spec-validation

Data provenance tracking classifies every claimed-done field in a codebase as REAL, AUTHORED, or MOCK. It provides file:line evidence for each classification, ensuring that features are genuinely implemented rather than just rendered or using mocked data.

Yes, you can audit a codebase to detect mocked data by validating the implementation against the specification document. This defect detection process confirms whether claimed-done features are genuinely implemented and generates a living gap report documenting the findings.

Implementation verification requires access to the target codebase and the corresponding specification document. You provide the path to the codebase branch and the specification to initiate the validation process and generate the gap report.

Specification validation goes beyond standard automated tests by confirming that a feature is genuinely implemented, not just passing tests. It audits data provenance to classify fields as REAL, AUTHORED, or MOCK, providing deeper evidence-based verification of the actual implementation.

A living gap report is the output of specification validation that ranks and documents the verification results. It details the validation process, classifies claimed-done fields, and provides evidence-based proof of whether features are genuinely implemented in the codebase.