# Provider comparison — EM-DAT-07

Claude and Grok agree that Analytical Study is a profile of WM-ACT-036, while WM-ACT-034 provides reusable assessment mechanics but does not master inquiry identity. Both keep reusable Analysis Method outside the study as an identifier-unassigned candidate and separate method definition from one study's pinned application and execution. Both reuse WM-KNW-007 for findings and recommendations, preserve reports as presentation records, and require WM-KNW-010 plus WM-ACT-024 before a recommendation has decision effect.

Grok sharpens the reconciled shape in four places. Claim mood must explicitly distinguish assertoric findings from deontic recommendations. Every finding pins study, method, sample and data versions. A causal finding cannot complete without measurand, comparator, design and competing-explanation analysis, and an undischarged sampling-frame defeater blocks causal closure. Changing any of those pins creates a successor finding instead of rewriting the issued claim.

The reconciled disposition is PROFILE plus one identifier-unassigned Analysis Method allocation candidate. No runtime or model identifier is created. Publication remains held because Analysis Method has no registry allocation, relation contracts and conflict-link vocabulary are incomplete, WM-REC-002 is not canonically publishable, and all artifacts remain research candidates.