All commands run from the repo root with the project venv (.venv/bin/python). The Arb backend must be live: .venv/bin/python -c "from zeta import rigor; print(rigor.BACKEND)" should print python-flint.
RF-C001 / RF-C002: sine-Gram moments (m_4 check; m_5, m_6 new)
.venv/bin/python hunts/rogue_frontier/sine_gram/exact_finite_N.py .venv/bin/python hunts/rogue_frontier/sine_gram/fast_moments.py .venv/bin/python hunts/rogue_frontier/sine_gram/lambda_structure.py .venv/bin/python hunts/rogue_frontier/sine_gram/mc_moments.py
The first two are exact integer arithmetic (no floats in any claimed value); the JSON checkpoints they write are committed alongside. The Monte Carlo is the statistical control.
RF-C003: the improved window (PROMOTED, RH-conditional)
First arm (derivation, optimization, enclosures):
.venv/bin/python hunts/rogue_frontier/window_opt/functional.py .venv/bin/python hunts/rogue_frontier/window_opt/optimize.py .venv/bin/python hunts/rogue_frontier/window_opt/crosscheck_finiteN.py .venv/bin/python hunts/rogue_frontier/window_opt/enclose.py
Headline check in one line (exact rationals; sympy):
.venv/bin/python -c "import sys; sys.path.insert(0,'hunts/rogue_frontier/window_opt'); from functional import moments_polyeven_exact, OPT_Q; print(moments_polyeven_exact(OPT_Q)[2])"
Expected: 2245228120295149280/3276332462159207451, and the final constant 1/2 + F/18 + (4/9)(19/27) = 50176758585216887915/58973984318865734118. The blinded verifier's independent scripts (structurally different derivation, same rationals) are preserved in the session scratchpad and its key outputs are quoted in RESULTS_LEDGER.md, which was not landed with this arm (it summarises the unlanded fkappa/ claim too, see LANDING.md); read it on origin/claude/riemann-hypothesis-research-ofds8s. An outside reader re-derives independently from the source paper's SS7.1 and SS7.5(g), which is the point.
RF-C004: truncated Weil form replication + enclosures + DH control
.venv/bin/python hunts/rogue_frontier/weil_trunc/run_replication.py .venv/bin/python hunts/rogue_frontier/weil_trunc/run_enclosures.py .venv/bin/python hunts/rogue_frontier/weil_trunc/run_dh.py
(See weil_trunc/RESULTS.md for the exact driver names and the JSON gates; replication.json must report 8/8, enclosures.json 27/27 conclusive-positive cells.) Sources and equation-level citations: weil_trunc/SOURCE.md.
RF-C006: Bian audit, NOT LANDED, see LANDING.md
fkappa/ was deliberately excluded from the salvage of 2026-08-21 (hunt R-F00E48). Its corrected kappa = 2 table contradicts main's landed hunts/higher_xi/ table from i = 2 onward with a conflicting diagnosis, and that is an adjudication rather than a merge. The commands below are recorded for whoever runs that adjudication; they do not resolve against this tree, and every path they name lives only on origin/claude/riemann-hypothesis-research-ofds8s:
.venv/bin/python hunts/rogue_frontier/fkappa/bian_engine.py --selftest .venv/bin/python hunts/rogue_frontier/fkappa/validate_pairs.py .venv/bin/python hunts/rogue_frontier/fkappa/validate_numeric.py .venv/bin/python hunts/rogue_frontier/fkappa/closed_form.py
The thesis itself: University of Rochester repository, institutional item 5500 (URL in fkappa/RESULTS.md); the transcribed Mathematica source is fkappa/bian_mathematica.txt, and the three defect witnesses are each a single exact rational computation printed by the validators.
RF-C005: Baez-Duarte distances
.venv/bin/python hunts/rogue_frontier/nyman_beurling/validate.py .venv/bin/python hunts/rogue_frontier/nyman_beurling/run_ladder.py .venv/bin/python hunts/rogue_frontier/nyman_beurling/analyze.py
Checkpointed coefficients and analysis are under nyman_beurling/results/.
Second wave additions
.venv/bin/python hunts/rogue_frontier/weil_trunc/dhneg_confirm.py .venv/bin/python hunts/rogue_frontier/window_opt/global_slice.py
Two further second-wave commands belong to the unlanded fkappa/ arm and are recorded here only so the adjudication can find them: fkappa/theory_validate.py and fkappa/extend_guess.py. The two-engine coefficient cross-check is one comparison of fkappa/coefficients_ext.json (14-fold-sum engine) against fkappa/row2_ext.json / row3_ext.json (resolvent engine): exact rational equality on every overlapping index is the claim.