This seventh paper presents the first data-level confrontation of the TCV-ϕ framework with Planck likelihoods. The analysis is organized as a staged validation ladder, starting from a CLASS bridge baseline and a native LCDM Planck run, then moving to a minimal TCV versus LCDM comparison with matched likelihood settings. In the reported benchmark, the external-spectrum interface reproduces the power-law branch at sub-percent level over the linear k range, the baseline LCDM run reaches converged Planck-era posteriors, and the minimal TCV configuration yields delta chi2 ~ +1.26 relative to LCDM. With two fewer free parameters, the same benchmark gives delta AIC ~ -2.74 and delta BIC ~ -14.38, treated here as first-pass model-comparison diagnostics within a conservative reproducibility-focused pipeline.
Cyrille Lecroq (Thu,) studied this question.