NORMTRACE-30: Evidence-preserving assurance for standards mappings

NORMTRACE-30 is a reproducible standards-assurance engineering capsule for checking join integrity across legal requirements, exact standard versions, mapping rows, dependencies, evidence objects, model outputs, and human dispositions.

The thesis-facing corpus is a CEN/CENELEC AI-assurance spine covering AI Act logging, transparency, human oversight, and quality-management evidence. A separate five-standard ETSI radio corpus is used only to validate the instrument under complete first-party source access. The release includes code, schemas, freezes, prompts, derived metadata, model receipts, reports, figures, a signed 34-artifact checkpoint, RFC 3161 time evidence, and the exported EATF verifier correction.

No standards full text, private model contexts, mutation key, credentials, or blind human-review packet is included. Construction-key scores remain provisional pending independent adjudication. Cryptographic evidence authenticates bytes and time, not semantic truth or legal conformity.

Mixed licensing: source code is MIT; original documentation, figures, schemas, and datasets are CC BY 4.0. Third-party standards and marks are not licensed. See LICENSE.md in the capsule.