Exchange and inspect schemas
Read types, keys and relationships. Preserve unfamiliar extensions and see which constraints a target format can express.
Check versions and operations →Schema interchange fabric
Preserve the native meaning of schemas across systems. Use the same metadata to generate tools and declare how records can change.
Experimental · Shared Rust engine · CLI, Python, Node/Bun and browser WASM
Tools read the metadata they need. Support varies by format and operation.
Diagram source and rendering limitsRead types, keys and relationships. Preserve unfamiliar extensions and see which constraints a target format can express.
Check versions and operations →Describe a change, coordinate committed actions with compensation, or model a workflow with a timed wait that can resolve early on a correlated event.
Actions · Sagas · WorkflowsGenerate Python clients for table and ontology bindings. Use typed reads and action calls through a host-provided transport.
Start with generated Python access →Browse example schemas and their relationships, or inspect a local JSON or YAML schema.
Browse the model catalog →Preview PostgreSQL DDL, SQLAlchemy models and Python CRUD operations for supported pack inputs.
Generate pack tools →Edit a UMF document, review its validation results and write it as JSON or YAML.
Open the playground →The generated reference covers the core model, extension packages, and SDK exports. Integration records give the tested versions, operations and subsets.