trust-systems-meta-model

TIS Executable Artifact Walkthrough

This walkthrough demonstrates how TSMM semantics become executable Trust Infrastructure Schemas artifacts without losing governance meaning.

Scenario

A relying-party verifier is asked to allow Example Agent Beta to submit a bounded status update. The verifier must not treat registry publication or identity-state evidence as automatic runtime authorization.

Step 1: Model the system in TSMM

Use examples/cross-repo/trust-decision-system.graph.json to identify:

Step 2: Project authority into a TIS boundary

The sponsor-to-agent authority edge is projected into 01-authority-boundary.example.json. This constrains scope and requires revocation checks.

Step 3: Package evidence

The evidence set is packaged in 02-evidence-bundle-manifest.example.json. This makes evidence replayable and auditable.

Step 4: Record verifier evaluation

The verifier records assessment output in 03-evaluation-envelope.example.json. This captures controls checked, result, assurance level, semantic bindings, and authority boundary context.

Step 5: Issue a decision receipt

The relying-party result is recorded in 04-decision-receipt.example.json. This is the audit pivot because it binds policy, authority, evidence, result, effect, and timestamps.

Step 6: Publish registry state

The discoverable registry state is published in 05-registry-entry.example.json. The entry makes the artifact chain discoverable but does not authorize future runtime action by itself.

Validation evidence

The TIS repository validates the composition artifacts through tools/validate-conformance.js and the artifact coverage manifest. TSMM validates the graph and binding artifacts through the repository validation scripts.