Chaplin (Customer Health and Planned Lifecycle Intelligence Nexus) is an open-source solution that brings self-service analytics to AWS Health events across multi-account environments. It uses a multi-agent AI architecture powered by Amazon Bedrock and exposed via the Model Context Protocol (MCP), allowing operations teams to query health events in natural language from MCP-compatible clients like Claude Code or Kiro CLI. The system combines rule-based pattern classification for cost efficiency with AI-powered contextual analysis for unstructured event descriptions. A three-tier architecture collects events via EventBridge and Lambda, stores them in DynamoDB and S3, and exposes MCP tools for querying. The solution supports both local and Lambda-based remote deployments and integrates with tools like JIRA, GitHub, and ServiceNow through MCP.
