Software Tree launches ORM_Skyway to connect relational databases to AI agents

Software Tree on August 20, 2026 launched ORM_Skyway, an automation product that turns existing relational databases into AI-ready, governed business-object services. The company says the system can help enterprises expose useful data to AI agents without opening up raw production schemas.

Why it matters: - Enterprise AI often stalls at the data layer because business systems store information in relational tables, while AI agents work better with business concepts. - ORM_Skyway is designed to create a governed path from existing databases to AI applications without a full data migration. - The approach could matter for AI platforms, data vendors, systems integrators and developers building enterprise agent workflows.

What happened: - Software Tree announced ORM_Skyway on August 20, 2026 in Campbell, California. - ORM_Skyway automates the process of making existing relational databases AI-ready. - The product starts with an on-premises or cloud relational database, reverse-engineers the schema, generates a business object model and object-relational mapping, packages the result as a RESTful microservice, and prepares the connection to AI applications through the Model Context Protocol, or MCP. - The company positioned ORM_Skyway as a bridge between relational schemas and AI agents that interact with curated business objects.

The details: - ORM_Skyway is built on three Software Tree technologies: JDX, Gilhari and ORMCP. - JDX provides declarative ORM between relational databases and business objects. - Gilhari exposes those business objects as JSON through a Docker-native RESTful microservice. - ORMCP provides MCP connectivity so AI agents and other MCP-compatible applications can interact semantically with those objects. - ORM_Skyway is design-time automation, and its job ends after the architecture is generated and configured. - Developers can curate the generated business object model to choose which tables, attributes and relationships are available. - Developers can also rename structures in business-friendly terms before exposing them to AI. - ORMCP can enforce read-only access across a deployment for added control and security. - The product is database-agnostic and works with JDBC-compliant databases. - Verified database support includes PostgreSQL, MySQL, Oracle, SQLite, Microsoft SQL Server, Snowflake, IBM DB2, CockroachDB and SAP HANA. - Support for additional databases is in the pipeline. - ORM_Skyway is available without a license fee under its published license terms. - JDX, Gilhari and ORMCP are available under 30-day trial or beta evaluation licenses. - Developers and organizations can learn more, watch an introductory video, review technical information and download ORM_Skyway from the Software Tree website. - Software Tree also listed its LinkedIn, Facebook, YouTube and X pages in the announcement.

Between the lines: - The launch targets a familiar enterprise AI problem: giving agents useful access to operational data without exposing an entire production database schema. - By mapping only selected business objects, ORM_Skyway creates a governance boundary that keeps unmapped data out of AI workflows. - The product also shifts effort away from custom integration plumbing and toward deciding which business objects AI should see. - Early user comments in the release suggest the value proposition is both governance and speed, especially for teams testing PostgreSQL and CockroachDB deployments. - Software Tree is also signaling that ORM_Skyway and ORMCP could be building blocks for other enterprise AI vendors and services providers.

What's next: - Software Tree says more database support is coming. - The company is inviting technology and services providers to evaluate ORM_Skyway and ORMCP for potential integration and collaboration. - Enterprises and developers can test the product now under its published terms.

The bottom line: - Software Tree is trying to turn decades of ORM tooling into an AI access layer for enterprise databases, with governance built in from the start.