[amaybaum]

Evidence-only integration of the five MR-1 stage-1 records from PR #1023 (branch research/measurement-record-formation-20261031, head af1e9be2ceabd5f6181b8d56b537b7ce1a4a3bba), under the owner decision of 2026-11-04.

What lands. The five files, at their branch paths under verification/programmes/oi-qm/reconstruction/measurement-record-formation/, byte-identical to that head (directory tree c59cd0b2). The preregistration is one of the five.

File	Blob	Bytes
preregistration.md (frozen at 1a236f33, unchanged since)	c36fb14b	60689
amendments/MR-star.md	45af1678	106948
amendments/KS-substitution.md	973b70ef	12384
amendments/D3-CC2-class-level.md	c16ce379	15966
amendments/stage1-log.json	378116ff	98129

No other file is added or changed. The stage-1 result comes from reading and analysis, not computation: the records state that no computation evaluating any physical, candidate, control or algebraic quantity was run. So no command or expected output is pinned (AGENTS.md §A.42 item 1), and nothing was run at landing beyond the repository's static checks.

What does not change.

• The records are not edited, and their scoped GAP conclusion is unchanged.
• MR-1 stage 2 remains unauthorized, and O-REC's outstanding prerequisites (O-RET, O-ACC) remain with P0: physical observable carrier and evolution selection #866.
• The records' off-main source anchors stay provenance only, as their preregistration §2.2 states (preregistration.md:77–87). No stage-1 conclusion depends on them. They remain branch-only dependencies:
  • research: SUBSTATE-2 recursive inverse-limit substructure and depth-feedback controls #1011, F2-1 at 4e7ddb70 (A-star.md and the F2-1 preregistration): potentially resolved by the planned research: SUBSTATE-2 recursive inverse-limit substructure and depth-feedback controls #1011 evidence landing;
  • Research (WIP): SUBSTATE-1 — internal sub-states and visible recurrence #1009, SUBSTATE-1 at 8af2d904 (later-access-result.md and related notes): remains branch-only;
  • Composition and access: finite-depth composition contract, logical acceptance circuit and complete-record laws (research record) #912 at 2af1d5d5 (REPORT.md): remains branch-only.
• No classification, manuscript, Lean module or other record changes.

Why. verification/programmes/oi-qm/reconstruction/o-ret-routes-20261103/SOURCES.md:53 on main quotes "MR-1 scoped GAP | #1023". Today the records behind that status exist only on #1023's branch. Landing them puts that support on main (§A.42 item 2).

Owner decision (amaybaum, session chat, 2026-11-04): "#1023 — MR-1 | Land the five Stage 1 records | Approved as evidence-only integration, followed by handoff to #866 and close without merging." "#1023: Retain the five Stage 1 records without altering their scoped GAP conclusion. Stage 2 remains unauthorized, and O-REC's outstanding prerequisites remain with #866." "I authorize preparing and pushing separate draft landing PRs, with independent exact-head review and CI. Neither landing is authorized to merge without a separate exact-SHA decision."

Local checks at the head:

• git hash-object equals the source blob for each file.
• duplicate_check, claims_check, voice_check and artifact_placement_check exit 0.
• stage1-log.json parses.
• The secret, sensitive-data and local-path scan found nothing.
• An independent review of cccd69cd returned CONFIRMED WITH NITS, with no required change.

Merge gate. This PR is a draft. Merging it requires green CI and the owner's separate approval of its exact head SHA, by merge commit only. #1023 itself is to be handed off to #866 and closed without merging, with its branch kept.

---

Generated by Claude Code