Three-coloured paths: partial-results candidate

Internal editorial decision

Decision: Accept with nonfatal notes / PASS_WITH_NOTES, 8 September 2026.
Scope: bounded partial-results candidate only; AIM-COMBINATORICS-0050 unresolved.

Frozen reviewed ZIP: ef9c40014e6ceb6e0b573b60739a50303a28f693bc1a12343a84c10972a88204.
PDF: a4ad630aae61c4230102d3d24ff01fd0d70f90393dea024cfb3b5364036c6975.
Manifest: 7b26f126788039e0391292e9a3796ff3331a3b90d536bfec0e04b082b35a6be8.

Five role reports are retained under editorial/. All were filed read-only
against these same hashes without reading other reports. The coordinator
served as Editor-in-Chief; one reviewer supplied domain and subsequently
applications roles, explicitly declaring this overlap. These are differentiated
internal roles, not five independent external people. All work is
producer-coordinated AI review, with no external assurance promotion.

Role	Decision	Confidence and main scope
Editor-in-Chief	Accept with presentation note	4/5 editorial fit; partial significance and claim calibration.
Methodology	Accept	High replay, moderate-to-high semantic interface.
Domain	Accept	High central proof/quantifiers, moderate dependency chain.
Applications	Accept	Safe reuse and explicit limitations; overlap declared.
Devil's Advocate	Accept with minor observations	Moderate inspected proof chain; no Critical or Major defect.

No P0/P1 scientific issue was found in this round. No mathematical claim or
evidence was expanded. The supplied review's required repairs are mapped in
REVIEW_RESPONSE.md. No further scientific confirmation round is needed for the
following tooling/presentation-only repairs under the publication policy:

• EIC-01 / applications informational note: REPORT.md now explicitly identifies
  its historical private-archive scope and points to current successor records.
• Methodology low / Devil minor 1: replay receipts now distinguish integrity-only
  mode using PASS_INTEGRITY_ONLY, explicit mode and scope, from the 16-check replay.
• Devil minor 2: the coordinator checked the primary journal theorem statement
  and the retained excess-survey page. The reviewer's failed retrieval remains
  disclosed; it is not represented as their successful source verification.
• Devil minor 3: hostile tests remain described as selected controls, not
  exhaustive verifier fault resistance. No stronger validation claim is added.

The final deterministic acceptance test covers the tooling repair and revised
archive boundary. DOI insertion, metadata, documentation corrections and final
asset regeneration are identity/presentation changes; the scientific notes and
canonical claims remain unchanged. Historical catalogue completeness, formal
verification, external specialist assessment and priority are still unestablished.

This decision is not evidence that public GitHub, Zenodo or site deployment is
complete. Those remain separate operational gates and must be read back.