Gaia Coherent Kinematics Survey: Second-Architecture Provenance and Follow-up Record

The first preregistered Gaia Coherent Kinematics Survey detector failed its calibration audit because its empirical reference tails saturated at the available null resolution. No real-sky detection claim was made from that architecture. A second, separately defined analysis architecture was subsequently developed to improve sensitivity while retaining the controls motivated by the failed calibration. The second full-sky run examined 1,085,109 Gaia DR3 stars within 300 pc and produced 747 consolidated structures. Of these, 37 persisted across at least two independent analysis configurations and contained 555 unique stars. Persistence was essentially uncorrelated with the previously identified core-population/density artifact (ρ≈0.003, p≈0.95\rho \approx 0.003,\ p \approx 0.95). This record establishes provenance for that frozen 555-star follow-up cohort without publicly disclosing the complete target list or the implementation of the second detector. SHA-256 commitments to the canonical 555-source set and the 37-structure membership map are included so that the complete cohort can later be disclosed and cryptographically verified against this timestamped record. Follow-up is proceeding by staged unblinding. Only the first 16 stars of the 555-star cohort have undergone detailed photometric pilot analysis to date. Several initially interesting signals were eliminated by improved source association, photometric quality control, cadence-alias rejection, and cross-instrument validation. A small number of candidates remain under investigation, including one source showing reproducible excess variability in ZTF and three independent TESS sectors. This document is a post-execution provenance and cryptographic-commitment record. It is not a preregistration and does not constitute a discovery claim. At the date of publication, no unexplained astrophysical signal has been established.