Back to vendors
A

AWS Transform

Also known as: AWS Transform custom, AWS Transform for mainframe, AWS Blu Insights

Visit site
Entry price$0.035 per active agent minute for the listed custom workloadsFull pricing detail

AWS Transform uses specialized agents to modernize applications and migrate workloads. A team can plan a VMware move, refactor mainframe code, port a .NET application or apply a reusable change across repositories. The service produces plans, code and migration artifacts with points for human approval. Developers can run transformations through the web workbench, a CLI or an MCP client.

AWS Transform is an Amazon Web Services product for migrating infrastructure and modernizing applications. Its agents cover mainframe modernization, Windows with .NET and SQL Server, VMware and storage migration, custom code transformations and the move of AI workloads from OpenAI, Gemini or Anthropic models to Amazon Bedrock, with model mappings and cost estimates. In the workbench a team can define the job, upload source information and refine the sequence of work.

For a mainframe application, the service can extract business rules, map relationships among programs and generate requirements connected to the source. It can then produce code for a new application and run equivalence tests against the legacy behavior. AWS Transform for mainframe is generally available, and mainframe refactor includes capabilities formerly sold as AWS Blu Insights.

For migration jobs, the workbench can build migration waves from dependencies, help convert network configurations and organize cutovers. A project team decides what is permitted to proceed and responds to approvals as the agent works. The worklog and artifacts preserve what happened during the job.

AWS Transform custom ships ready made Java, Node.js and Python upgrades and can apply an organization's own transformation definition across repositories, such as a runtime migration or replacing a common library. Continuous modernization, in preview, can schedule code analyses across thousands of repositories and produce pull requests or other changes for teams using GitHub, GitLab or Bitbucket.

Developers can work through the console, the AWS Transform CLI, a Kiro power and other IDE integrations, or the AWS Transform MCP server from Claude, Cursor or Codex. The CLI can resume a prior conversation. The MCP service can work with jobs, artifacts and review tasks under the buyer’s AWS access. The platform supports AWS IAM Identity Center and federation with Okta or Microsoft Entra. AWS Transform custom and the .NET CLI run in eight AWS Regions. Some continuously scheduled tasks can execute using compute in the customer’s account.

The Windows, mainframe and VMware agents and migration assessments are available at no charge for their agent work. The standard .NET transformation includes up to 50,000 agent minutes each month. Custom transformation pricing uses $0.035 per active agent minute in AWS’s examples. A task using multiple agents can consume more minutes than the time on the clock. Other AWS resources used during a migration can incur their standard fees.

Thomson Reuters achieved four times faster transformation and 30% lower costs moving .NET workloads from Windows to Linux with AWS Transform, IDEMIA modernized .NET four times faster, Bridgestone modernized its mainframe in seven months and CSL planned VMware migration waves ten times faster. Customers have analyzed 15 billion lines of code with it, and Gartner named AWS a Leader in its 2026 Magic Quadrant for AI Augmented Code Modernization Tools. A developer can use the standard .NET transformation to rewrite application code, test the result and review the artifacts before a team rolls out the updated application.

Vendor details

Canonical URL

https://aws.amazon.com/transform/

Category

Coding agent

Subcategory

Application modernization and migration agents

Company status

first party product

Use cases & customers

Primary use cases

Modernize .NET or mainframe application codePlan and execute workload migrationsRoll out repeatable changes across repositories

Target customers

Enterprise application modernization teamsMainframe and migration program ownersDevelopers maintaining large code portfolios

Deployment options

AWS Transform web workbench in supported AWS RegionsAWS Transform CLI and IDE integrationsCustomer account compute for supported scheduled analyses

Integrations

GitHub, GitLab, Bitbucket, AWS Systems Manager and AWS account connectors. AWS Transform CLI, Kiro Power and a dedicated MCP server.

In practice

A team extracts rules from a mainframe application, checks the generated requirements and validates the new implementation with an equivalence test.

A developer applies the standard .NET transformation to a Windows application and reviews the resulting code and build outputs.

An application portfolio team schedules runtime checks across repositories and reviews the pull requests proposed for supported upgrades.

Agentic Index coverage score

11.0 / 14 capabilities · 79%

Integrations & Tool Calling Full

An agent can connect to GitHub, GitLab or Bitbucket and create a change through an authenticated source connection. Continuous modernization accepts a selected remediation and applies the appropriate transformation definition to the repository. The output can be a pull request or merge request. Migration work can also use connected AWS accounts to create infrastructure approved by the project team.

Sourcedocs.aws.amazon.com/transform/latest/userguide/continuous-modernization.htmlread 2026-11-20

Workflow Orchestration Full

Separate agents contribute to migration and modernization jobs. A mainframe project can move through portfolio assessment, extraction of business rules, code generation and verification. Migration workflows can group workloads into waves, prepare the network and manage a sequence of cutovers. A job plan connects the tasks and tracks the stages that need buyer input.

Sourcedocs.aws.amazon.com/transform/latest/launchguide/aws-transform.htmlread 2026-11-20

Knowledge Grounding & RAG Partial

AWS Transform can use supplied code, network information and deployment details to reason about a migration. It can generate an interactive diagram of application dependencies from the connection data for a job. Teams can inspect the diagram and use related findings to organize migration waves. Mainframe analysis ties extracted requirements and business rules back to source files.

Sourcedocs.aws.amazon.com/transform/latest/userguide/transform-vmware-diagrams-and-reports.htmlread 2026-11-20

Human Oversight & Guardrails Full

A project member can respond to assignments requiring human input in the collaboration pane. The Approvals tab presents items that need a decision, such as reviewing a proposed transformation plan. A user can retry a completed task with revised input when the first result needs changes. AWS Transform carries the updated plan forward after the project’s decisions are recorded.

Sourcedocs.aws.amazon.com/transform/latest/userguide/transform-environment.htmlread 2026-11-20

Security, Identity & Governance Full

Access to the AWS Transform workspace can use IAM, IAM Identity Center or direct federation. User activity is attributed to the caller’s IAM identity in CloudTrail, while users with federated access can sign in through Okta or Microsoft Entra. Stored profile information can use a customer managed key. AWS Transform is in scope for FedRAMP Class C in the AWS Northern Virginia Region.

Sourcedocs.aws.amazon.com/transform/latest/userguide/transform-setup.htmlread 2026-11-20

Observability & Auditability Full

A transformation worklog and the CLI conversation logs show the actions taken during a project. The CLI also keeps separate subagent logs so a developer can inspect what happened while different parts of the work ran. Build and test output can be used to locate a failed step. AWS CloudTrail captures API calls and the identity of the caller for account audit.

Sourcedocs.aws.amazon.com/transform/latest/userguide/custom-troubleshooting.htmlread 2026-11-20

Memory & State Persistence Partial

The AWS Transform CLI can resume the latest interactive conversation or open one by its conversation identifier. That gives a developer the context from an earlier portion of the same transformation when returning to the job. Project artifacts can remain available for later review. Reusable transformation definitions carry instructions that a team can deliberately apply to another repository.

Sourcedocs.aws.amazon.com/transform/latest/userguide/custom-command-reference.htmlread 2026-11-20

Deployment & Data Residency Full

AWS Transform custom and the .NET command line service are available in eight AWS Regions including Northern Virginia, Canada Central, Frankfurt and locations in Asia Pacific. Teams can choose where supported jobs run. Continuous modernization also supports scheduled analyses using compute in the customer’s AWS account. The CLI lets a developer work from their own environment.

Sourceaws.amazon.com/about-aws/whats-new/2026/09/aws-transform-dotnet-cliread 2026-11-20

Prebuilt Agents / Templates / Packs Full

AWS Transform includes ready to use recipes for common modernization work. A developer can select the AWS managed .NET modernization to port a supported project or use another supplied recipe for a common change. Separate managed agents handle migration, Windows and mainframe modernization workflows. An organization can also define a transformation of its own to share with its teams.

Sourcedocs.aws.amazon.com/transform/latest/launchguide/aws-transform.htmlread 2026-11-20

Triggers & Channel Coverage Full

Continuous modernization can run an analysis on a recurring schedule or when a user starts it. A team can schedule daily, weekly or monthly checks across connected repositories using EventBridge Scheduler. The command line and workspace also let developers start a custom transformation when they select a repository or supply a definition they want the agent to apply.

Sourcedocs.aws.amazon.com/transform/latest/userguide/continuous-modernization.htmlread 2026-11-20

Model Flexibility & Routing Not documented

A developer selects the transformation definition or job type and chooses how to start the work. The AWS Transform service handles the reasoning used for the selected task. The developer controls the instructions, input files and resulting code. An integration can also pass a supported transformation request from an IDE or an outside assistant to the service.

Sourcedocs.aws.amazon.com/transform/latest/userguide/custom-command-reference.htmlread 2026-11-20

APIs / SDKs / MCP Extensibility Full

The AWS Transform MCP server exposes 19 functions for workspace and job management, approvals, artifacts and other transformation resources. It uses the caller’s AWS credentials. An external coding assistant can work with jobs and review requests through this interface, including agents for mainframe or .NET work. AWS provides integrations with Kiro and other MCP clients.

Sourcedocs.aws.amazon.com/transform/latest/userguide/developer-tools.htmlread 2026-11-20

Testing, Debugging & Optimization Full

Mainframe modernization has automated equivalence tests that compare legacy behavior with the resulting application. In a custom transformation, build and test steps can verify whether the generated result passes the project’s checks. The team can inspect the test output and the latest generated files when the result needs further work. A migration job can also launch test instances before cutover.

Sourcedocs.aws.amazon.com/transform/latest/launchguide/aws-transform.htmlread 2026-11-20

Browser / Computer-use Not documented

Developers control the AWS Transform workbench from the web application, CLI or an external coding assistant. The service uses connected application code, configuration and artifacts as inputs and outputs for its jobs. Its developer tools work through the AWS Transform interface, while migration and infrastructure changes go through the configured AWS services and connectors.

Sourcedocs.aws.amazon.com/transform/latest/userguide/developer-tools.htmlread 2026-11-20

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Pricing

$0.035 per active agent minute for the listed custom workloads

Active agent minutes for paid transformations and analyses

Free tier

Included quota

The AWS managed .NET transformation includes up to 50,000 agent minutes per month. Assessments and the Windows, mainframe and VMware agents are available at no charge for agent work.

What is public

AWS Transform includes assessments and the Windows, mainframe and VMware agents. Its managed .NET recipe includes up to 50,000 agent minutes monthly. Custom transformation examples use $0.035 per active minute.

Billing mechanics

Active planning and code changes count toward agent minutes. Local compilation, test execution and idle time are excluded. Work by parallel agents is added together.

Variable cost rationale

Active work by parallel agents adds to billed minutes. Continuous analyses and remediations consume agent time, and resources required by a migration can have separate AWS charges.

Overage / add-ons

Custom transformation examples use $0.035 per active agent minute. AWS meters the work of concurrent agents separately and charges for active work even if the resulting transformation fails validation.

Sales call required

No, self serve available

Free / trial

Included assessment and selected migration or modernization agents

Lowest paid plan

AWS Transform custom

Commercial notes

Continuous modernization bills ordinary analyses and remediations in custom agent minutes. Security analysis has separate AWS Security Agent billing.

Missing data

Total cost depends on transformation work and the other AWS resources used for the project.

Agentic Index verified 2026-11-20

Alternatives to AWS Transform

Other products that do the work AWS Transform does, and what each one is.

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.