← Back to all announcements
★★★★☆ 23/06/2026

Amazon OpenSearch Service now offers AI-assisted migrations

Describe your migration goal in plain language and let an AI agent handle planning, infra deployment, and cutover—weeks of manual work, automated.

View original announcement →

Visual Summary

graph TD A{{AI-Assisted Migration Assistant}}:::announced B((Self-Managed Clusters)):::external C(Amazon OpenSearch Service):::compute D(OpenSearch Serverless):::compute E(Amazon EKS):::compute F([Agent-Guided Workflow]):::feature G([Live Traffic Replay]):::feature H((AI Coding Tools)):::external I(Amazon MSK):::storage H ==>|"natural language"| A A -->|"discovers"| B A -->|"deploys stack"| E E -->|"streams via"| I A ==>|"migrates to"| C A ==>|"migrates to"| D A -->|"orchestrates"| F B -.->|"captures traffic"| G G -.->|"replays to"| C classDef announced fill:#ff9900,stroke:#ec7211,color:#fff,font-weight:bold classDef compute fill:#e3f2fd,stroke:#1565c0,color:#1565c0 classDef storage fill:#e8f5e9,stroke:#2e7d32,color:#2e7d32 classDef feature fill:#fff3e0,stroke:#e65100,color:#e65100 classDef external fill:#f5f5f5,stroke:#616161,color:#616161

What's New

Amazon OpenSearch Service's Migration Assistant now includes an AI-assisted experience that lets engineers describe their migration goals in plain language and have an agent handle planning, infrastructure deployment, and execution. The feature supports migrations from self-managed Apache Solr, Elasticsearch, and OpenSearch clusters to either OpenSearch Serverless or Managed Clusters, and works with popular AI coding tools such as Kiro, Claude Code, Codex, and Cursor. This release also adds live traffic capture and replay support for Apache Solr, a capability previously unavailable for that source platform.

How It Works

  • Agent mode activation: Install the Migration Assistant CLI via a single curl installer script, then choose "Agent mode" at setup or invoke it directly with migration-assistant agent; the agent operates as an interactive CLI session driven by natural-language instructions.
  • Environment discovery: The agent automatically identifies source cluster endpoints, validates network connectivity, checks authentication methods, and confirms all prerequisites (AWS CLI v2, kubectl, Helm 3) are in place before proceeding.
  • Migration planning: Based on your stated goals and source cluster characteristics, the agent proposes a migration path—historical backfill only, live traffic migration, or both—and presents the plan for your approval before any action is taken.
  • Infrastructure provisioning: The agent deploys the full EKS-based Migration Assistant 3.0 stack, including networking, Kafka, Pod Identity Agent add-ons, and supporting services, using Helm charts without requiring manual CloudFormation or Kubernetes authoring.
  • Workflow configuration: It generates the workflow JSON configuration (source, target, snapshot settings, schema transforms) and surfaces it for human review at defined approval gates before submission.
  • Migration execution and monitoring: The agent submits the workflow, monitors backfill and live traffic replay progress, and pauses at key checkpoints requiring explicit operator approval before advancing.
  • Validation: Post-migration, the agent compares document counts between source and target, verifies schema translation accuracy, and checks replayer health to confirm a successful migration.
  • Solr live traffic support: A new capture-and-replay proxy can now intercept and record live Solr traffic, replay it against the OpenSearch target, and compare latency and response codes for pre-cutover validation.

Why It's Important

  • Dramatic reduction in planning overhead: Migrations that previously required weeks of manual planning—constructing workflow JSON, running CLI commands, and troubleshooting step by step—can now be initiated and structured through conversational prompts, compressing the pre-migration phase significantly.
  • Lower error risk: The agent-guided workflow enforces a structured, repeatable sequence with built-in approval gates, reducing the ad-hoc mistakes that commonly occur when engineers manually orchestrate complex multi-step migrations.
  • Broader source platform coverage: Adding live traffic capture and replay for Apache Solr closes a major gap, enabling organizations running Solr-based search infrastructure to validate OpenSearch behavior under real production traffic before cutting over.
  • Democratizes complex migrations: Teams without deep OpenSearch or Kubernetes expertise can now execute production-grade migrations by describing their intent in plain language, lowering the skill barrier for cloud adoption.
  • Risk mitigation preserved: The AI layer sits on top of the existing low-risk migration architecture—abort capabilities, source cluster preservation, snapshot-based backfill, and multi-hop support—so automation does not come at the cost of safety controls.

How It's Different

  • Natural language vs. manual CLI: Unlike the original Migration Assistant (launched December 2023), which required engineers to manually construct workflow JSON and run CLI commands, the new AI-assisted mode accepts plain-language descriptions and translates them into executable steps automatically.
  • Tool-agnostic AI integration: Rather than locking users into a single AI platform, the agent works with any MCP-compatible or CLI-driven coding agent (Kiro, Claude Code, Codex, Cursor, and others), fitting into existing developer toolchains.
  • End-to-end orchestration vs. task automation: The original Migration Assistant automated individual manual tasks; the AI agent orchestrates the entire lifecycle—assessment, infrastructure deployment, configuration, execution, and validation—as a single guided workflow.
  • Solr live traffic parity: Competing migration tools and the prior version of Migration Assistant lacked live traffic capture and replay for Solr; this release brings Solr to feature parity with Elasticsearch and OpenSearch sources.
  • Approval-gated autonomy: Unlike fully autonomous migration scripts, the agent pauses at defined checkpoints for human review, providing a balance between automation speed and operator control that is uncommon in self-service migration tooling.

When to Prefer It

  • Large or complex self-managed clusters: Use the AI-assisted mode when migrating clusters with many indices, custom analyzers, or complex schema mappings where manually authoring workflow JSON would be time-consuming and error-prone.
  • Teams new to OpenSearch or Kubernetes: Ideal when the migration team lacks deep familiarity with OpenSearch internals or the EKS-based Migration Assistant infrastructure, as the agent handles configuration generation and troubleshooting automatically.
  • Apache Solr migrations requiring live validation: Choose this path when migrating from Solr and needing to validate OpenSearch behavior under real production traffic before cutover, now enabled by the new live traffic capture and replay feature.
  • Time-constrained migration projects: When business deadlines compress the planning window, the agent-guided workflow can accelerate the assessment and infrastructure-deployment phases that traditionally consume the most calendar time.
  • Proof-of-concept or iterative migrations: Well-suited for teams wanting to run a low-risk trial migration to OpenSearch Serverless or a Managed Cluster before committing to full production cutover, since the agent structures and validates each phase independently.
  • Organizations standardizing on AI coding tools: If your engineering workflow already uses Kiro, Claude Code, or similar agents for infrastructure-as-code tasks, the Migration Assistant agent mode integrates naturally into that existing toolchain.

Availability

  • GA status: Generally available as of June 23, 2026; this is a production-ready release, not a preview.
  • Supported regions: Available in all commercial AWS Regions and AWS GovCloud (US) Regions where Amazon OpenSearch Service is offered.
  • Supported sources: Apache Solr 6.x–9.x, Elasticsearch 1.x–8.x, and OpenSearch 1.x–3.x; Elasticsearch 1.x–2.x migrations are backfill only (live traffic capture and replay not supported for those versions).
  • Supported targets: Amazon OpenSearch Service Managed Clusters (OpenSearch 1.x, 2.x, 3.x) and Amazon OpenSearch Serverless NextGen; document ID preservation on Serverless applies to SEARCH collections only.
  • Pricing: Migration Assistant is an AWS Solution (open-source, deployed into your account); you pay for the underlying AWS resources consumed (EKS, EC2, Kafka/MSK, networking) at standard AWS rates—there is no additional charge for the Migration Assistant software or AI agent mode itself.
  • Prerequisites: AWS CLI v2, kubectl, Helm 3, valid AWS credentials, and a supported AI coding agent CLI must be available on the operator's workstation before launching agent mode.

Tags

Servicesopensearch
Typenew-feature
Conceptsagentic-aigenai
Use Casesmigrationdeveloper-tools
Providersanthropic
GeographyGlobal

Related Resources

AI Radar AWS

AWS AI/ML news — curated, researched, explained

An automated intelligence platform that curates, researches, and analyzes AWS AI/ML/GenAI announcements daily. Every report is backed by real research — the system reads linked blog posts and documentation to provide accurate, in-depth analysis.

How Each Report Is Generated

  1. Collection — Daily monitoring of the AWS "What's New" RSS feed
  2. Filtering — AI-powered relevance detection for AI/ML/GenAI topics
  3. Taxonomy Tagging — LLM-based classification across 6 dimensions
  4. Importance Scoring — Point-based system with tag bonuses (1-5 stars)
  5. Research Phase — Follows links to blog posts and documentation
  6. Report Generation — Claude Sonnet produces structured 6-section analysis
  7. Visual Summary — Claude Opus generates Mermaid diagrams for key items
  8. Publishing — Static website rebuilt and deployed via CloudFront

Features

  • Faceted filtering by service, type, concept, and more
  • Multi-dimensional taxonomy with 80+ tags across 6 dimensions
  • Geographic availability badges (Global, APJ, EMEA, AMER) with filtering
  • Timeline visualization of announcement volume
  • PDF export for offline reading
  • Mermaid visual summaries for key announcements
  • Daily automated updates — no manual curation
What makes this different: Each report involves a dedicated research phase where the system reads linked blog posts and AWS documentation pages. This produces analysis that goes beyond the original announcement text.

Technology

Built with Python, AWS Lambda, Amazon Bedrock (Claude Sonnet 4.6, Opus 4.6, Haiku 4.5), S3, CloudFront, WAF, EventBridge, and CDK.

Open Source

This project is open source. Fork it, customize it for your needs, and deploy your own instance.
📦 github.com/bbonik/ai-radar-aws

How Importance Scoring Works

Each announcement receives a point score based on multiple factors. The total score maps to a 1-5 star rating:

1★ < 2 pts 2★ ≥ 2 pts 3★ ≥ 3.5 pts 4★ ≥ 5 pts 5★ ≥ 6.5 pts

Point Breakdown

FactorPointsWhen
Core AI service (Bedrock, AgentCore, SageMaker AI)+4Service named in title
Key AI service (SageMaker, Kiro, QuickSight)+2Service named in title
Other AI-related service+1Default
Blog post link+3Link to aws.amazon.com/blogs/
GitHub samples link+2Link to github.com/aws*
Documentation link+1Link to docs.aws.amazon.com/
New model+1.5Tagged as "new-model"
New service+1Tagged as "new-service"
New feature+0.5Tagged as "new-feature"
Anthropic / OpenAI provider+2Provider explicitly mentioned
Instance / notebook announcement-2Hardware/capacity, not feature
Performance / pricing / security-0.5Incremental updates
Region expansion to APJ+1Expands to Asia Pacific
Region expansion (non-APJ only)-1.5Only expands to other regions

Geographic Relevance Badges

Each announcement card shows a small badge indicating whether the feature is available in your region:

🌐 Global Available in all regions
🌏 APJ Asia Pacific
🌍 EMEA Europe / Middle East / Africa
🌎 AMER Americas (US, Canada, South America)
No badge Geography unknown
How geography is detected: The system detects ALL geographies mentioned in each announcement. If the text mentions specific regions (Tokyo, Frankfurt, Oregon, etc.), the corresponding geography badges are shown. If it says "all regions" or is a new feature with no region specified, it gets the Global badge. Geography is also filterable — click a geo chip to see only announcements available in that region.