Export a Da Vinci ATR attribution list using FHIR Bulk Data $export scoped to a Group, then reconcile the exported membership against a provider roster
GET /Group/{atr-group-id}/$export with headers Accept: application/fhir+json and Prefer: respond-async; capture the 202 response Content-Location URL
Poll the Content-Location URL until the response is 200 OK (not 202); parse the response body for output file URLs with resourceType=Patient and resourceType=Group
Download NDJSON files for Patient and Group resources; the Group file contains the updated attribution list with member references and extensions
Parse each NDJSON line to extract attributed Patient IDs and their associated member-relationship and contract extensions
Cross-reference the exported Patient list against the provider's internal roster, identifying patients attributed to the provider but not in the roster (new) and roster patients absent from the export (removed)
Known gotchas
ATR bulk exports may include multiple resource types beyond Patient (e.g., Coverage, RelatedPerson); filter by resourceType in the output manifest to avoid processing irrelevant files
The export includes Group resources that are sub-groups or nested attribution lists; ensure you are processing the correct root ATR Group and not a subgroup when reconciling membership
NDJSON files are transient and hosted at pre-signed URLs that expire (typically within 24 hours); download all files immediately after the export completes rather than storing URLs for later retrieval
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp