docs/CLAIMS.md

Claim ledger — v3.0.0

“New candidate” means developed here and not matched by the targeted comparison. It does not certify worldwide priority. The manuscript contains full mathematical hypotheses and proofs.

Result	Mathematical scope	Status
Restricted-block row ideals	Exact direct-sum row module and supported generators	Retained candidate operational integration
Finite analytic ideal realization	Prescribed generators explicitly stored in coefficients	Retained construction; not free computation
Arc-blind examples	Same arc orders, different ideal membership	Observation realization of an established algebraic distinction
Elliptic m-fold point	Second-deficit axes-plus-diagonal curve	Known geometry; Smyth/Stevens credited
Third-deficit classification	m >= 6, independent parameters, reduced analytic germ near exterior point	New candidate theorem with proof
Six-channel threshold	Dense rational m=5 counterexample	Exact counterexample
Intrinsic linear rigidity	Aut_lin(Z_m) = C* x S_m	New candidate theorem with proof
Selected-component budget	ceil(2(m-1)/3) whole twin constraints	New candidate sharp theorem; not sample complexity
Full-state rank loss	Exact indicator formula; double-star locus	New candidate consequence with proof
Scalar certificate algorithm	O(m^4) exact rational scalar operations	Proven algorithm; bit complexity excluded
Core-capacity law	Generic rank for arbitrary core size and r >= 1	New candidate model-specific law
Analytic feedback factorization	Complete joint analytic frame, finite analytic transcript	Explicit theorem under restrictive hypotheses; not a universal claim about adaptivity

The all-size results are supported by written mathematical arguments. The 1,028 test cases and the two enumerations do not independently certify all analytic proofs. No theorem is counted as globally original merely because its notation or project name is new.