Experimental Verdict Governance for Quantum Traction Theory

When does an experimental null actually count against a theory? An experimental verdict is meaningful only when the intended physical object was built, ordinary alternatives were controlled, the covariance and power were adequate, and the analysis was frozen before the target was opened. This paper turns those requirements into an executable governance object for Quantum Traction Theory (QTT). The registered test object is \[ \boxed{ \mathfrak T=(\mathcal O,\mathcal P,\mathcal C,\mathcal D,q,\Sigma, H_0,H_{\rm QTT},\mathcal K) } \] and its eligibility is the logical product \[ \boxed{ G_{\rm eligible} =G_{\rm physical}G_{\rm ordinary}G_{\rm covariance} G_{\rm power}G_{\rm chronology}. } \] If any gate is zero, the packet is ineligible and carries no theory verdict. If all gates pass, the executor must retain five distinct possibilities: registered falsification, QTT candidate, qualified other/cross-channel anomaly, underpowered result, or ineligible packet. A reproducible effect outside the primary channel is preserved rather than silently discarded or relabelled. \[ \boxed{ \text{protect against wrong falsification} \Longleftrightarrow \text{protect against post-result theory immunization}. } \] The paper also defines a scope lattice: failure of a registered access row does not automatically falsify a source theorem or the whole framework unless a prospectively printed dependency theorem proves that propagation. The same discipline prevents a favorable but ineligible result from becoming evidence. The audit applies this standard to the current deposited QTT test families and provides prospective successor specifications for the older A2 tabletop and HVP blind-lock packets without altering their frozen targets, hashes, or historical seals. Stable anchors: Concept DOI: 10.5281/zenodo.21775175 Main book: Quantum Traction Theory: Main Book v10.01 Website: quantumtraction.org Included files: PDF paper, Version 1.0 Reconstruction ZIP containing LaTeX source, validator, audit table, example packet, metadata, render audit, and SHA-256 manifest