Re-runnable Scripts
Generate deployment-ready SQL that resolves keys using business identifiers at runtime. Components execute consistently across all environments, with built-in support for safe re-runs, targeted revert, and controlled rollback.
DataStar v3 for Windows
Author, version, deploy, and reverse database changes with confidence. One tool for your schema, your reference data, and the release that puts them in production: Oracle, SQL Server, Git or TFVC, Jira or Azure Boards.
Requires a licence key. How licensing works
DataStar is built around the concept of a component, a re-runnable SQL artefact representing either a schema object or a set of related data.
Using a flexible, template-driven approach, DataStar can model and manage both database structures and data across any platform. Clients define templates to represent the objects relevant to their environment, from tables, views, and procedures to complex data sets and application configurations. Platforms such as Charles River Development (CRD) are a common use case, but the approach is fully general.
All components are version-controlled, promoted through environments in a controlled manner, and fully traceable from change request to deployment, ensuring strong governance and auditability.
Generate deployment-ready SQL that resolves keys using business identifiers at runtime. Components execute consistently across all environments, with built-in support for safe re-runs, targeted revert, and controlled rollback.
A structured and repeatable release process that takes changes from work item through to deployment in a controlled and governed manner.
Supports both Git and TFVC, including branching, merging, and conflict resolution. Provides a complete, auditable history with every change traceable back to business requirements or work items.
Supports both Oracle and SQL Server within a consistent delivery model, enabling teams to manage database changes across platforms with a unified approach.
Maintain a single source of truth while supporting environment-specific configuration at deployment time, ensuring consistent and controlled releases across all environments.
Provides end-to-end traceability from change request through to deployment, delivering a complete audit trail of what changed, when, and why.
Integrates with existing delivery pipelines, including Azure DevOps, Octopus Deploy, GitHub Actions, and Bitbucket Pipelines.
Includes a Model Context Protocol (MCP) server that enables controlled interaction with components, queries, and deployments from AI tools.
No rewrites, no migrations. DataStar sits inside whatever tooling you already use.
New in v3
A guided, four-step flow that takes a work item from "I've finished my change" to "it's pushed, tagged, and ready for CI/CD", in about 60 seconds.
Link the release to a tracked ticket (a Jira story or task, an Azure Boards user story or work item, or equivalent) so every change is traceable to a business requirement.
Review and commit your pending changes inline before moving on to the deployment file.
Curate the manifest of what actually deploys. Save to workspace or attach to the ticket.
Complete the release in a single action that pushes your changes and triggers CI/CD to begin promotion through environments.
New in v3
Work directly with your data, schema, and deployment pipeline through an AI assistant that understands your environment. Explore metadata, interrogate components, and interact with your backlog in Jira or Azure Boards from a single, permission-governed interface.
Powered by an embedded Model Context Protocol (MCP) server, DataStar connects AI tools such as Claude and Kiro to your existing workflows. Investigate data, draft changes, and stage deployments with precise control over what the assistant can read and write. Read access is on by default; write access is gated behind a single toggle, off until you turn it on. Every operation is fully auditable.
Enable the MCP server →