From Fold to Physics: An Exact, Parameter-Free and Machine-Closed Reconstruction of Physical Science from Smithian Fold Theory

From Fold to Physics, version 1.1.0, reports the complete current 349-claim Physics reconstruction led by the exact first-principles result alpha^-1 = 503846395469/3676744786 = 137.035999177180855..., sealed and validated inside the complete CODATA 2022 interval. The same exact derivational constitution produces charged-lepton cubic and Koide relations, electron and muon magnetic anomalies, the on-shell electroweak share, Higgs mass and self-coupling, the Planck/proton hierarchy, proton radius, vacuum and cosmological ratios, three-dimensional stability, rank-two boundary support, inverse-square dilution, nuclear closure numbers, hadron trajectories, relativistic and quantum laws, thermodynamic and extraction boundaries, gravitational-wave ordering and the complete current Physics inventory. All 349 current Physics claims carry immutable model-admitted receipts. Their declared grammars preserve 249,072 explicitly generated candidates, 249,072 one-for-one decisions, 349 unique survivors and 1,396 passing mandatory adverse controls. The paper documents every derivation's root dependencies, exact theorem, candidate generator, closure boundary, survivor, axis-level uniqueness witnesses, controls, independent reconstruction, empirical custody where registered and receipt identity. The categorical clean-room audit closes all 488 Physics-owned V1/V2 atoms at the declared current-evidence boundary with no open Physics atom or gap family. Formal Grand Lock 075 binds the root trace and structural perturbation census. Empirical Grand Lock 076 reconciles all 234 pre-lock empirical claims, 147 distinct external source identities, every available measurement receipt, all disclosed legacy receipt shapes and every detected unfavorable-result or scope-boundary claim. This is also an explicit Ernos Labs open-science statement. Maria Smith developed SFT outside formal academic education, institutional research employment and conventional grant funding. The work argues that credentials, funding success, consensus and opaque predictive performance cannot substitute for derivation or empirical closure. It publishes the complete scientific chain so independent researchers can reproduce, criticize, extend or invalidate claims without institutional permission. Maria Smith retains copyright; papers and documentation are CC BY 4.0, code is Apache-2.0, and use of the Ernos Labs designation requires conformance to the public empirical and community standards. This authorized open-access release remains open to lawful extension, correction and falsification. Submissions and challenges are invited through [email protected] and https://discord.gg/ucwGryVxGr; the public project is at https://github.com/MettaMazza/ernos-labs-sft-platform.