ProductionEvidence: Medium50/100

Customer Health and Planned Lifecycle Intelligence Nexus (Chaplin) — self-service AWS Health analytics with AI agents on Amazon Bedrock

Use case typeConversational analyticsUpdated Jun 25, 2026

Chaplin is an open source solution for enterprise operations teams that turns AWS Health notifications into self-service analytics through AI agents exposed via the Model Context Protocol (MCP). It centralizes event ingestion from multiple AWS accounts into Amazon S3 and Amazon DynamoDB, then lets users ask natural-language questions in MCP-compatible assistants to get precise counts, contextual impact analysis, and remediation guidance.

Organization
AWS Technical Field Communities
Industry
Other
Location
United States
Published
June 2026

Reported outcomes

60-90 days

quantified impactTime & speed

Strategic outcomes

Cost efficiencyProactive lifecycle management and reduced manual triageRisk & complianceEarly visibility into security and compliance eventsCost efficiencyLower inference and support overhead
Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Quantified impact: 60-90 days

AWS Machine Learning BlogJun 25, 2026Blog postInferred claimMedium evidence strength

Enables proactive identification of upcoming migrations 60–90 days in advance.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
Customer
AWS Technical Field Communities
Provider
AWS
Maturity
Production

Improves operational efficiency, cost optimization, and risk mitigation through early visibility into lifecycle changes, security events, and operational impact

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

  • 1Conversational analytics
  • 2AI agents
  • 3Real-time analytics
  • Operations teams receive many AWS Health notifications across 50+ accounts and cannot quickly prioritize lifecycle events, security patches, and planned maintenance.
  • They often depend on Technical Account Managers (TAMs) for interpretation, which slows decisions and keeps teams in reactive firefighting mode.
  • Built a multi-account ingestion pipeline using AWS Health API, Amazon EventBridge, AWS Lambda, Amazon S3, and Amazon DynamoDB.
  • Exposed a Model Context Protocol (MCP) server with specialized agents for natural language to structured query conversion, contextual impact analysis, and pattern-based classification.
  • Used a pattern-first processing approach so routine categorization is handled by rules and only deeper unstructured analysis uses Amazon Bedrock with Claude 4.5 Sonnet.
  • Integrated with MCP-compatible assistants such as Claude Code and Kiro CLI so teams can query health events conversationally and plan remediation in the same workflow.
Technologies
Amazon BedrockClaude 4.5 SonnetAmazon DynamoDBAmazon S3AWS LambdaAmazon EventBridgeAWS Health APIAWS OrganizationsModel Context ProtocolStrands Agents SDK
  • Enables proactive identification of upcoming migrations 60–90 days in advance.
  • Reduces manual triage by categorizing events automatically.
  • Removes dependency on TAMs for routine analysis.
  • Improves operational efficiency, cost optimization, and risk mitigation through early visibility into lifecycle changes, security events, and operational impact.
Architecture

The solution uses AWS Health API and Amazon EventBridge in member accounts, cross-account AWS Lambda collectors, a centralized Amazon S3 data lake, and Amazon DynamoDB for structured querying. An MCP server hosts three specialized agents: a natural-language-to-structured-query agent, a contextual impact analysis agent, and a DB query builder. Pattern-based classification handles routine event categorization, while Amazon Bedrock and Claude 4.5 Sonnet handle deeper analysis. MCP-compatible assistants such as Claude Code or Kiro CLI serve as the presentation layer.

Sources & evidence1
Evidence: Medium50/100Evidence strength
  • Customer explicitly identified
  • Deployment status explicitly supported
  • Quantified outcome available
  • Technical implementation details available
Type: Blog PostPublished: Jun 25, 2026Publisher: AWSEvidence: VendorConfidence: Medium
Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases
Customer evidence
More from AWS Technical Field Communities
Provider evidence
AWS provider pageAWS ranked cases
Market context
Other AI use casesUnited States country insights
AI patterns
AgentMulti-agent
Browse the catalog
All AI use casesAI use-case typesConversational analytics use cases

Was this useful?

Community

Comments

Loading comments...

Similar cases