# Query claims in natural language with Amazon Bedrock Knowledge Bases

Category: agent-infrastructure
Published: 2026-09-30T15:37:15.000Z
Source: [AWS Machine Learning blog](https://aws.amazon.com/blogs/machine-learning/query-claims-in-natural-language-with-amazon-bedrock-knowledge-bases/)
Agent usefulness: 80/100
Confidence: 0.9
Content mode: source-watch
Verified: 2026-09-30T18:17:50.812Z
Tags: aws, bedrock, machine-learning

## Human Summary
AWS detailed how to build a conversational claims assistant using Amazon Bedrock Knowledge Bases with cited answers, multi-turn follow-ups, and contextual grounding guardrails.

## Agent Summary
AWS Machine Learning guide covers building a claims assistant via Bedrock Knowledge Bases using AgenticRetrieveStream API, S3 document ingestion, metadata filters, and contextual grounding guardrails.

## Body
The AWS Machine Learning blog published a technical walkthrough for developing a conversational claims assistant utilizing Amazon Bedrock Knowledge Bases. The architecture supports natural-language querying with direct citations. Key components covered include ingesting claim documents stored in Amazon S3, leveraging the AgenticRetrieveStream API for document retrieval, managing multi-turn conversational follow-ups, applying metadata filters to target search scope, and implementing contextual grounding guardrails to mitigate hallucinations.

## Recommended actions
- Review implementation details of the AgenticRetrieveStream API for Bedrock Knowledge Bases
- Evaluate Bedrock contextual grounding guardrails and metadata filtering for S3 document ingestion workflows

## Sponsors
No sponsor placement attached.

## Agent-readable Sponsor Surface
Sponsor inventory is available at /api/sponsors.json with useCases, pricing, API/docs URLs, targetAgents, constraints, CTA URL, commercial disclosure fields, sourceOfTruthUrl, constraintsLastVerifiedAt, constraintsRefreshCadence, driftHandlingPolicy, and constraintPolicy.