AWS amazonq documentation change
Summary
Added documentation for 'Agentic RAG' concept, describing enhanced RAG capabilities using multiple intelligent agents and tools
Security assessment
The change introduces a new feature (Agentic RAG) but does not address security vulnerabilities or document security-specific capabilities. It focuses on improved query processing and response generation mechanics.
Diff
diff --git a/amazonq/latest/qbusiness-ug/concepts-terms.md b/amazonq/latest/qbusiness-ug/concepts-terms.md index 6a4109a72..fb6c7629d 100644 --- a//amazonq/latest/qbusiness-ug/concepts-terms.md +++ b//amazonq/latest/qbusiness-ug/concepts-terms.md @@ -5 +5 @@ -Application environmentACLs (Access Control Lists)Amazon Q AppsAnalytics dashboardAudio and video extractionBrowser extensionsChat orchestrationCustom document enrichmentData accessorsData sourceData source connectorDocumentDocument attributesField mappingsFiltering using document attributesFoundation modelGuardrailsHallucinationHallucination mitigationIAM Identity CenterIdentity Federation through IAMIdentity providerIndexIndex capacityIntegrationsISV integrationLarge language modelPrincipal MappingPluginsQuick promptsResponse personalizationRetrieverRetrieval Augmented GenerationRelevance tuningSubscription tiersTagsVisual content extractionUser storeWeb experience +Agentic RAGApplication environmentACLs (Access Control Lists)Amazon Q AppsAnalytics dashboardAudio and video extractionBrowser extensionsChat orchestrationCustom document enrichmentData accessorsData sourceData source connectorDocumentDocument attributesField mappingsFiltering using document attributesFoundation modelGuardrailsHallucinationHallucination mitigationIAM Identity CenterIdentity Federation through IAMIdentity providerIndexIndex capacityIntegrationsISV integrationLarge language modelPrincipal MappingPluginsQuick promptsResponse personalizationRetrieverRetrieval Augmented GenerationRelevance tuningSubscription tiersTagsVisual content extractionUser storeWeb experience @@ -12,0 +13,2 @@ This section describes the key concepts and terms related to Amazon Q Business. + * Agentic RAG + @@ -93,0 +96,17 @@ This section describes the key concepts and terms related to Amazon Q Business. +## Agentic RAG + +Agentic RAG (Retrieval Augmented Generation) is an advanced natural language processing technique that enhances Amazon Q Business's standard RAG capabilities. Using Agentic RAG, Amazon Q Business employs multiple intelligent agents and specialized tools to process queries and retrieve and generate responses from your enterprise data using its underlying large language model, while continuously monitoring quality. + +With Agentic RAG system processes queries through a combination of the following coordinated steps: + + * Analyzes both the user's question and conversation history and determines which retrieval tools to use + + * Intelligently triggers multiple retrieval operations as needed + + * Synthesizes information from various sources, and generate responses using its underlying large language model + + + + +For more information, see [Agentic Retrieval Augmented Generation (RAG)](https://docs.aws.amazon.com/amazonq/latest/qbusiness-ug/agentic-rag.html). +