trust-systems-meta-model

Validation and Testability

TSMM should be testable as a model, not only readable as documentation. The current validation layer is intentionally lightweight, but it is designed to catch structural drift before it turns into interpretive drift.

Validation surfaces

The repository now validates six surfaces:

1. core and extension examples against their schemas
2. schema coverage so examples exercise modeled properties
3. ecosystem bindings against the binding schema
4. binding constraint sets against the constraint-set schema
5. representative valid and invalid test vectors
6. YAML-based authority, delegation, lifecycle, assurance, and interoperability artifacts

Validation tree

• validation/schemas/tsmm-binding-constraints.schema.json
• validation/test_vectors/valid/
• validation/test_vectors/invalid/
• validation/conformance/tsmm-validation-profile.json

Validation scripts

• python scripts/validate_examples.py
• python scripts/check_schema_coverage.py
• python scripts/validate_bindings.py
• python scripts/validate_test_vectors.py
• python scripts/check_docs.py
• python scripts/validate_yaml_models.py

Design intent

This layer does not claim external certification. It does something more immediate and more useful for an evolving reference model: it verifies that the catalog, examples, bindings, and documentation still fit together coherently after each increment on the main branch.

This site is open source. Improve this page.