie | vector | geo-cellnoinvertedinherited
Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

derived-store-without-cdc-or-rebuild

A search index is derived data and needs an asynchronous feed from a stream, change data capture or an indexing worker. Without one it is written inline on the request path, and it can never be rebuilt after it drifts.

  • — this does not hold:
  • — the node has at least one asynchronous incoming edge from a stream, change-data-capture or worker
Use it in a design
nodes:
  - { id: "foundry-iq", component: "azure-foundry-iq" }
Other Search index products · 31
Source
# yaml-language-server: $schema=../../schema/component.schema.json
# Foundry IQ: a knowledge base over enterprise knowledge sources that agents query through agentic retrieval, built on Azure AI Search indexing.
# "What is Foundry IQ?" states indexed sources are chunked, embedded and refreshed by scheduled indexer runs, that results are semantically reranked and returned with citations, and that permissions are enforced at query time under the caller's Entra identity.
schemaVersion: "1"
type: "component"
id: "azure-foundry-iq"
kind: "search-index"
provider: "azure"
docs: "https://learn.microsoft.com/en-us/azure/foundry/agents/concepts/what-is-foundry-iq"
defaults:
  managed: true
  indexing: "batch"
properties:
  knowledgeSourceType:
    type: "enum"
    values: ["azure-blob", "sharepoint", "onelake", "web", "search-index"]
  retrievalReasoningEffort:
    type: "enum"
    values: ["minimal", "low", "medium"]
  permissionEnforcement:
    type: "boolean"
    default: true
  semanticReranking:
    type: "boolean"
    default: true
  citations:
    type: "boolean"
    default: true

The same page as markdown, for an agent: /catalogue/components/azure-foundry-iq.md