[JoyboyBrian]

Result

Bind the selected #5413 raw145-axis caller to complete compact-control reservations: all119 main/helper/IDLE routes,16,748 literal swaps,18 local planes and16 central translations. The source uses opposite pre-route directions for main/helper versus IDLE; the checker preserves those directions and the15 actual IDLE child prefixes with their full spectator complements.

research/selected-control-reservations/PROOF.md gives a parametric address-placement implication and whole-row/child return boundary. This is a raw-caller source binding, not a new compact primitive or a final compiled-template proof. Actual final W/O/N/stock and expanded occurrence injections remain a precise open obligation; raw10928/418 are not substituted for them. No new κ, invoice, frontier change or native-all-columns closure.

The generic rotation/guard/repair scope was withdrawn after finding the retained compact-carry/seed29/seed73 audits and PR3055 physical implementation. This work follows #5445's distinct actual-reservation question. Canonical investigation: Discussion5477.

Evidence

• Python3.12.2 registered normal/optimized replay passes:207,060 lifted coordinate checks,45 actual IDLE child cases,21 row-boundary cases and8 semantic mutation rejections.
• Ten portable controls pass with Git unavailable/empty PATH, including altered source bytes and changed/missing/type-altered receipts.
• Independent Codex peer review passes, including34,510 separate basis-bit route trajectories; two receipt-boundary findings were repaired. No independent human or formal-kernel review is claimed.
• Exact committed-byte/source-history check passes; explicit base-input export and offline integrity check pass; final exported verifier passes normally and optimized without Git or PATH tools.

python3.12 -B scripts/research_package.py --package research/selected-control-reservations
python3.12 -B scripts/check_research_commit.py --package research/selected-control-reservations --check-source-history

Synthetic interval fixtures are address controls, not production parameter or tape-tariff instances. No selected full replay, physical execution, Gaussian endpoint replay, Lean or Docker build was run.

Credit

Prepared for JoyboyBrian with substantial OpenAI Codex assistance. Retain Douglas Colkitt's compact/layout argument; John-Michael Smith, Hanson Sun and Aaron Wang's audits; Jose Manuel's physical implementation; Chris R / Clear1nsight's #5413 caller and its Dillon, Ali Romman, Cao Yifan, Cinna Davis, zxv, Jacob Alan Sussman and cumulative source lineage. Apache-2.0; frozen sources unchanged.