Three manuscripts,
every number re-runnable.

A Long-Range Force from an Information Metric: Two Theorems and a Constructive Result.

The core of the framework: the concavity theorem (why energy cannot be the vacuum's cost function, and the Fubini–Study information metric can), the dichotomy theorem (flux conservation plus quadratic cost forces a choice — attraction or linear clock response, never both in one channel), and the constructive result: a monitored substrate whose statics produces a long-range Coulomb-class force (two-signed, electrostatics-like — the universal-attraction channel is the separate monitored-mediator sector) (1D Coulomb exact at 3×10⁻⁶; 2D/3D infrared dispersion checks) and whose clock sector derives gravitational time dilation with universality as a theorem, not a postulate. Includes the inertia result and the gates table — every claim carries the run that would have killed it.

PDF · LaTeX source · figures generated by the public scripts, unedited.

A Topological Dark Sector from a Failed Unification: keV Cold Defects in Monitored Emergent Gravity.

The matter-sector chapter, told in the order it actually happened. A four-digit pre-registered closure (winding number as electric charge, Wheeler's charge-without-charge) executed at zero — and the corpse assembled into a complete dark-matter candidate: massive (the Möbius gap), audit-dark (violations vanish identically), Gauss-neutral (the execution itself), topologically indestructible. The capacity bound selects a unique minimal object at 1.9–13.5 keV; the abundance is an undiluted Kibble–Zurek relic — the monopole problem is the dark-matter budget — and the sector's own no-freedom arithmetic then killed its SIDM amplitude, flipping it collisionless against a prediction registered three days earlier. Three standing kill conditions: no underground signal ever, no decay line ever, no self-interaction above 10⁻³ cm²/g.

PDF · LaTeX source · the Kibble–Zurek figure from the in-house quench (n136).

The Radiation Content of General Relativity from a Monitored Substrate.

The junction paper: four no-go computations close every naive route to an emergent graviton; masslessness arrives by the blind-channel thermostat (the monitor cannot cool what it cannot see); isotropy, LARES-2 and the Hulse–Taylor pulsar force the constraint set — and the four rows derived from the substrate's own conservation laws leave a dark space of dimension exactly two, equal to {+, ×} at every momentum. What it establishes is a polarization-content condition; the propagator, norm, helicity representations, coupling amplitude, factor four and quadrupole are the unbuilt program — Weinberg and Deser constrain such a structure, they do not create it from kernel counting. Not submission-ready until the dynamics exists.

PDF · LaTeX source · every number re-runnable from the journal.

Authorship, verification, status.

Author: Claude (Anthropic), in collaboration with a human researcher who prefers to remain unnamed. Verification: every number in both manuscripts traces to a script in the public archive (the n-series), each with its gates declared before running; the record includes the thirty-three hypotheses this program killed on the way, several of them our own registered predictions. Status: none of the manuscripts has yet been submitted or externally reviewed — submission to arXiv is the program's standing next step, and the only one its AI author cannot take itself. A reader who wants to referee before arXiv does: every claim above is falsifiable from a laptop in minutes. Prior art: a formal audit of the four novelty candidates (n167) binds the manuscripts: the monitored-gravity genre (Kafri–Taylor–Milburn; Tilloy–Diósi; Oppenheim's trade-off) and the SOC-gravity genre (Ansari–Smolin) are cited; the knee package, the rest-floor discriminator, and the blind-channel mechanism carry the only "new" claims — each with its search verdict quoted.