diff --git a/CHANGELOG.md b/CHANGELOG.md
index 304fc957..1fe7703b 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -7,6 +7,14 @@
 
 # Changelog
 
+## Sprint 190 Regional Kempen GRB buildings (2026-07-14)
+
+- Added an explicit regional GRB building operator that fetches the approved Kempen scope in 28 resumable municipality partitions and follows every OGC API pagination link.
+- Assigned boundary-crossing GRB features to exactly one partition using maximum municipality intersection, with deterministic NIS-code tie breaking and one retained source identity.
+- Added streaming artifact copy and batch-wise partition indexing through `DatasetService` and `VectorFeatureService`, avoiding one giant multipart parse while preserving one normal regional dataset for existing map and PostGIS selection flows.
+- Added truncation guards, source/checksum manifests, immutable observation dates, duplicate-source rejection and focused service/operator tests.
+- Kept the provider endpoint contract unchanged and retained explicit operator-only fetching; no source request runs during application startup or interactive map use.
+
 ## Sprint 189 Official Kempen operational scope (2026-07-14)
 
 - Defined `Kempen` operationally as the official 28-municipality Vlaamse vervoerregio, with an explicit warning that this policy boundary is not the wider cultural or landscape Kempen.
diff --git a/backend/README.md b/backend/README.md
index 46a6fca6..6db4be20 100644
--- a/backend/README.md
+++ b/backend/README.md
@@ -956,6 +956,22 @@ Use `--fetch-only` for a source/geometry/checksum audit. The scope pass does
 not fetch thematic GRB, population or land-use data; those remain separate,
 bounded operator jobs.
 
+Provision the regional GRB building theme after the scope pass:
+
+```bash
+docker exec geointel python /app/scripts/provision_regional_grb_buildings.py \
+  --scope kempen-transport-region
+```
+
+The operator retains 28 checksummed municipality partitions but exposes one
+normal regional reference dataset. `StorageService` copies the combined
+artifact without materializing it as upload bytes; `DatasetService` creates
+the Dataset and immutable DatasetVersion; `VectorFeatureService` validates and
+flushes partition features in bounded batches. The transaction must index the
+exact manifest feature count or it rolls back and removes the managed copy.
+No public API contract or provider readiness claim is changed by this
+operator-only path.
+
 ## Temporal Mol data and evolution
 
 Dataset uploads accept `temporal_series_key`, `observed_at`, `valid_from`,
diff --git a/backend/app/services/dataset_service.py b/backend/app/services/dataset_service.py
index 6c171229..e2bf38e8 100644
--- a/backend/app/services/dataset_service.py
+++ b/backend/app/services/dataset_service.py
@@ -12,7 +12,7 @@ from fastapi import UploadFile
 from sqlalchemy.orm import Session
 
 from app.core.errors import AppError
-from app.models import Dataset, DatasetVersion, Project
+from app.models import Area, Dataset, DatasetVersion, Project
 from app.schemas.dataset import (
     DatasetCreateResponse,
     DatasetStorageResponse,
@@ -410,6 +410,136 @@ class DatasetService:
 
         return DatasetService._to_response(dataset)
 
+    @staticmethod
+    def import_partitioned_vector_artifact(
+        db: Session,
+        *,
+        project_id: UUID,
+        area_id: UUID,
+        artifact_path: str | Path,
+        partition_paths: list[str | Path],
+        original_filename: str,
+        source: str,
+        dataset_role: str,
+        source_name: str,
+        reference_layer_name: str | None,
+        metadata_json: dict[str, Any],
+        source_metadata: dict[str, Any],
+        provenance_metadata: dict[str, Any],
+        temporal_series_key: str,
+        observed_at: datetime,
+        temporal_granularity: str = "snapshot",
+        source_version: str | None = None,
+        batch_size: int = 1000,
+    ) -> DatasetCreateResponse:
+        if not db.get(Project, project_id):
+            raise AppError(code="PROJECT_NOT_FOUND", message="Project not found", status_code=404)
+        area = db.get(Area, area_id)
+        if not area:
+            raise AppError(code="AREA_NOT_FOUND", message="Area not found", status_code=404)
+        if area.project_id != project_id:
+            raise AppError(code="INVALID_DATASET_SCOPE", message="Area does not belong to this project", status_code=400)
+        if not partition_paths:
+            raise AppError(
+                code="INVALID_GEOJSON_PARTITIONS",
+                message="At least one GeoJSON partition is required",
+                status_code=400,
+            )
+
+        filename = DatasetService._validate_upload_filename(original_filename)
+        if DatasetService._extension_for_path(filename) not in DatasetService.VECTOR_EXTENSIONS:
+            raise AppError(code="INVALID_UPLOAD", message="Vector artifacts require .geojson or .json files", status_code=415)
+        normalized_role = DatasetService._normalize_dataset_role(dataset_role)
+        temporal = DatasetService._validate_temporal_metadata(
+            temporal_series_key=temporal_series_key,
+            observed_at=observed_at,
+            valid_from=observed_at,
+            valid_to=None,
+            temporal_granularity=temporal_granularity,
+            source_version=source_version,
+        )
+        metadata = dict(metadata_json)
+        expected_feature_count = int(metadata.get("feature_count") or 0)
+        if expected_feature_count <= 0:
+            raise AppError(
+                code="INVALID_GEOJSON_PARTITIONS",
+                message="Partition metadata must declare a positive feature_count",
+                status_code=400,
+            )
+
+        dataset_id = uuid.uuid4()
+        storage_info = StorageService.persist_dataset_file_from_path(
+            project_id=str(project_id),
+            dataset_id=str(dataset_id),
+            dataset_type="vector",
+            original_filename=filename,
+            source_path=artifact_path,
+            content_type="application/geo+json",
+        )
+        dataset = Dataset(
+            id=dataset_id,
+            project_id=project_id,
+            area_id=area_id,
+            name=filename,
+            dataset_type="vector",
+            source=source,
+            dataset_role=normalized_role,
+            source_name=source_name,
+            reference_layer_name=reference_layer_name if normalized_role == "reference" else None,
+            source_metadata=source_metadata,
+            provenance_metadata=provenance_metadata,
+            imported_at=datetime.now(timezone.utc),
+            **temporal,
+            storage_path=storage_info["storage_path"],
+            original_filename=storage_info["original_filename"],
+            stored_filename=storage_info["stored_filename"],
+            content_type=storage_info["content_type"],
+            size_bytes=storage_info["size_bytes"],
+            checksum_sha256=storage_info["checksum_sha256"],
+            crs=str(metadata.get("crs") or "EPSG:4326"),
+            bounds_json=metadata.get("bounds_json"),
+            metadata_json=metadata,
+            status="ready",
+        )
+        try:
+            db.add(dataset)
+            db.add(
+                DatasetVersion(
+                    dataset_id=dataset.id,
+                    version=1,
+                    storage_path=dataset.storage_path,
+                    source_version=dataset.source_version,
+                    observed_at=dataset.observed_at,
+                    valid_from=dataset.valid_from,
+                    checksum_sha256=dataset.checksum_sha256,
+                    source_metadata=dataset.source_metadata,
+                    provenance_metadata=dataset.provenance_metadata,
+                )
+            )
+            persisted_count = VectorFeatureService.persist_geojson_partitions(
+                db,
+                dataset.id,
+                partition_paths,
+                feature_class=reference_layer_name if normalized_role == "reference" else None,
+                batch_size=batch_size,
+            )
+            if persisted_count != expected_feature_count:
+                raise AppError(
+                    code="PARTITION_FEATURE_COUNT_MISMATCH",
+                    message=(
+                        f"Regional artifact declares {expected_feature_count} features but "
+                        f"{persisted_count} queryable features were indexed"
+                    ),
+                    status_code=400,
+                )
+            db.commit()
+            db.refresh(dataset)
+        except Exception:
+            db.rollback()
+            StorageService.remove_dataset_file(storage_info["storage_path"])
+            raise
+        return DatasetService._to_response(dataset)
+
     @staticmethod
     def refresh_metadata(db: Session, dataset_id: UUID) -> DatasetCreateResponse:
         dataset = DatasetService._get_dataset(db, dataset_id)
diff --git a/backend/app/services/storage_service.py b/backend/app/services/storage_service.py
index fc3694f3..62a091f2 100644
--- a/backend/app/services/storage_service.py
+++ b/backend/app/services/storage_service.py
@@ -95,6 +95,39 @@ class StorageService:
         }
         return metadata
 
+    @staticmethod
+    def persist_dataset_file_from_path(
+        project_id: str,
+        dataset_id: str,
+        dataset_type: str,
+        original_filename: str,
+        source_path: str | Path,
+        content_type: str | None,
+    ) -> dict[str, Any]:
+        source = Path(source_path).resolve()
+        if not source.is_file():
+            raise FileNotFoundError(f"Dataset source artifact does not exist: {source}")
+
+        normalized_type = StorageService.normalize_dataset_type(dataset_type)
+        file_path = Path(StorageService.dataset_file_path(project_id, dataset_id, normalized_type, original_filename))
+        file_path.parent.mkdir(parents=True, exist_ok=True)
+        digest = hashlib.sha256()
+        size_bytes = 0
+        with source.open("rb") as input_stream, file_path.open("wb") as output_stream:
+            for chunk in iter(lambda: input_stream.read(8 * 1024 * 1024), b""):
+                output_stream.write(chunk)
+                digest.update(chunk)
+                size_bytes += len(chunk)
+
+        return {
+            "original_filename": StorageService._safe_filename(original_filename),
+            "stored_filename": file_path.name,
+            "content_type": content_type or "application/octet-stream",
+            "size_bytes": size_bytes,
+            "checksum_sha256": digest.hexdigest(),
+            "storage_path": str(file_path),
+        }
+
     @staticmethod
     def persist_file(
         storage_path: str,
diff --git a/backend/app/services/vector_feature_service.py b/backend/app/services/vector_feature_service.py
index 4795e4f4..b423c9a1 100644
--- a/backend/app/services/vector_feature_service.py
+++ b/backend/app/services/vector_feature_service.py
@@ -1,6 +1,8 @@
 from __future__ import annotations
 
-from typing import Any
+import json
+from pathlib import Path
+from typing import Any, Iterable
 from uuid import UUID
 
 from geoalchemy2.functions import ST_Intersects, ST_MakeEnvelope
@@ -17,6 +19,37 @@ from app.models import Dataset, VectorFeature
 
 
 class VectorFeatureService:
+    @staticmethod
+    def _feature_row(dataset_id: UUID, feature: dict[str, Any], index: int, feature_class: str | None) -> VectorFeature | None:
+        geometry_payload = feature.get("geometry")
+        if geometry_payload is None:
+            return None
+        try:
+            geometry = shape(geometry_payload)
+        except Exception as exc:
+            raise AppError(code="INVALID_GEOJSON", message=f"Invalid feature geometry at index {index}", status_code=400) from exc
+        if geometry.is_empty:
+            return None
+        if not geometry.is_valid:
+            geometry = make_valid(geometry)
+        if geometry.is_empty or not geometry.is_valid:
+            raise AppError(code="INVALID_GEOMETRY", message=f"Invalid feature geometry at index {index}", status_code=400)
+        if geometry.has_z:
+            geometry = transform_geometry(lambda x, y, z=None: (x, y), geometry)
+
+        properties = feature.get("properties") if isinstance(feature.get("properties"), dict) else {}
+        source_feature_id = feature.get("id")
+        if source_feature_id is None:
+            source_feature_id = properties.get("id") or properties.get("source_feature_id")
+
+        return VectorFeature(
+            dataset_id=dataset_id,
+            feature_class=feature_class,
+            source_feature_id=str(source_feature_id) if source_feature_id is not None else None,
+            properties_json=properties,
+            geometry=from_shape(geometry, srid=4326),
+        )
+
     @staticmethod
     def _normalize_selection_bbox(bbox: dict[str, Any]) -> dict[str, float | str]:
         try:
@@ -263,34 +296,9 @@ class VectorFeatureService:
         for index, feature in enumerate(features):
             if not isinstance(feature, dict):
                 raise AppError(code="INVALID_GEOJSON", message=f"Feature {index} must be an object", status_code=400)
-            geometry_payload = feature.get("geometry")
-            if geometry_payload is None:
+            row = VectorFeatureService._feature_row(dataset_id, feature, index, feature_class)
+            if row is None:
                 continue
-            try:
-                geometry = shape(geometry_payload)
-            except Exception as exc:
-                raise AppError(code="INVALID_GEOJSON", message=f"Invalid feature geometry at index {index}", status_code=400) from exc
-            if geometry.is_empty:
-                continue
-            if not geometry.is_valid:
-                geometry = make_valid(geometry)
-            if geometry.is_empty or not geometry.is_valid:
-                raise AppError(code="INVALID_GEOMETRY", message=f"Invalid feature geometry at index {index}", status_code=400)
-            if geometry.has_z:
-                geometry = transform_geometry(lambda x, y, z=None: (x, y), geometry)
-
-            properties = feature.get("properties") if isinstance(feature.get("properties"), dict) else {}
-            source_feature_id = feature.get("id")
-            if source_feature_id is None:
-                source_feature_id = properties.get("id") or properties.get("source_feature_id")
-
-            row = VectorFeature(
-                dataset_id=dataset_id,
-                feature_class=feature_class,
-                source_feature_id=str(source_feature_id) if source_feature_id is not None else None,
-                properties_json=properties,
-                geometry=from_shape(geometry, srid=4326),
-            )
             db.add(row)
             persisted.append(row)
 
@@ -298,3 +306,69 @@ class VectorFeatureService:
             db.flush()
             db.commit()
         return persisted
+
+    @staticmethod
+    def persist_geojson_partitions(
+        db,
+        dataset_id: UUID,
+        partition_paths: Iterable[str | Path],
+        feature_class: str | None = None,
+        *,
+        batch_size: int = 1000,
+    ) -> int:
+        if batch_size <= 0:
+            raise ValueError("batch_size must be positive")
+
+        persisted_count = 0
+        source_feature_ids: set[str] = set()
+        for partition_path in partition_paths:
+            path = Path(partition_path)
+            try:
+                payload = json.loads(path.read_text(encoding="utf-8"))
+            except (OSError, json.JSONDecodeError) as exc:
+                raise AppError(
+                    code="INVALID_GEOJSON_PARTITION",
+                    message=f"Could not read GeoJSON partition {path.name}",
+                    status_code=400,
+                ) from exc
+            features = payload.get("features")
+            if payload.get("type") != "FeatureCollection" or not isinstance(features, list):
+                raise AppError(
+                    code="INVALID_GEOJSON_PARTITION",
+                    message=f"GeoJSON partition {path.name} must be a FeatureCollection",
+                    status_code=400,
+                )
+
+            batch: list[VectorFeature] = []
+            for index, feature in enumerate(features):
+                if not isinstance(feature, dict):
+                    raise AppError(
+                        code="INVALID_GEOJSON_PARTITION",
+                        message=f"Feature {index} in {path.name} must be an object",
+                        status_code=400,
+                    )
+                row = VectorFeatureService._feature_row(dataset_id, feature, index, feature_class)
+                if row is None:
+                    continue
+                if row.source_feature_id:
+                    if row.source_feature_id in source_feature_ids:
+                        raise AppError(
+                            code="DUPLICATE_SOURCE_FEATURE",
+                            message=f"Duplicate source feature {row.source_feature_id} across regional partitions",
+                            status_code=400,
+                        )
+                    source_feature_ids.add(row.source_feature_id)
+                db.add(row)
+                batch.append(row)
+                persisted_count += 1
+                if len(batch) >= batch_size:
+                    db.flush()
+                    for persisted in batch:
+                        db.expunge(persisted)
+                    batch.clear()
+            if batch:
+                db.flush()
+                for persisted in batch:
+                    db.expunge(persisted)
+
+        return persisted_count
diff --git a/backend/tests/test_sprint190_regional_grb_buildings.py b/backend/tests/test_sprint190_regional_grb_buildings.py
new file mode 100644
index 00000000..a281699e
--- /dev/null
+++ b/backend/tests/test_sprint190_regional_grb_buildings.py
@@ -0,0 +1,195 @@
+from __future__ import annotations
+
+import importlib.util
+import json
+from pathlib import Path
+import sys
+from uuid import uuid4
+
+import pytest
+from shapely.geometry import Polygon
+from shapely.ops import unary_union
+
+from app.core.errors import AppError
+from app.services.storage_service import StorageService
+from app.services.vector_feature_service import VectorFeatureService
+
+
+ROOT = Path(__file__).resolve().parents[2]
+SCRIPTS = ROOT / "scripts"
+
+
+def load_operator():
+    scripts_path = str(SCRIPTS)
+    if scripts_path not in sys.path:
+        sys.path.insert(0, scripts_path)
+    spec = importlib.util.spec_from_file_location(
+        "provision_regional_grb_buildings_test",
+        SCRIPTS / "provision_regional_grb_buildings.py",
+    )
+    assert spec is not None
+    assert spec.loader is not None
+    module = importlib.util.module_from_spec(spec)
+    sys.modules[spec.name] = module
+    spec.loader.exec_module(module)
+    return module
+
+
+def feature(feature_id: str, polygon: Polygon) -> dict:
+    return {
+        "type": "Feature",
+        "id": feature_id,
+        "geometry": polygon.__geo_interface__,
+        "properties": {"UIDN": feature_id},
+    }
+
+
+def test_partition_assignment_is_deterministic_and_has_no_cross_member_duplicates() -> None:
+    operator = load_operator()
+    scopes = importlib.import_module("geographic_scopes")
+    alpha = scopes.ScopeMember("Alpha", "10001")
+    beta = scopes.ScopeMember("Beta", "10002")
+    scope = scopes.GeographicScope(
+        key="test-region",
+        display_name="Test region",
+        project_name="Test project",
+        project_region="Test",
+        area_name="Test operation boundary",
+        authority_name="Test authority",
+        authority_url="https://example.test/scope",
+        scope_type="policy_region",
+        limitation_message="Test limitation.",
+        members=(alpha, beta),
+    )
+    alpha_boundary = Polygon([(0, 0), (1, 0), (1, 1), (0, 1)])
+    beta_boundary = Polygon([(1, 0), (2, 0), (2, 1), (1, 1)])
+    members = [(alpha, alpha_boundary), (beta, beta_boundary)]
+    region = unary_union([alpha_boundary, beta_boundary])
+    alpha_building = feature("GBG.alpha", Polygon([(0.1, 0.1), (0.2, 0.1), (0.2, 0.2), (0.1, 0.2)]))
+    beta_building = feature("GBG.beta", Polygon([(1.2, 0.1), (1.3, 0.1), (1.3, 0.2), (1.2, 0.2)]))
+    crossing = feature("GBG.crossing", Polygon([(0.8, 0.3), (1.1, 0.3), (1.1, 0.5), (0.8, 0.5)]))
+    page = ({"type": "FeatureCollection", "features": [alpha_building, beta_building, crossing]}, "https://example.test/grb")
+
+    alpha_features, alpha_summary = operator.build_partition_features(
+        [page], member=alpha, members=members, regional_boundary=region, scope=scope, max_features=10
+    )
+    beta_features, beta_summary = operator.build_partition_features(
+        [page], member=beta, members=members, regional_boundary=region, scope=scope, max_features=10
+    )
+
+    assert {item["id"] for item in alpha_features} == {"GBG.alpha", "GBG.crossing"}
+    assert {item["id"] for item in beta_features} == {"GBG.beta"}
+    assert alpha_summary["reference_truncated"] is False
+    assert beta_summary["reference_truncated"] is False
+    assert alpha_features[1]["properties"]["partition_assignment"] == "maximum_boundary_intersection"
+
+
+def test_combined_artifact_streams_partitions_and_rejects_duplicate_source_ids(tmp_path: Path) -> None:
+    operator = load_operator()
+    scope = importlib.import_module("geographic_scopes").KEMPEN_TRANSPORT_REGION_SCOPE
+    first = tmp_path / "first.geojson"
+    second = tmp_path / "second.geojson"
+    first.write_text(json.dumps({"type": "FeatureCollection", "features": [feature("GBG.1", Polygon([(4, 51), (4.01, 51), (4.01, 51.01), (4, 51.01)]))]}), encoding="utf-8")
+    second.write_text(json.dumps({"type": "FeatureCollection", "features": [feature("GBG.2", Polygon([(4.1, 51), (4.11, 51), (4.11, 51.01), (4.1, 51.01)]))]}), encoding="utf-8")
+    combined = tmp_path / "combined.geojson"
+
+    summary = operator.write_combined_artifact(
+        combined,
+        scope=scope,
+        observed_date=operator.date(2026, 7, 14),
+        partition_paths=[first, second],
+        expected_feature_count=2,
+    )
+
+    payload = json.loads(combined.read_text(encoding="utf-8"))
+    assert summary["feature_count"] == 2
+    assert summary["sha256"] == operator.sha256_file(combined)
+    assert [item["id"] for item in payload["features"]] == ["GBG.1", "GBG.2"]
+
+    second.write_text(json.dumps({"type": "FeatureCollection", "features": [feature("GBG.1", Polygon([(4.1, 51), (4.11, 51), (4.11, 51.01), (4.1, 51.01)]))]}), encoding="utf-8")
+    with pytest.raises(RuntimeError, match="Duplicate regional source feature"):
+        operator.write_combined_artifact(
+            combined,
+            scope=scope,
+            observed_date=operator.date(2026, 7, 14),
+            partition_paths=[first, second],
+            expected_feature_count=2,
+        )
+
+
+class FakeDb:
+    def __init__(self) -> None:
+        self.rows = []
+        self.flush_count = 0
+        self.expunge_count = 0
+
+    def add(self, row) -> None:
+        self.rows.append(row)
+
+    def flush(self) -> None:
+        self.flush_count += 1
+
+    def expunge(self, row) -> None:
+        assert row in self.rows
+        self.expunge_count += 1
+
+
+def test_partition_persistence_batches_rows_and_guards_source_identity(tmp_path: Path) -> None:
+    first = tmp_path / "one.geojson"
+    second = tmp_path / "two.geojson"
+    first.write_text(json.dumps({"type": "FeatureCollection", "features": [feature("GBG.1", Polygon([(4, 51), (4.01, 51), (4.01, 51.01), (4, 51.01)]))]}), encoding="utf-8")
+    second.write_text(json.dumps({"type": "FeatureCollection", "features": [feature("GBG.2", Polygon([(4.1, 51), (4.11, 51), (4.11, 51.01), (4.1, 51.01)]))]}), encoding="utf-8")
+    db = FakeDb()
+
+    persisted = VectorFeatureService.persist_geojson_partitions(
+        db,
+        uuid4(),
+        [first, second],
+        feature_class="buildings",
+        batch_size=1,
+    )
+
+    assert persisted == 2
+    assert db.flush_count == 2
+    assert db.expunge_count == 2
+    assert {row.source_feature_id for row in db.rows} == {"GBG.1", "GBG.2"}
+
+    second.write_text(first.read_text(encoding="utf-8"), encoding="utf-8")
+    with pytest.raises(AppError) as error:
+        VectorFeatureService.persist_geojson_partitions(FakeDb(), uuid4(), [first, second])
+    assert error.value.code == "DUPLICATE_SOURCE_FEATURE"
+
+
+def test_storage_service_copies_large_artifacts_without_loading_them_as_upload_bytes(tmp_path: Path, monkeypatch) -> None:
+    source = tmp_path / "source.geojson"
+    source.write_bytes((b"0123456789abcdef" * 1024 * 1024) + b"tail")
+    storage_root = tmp_path / "storage"
+    monkeypatch.setattr(StorageService, "_base_dir", staticmethod(lambda: storage_root))
+
+    metadata = StorageService.persist_dataset_file_from_path(
+        "project",
+        "dataset",
+        "vector",
+        "regional.geojson",
+        source,
+        "application/geo+json",
+    )
+
+    stored = Path(metadata["storage_path"])
+    assert stored.read_bytes() == source.read_bytes()
+    assert metadata["size_bytes"] == source.stat().st_size
+    assert metadata["checksum_sha256"] == load_operator().sha256_file(source)
+
+
+def test_regional_operator_is_packaged_documented_and_uses_service_boundaries() -> None:
+    dockerfile = (ROOT / "deploy/unraid/Dockerfile.all-in-one").read_text(encoding="utf-8")
+    readiness = (ROOT / "scripts/run_readiness_check.sh").read_text(encoding="utf-8")
+    operator = (SCRIPTS / "provision_regional_grb_buildings.py").read_text(encoding="utf-8")
+    dataset_service = (ROOT / "backend/app/services/dataset_service.py").read_text(encoding="utf-8")
+
+    assert "provision_regional_grb_buildings.py" in dockerfile
+    assert "py_compile scripts/provision_regional_grb_buildings.py" in readiness
+    assert "DatasetService.import_partitioned_vector_artifact" in operator
+    assert "VectorFeatureService.persist_geojson_partitions" in dataset_service
+    assert "insert into vector_features" not in operator.lower()
+    assert "db.add(VectorFeature" not in operator
diff --git a/deploy/unraid/Dockerfile.all-in-one b/deploy/unraid/Dockerfile.all-in-one
index c35db4ee..55058bea 100644
--- a/deploy/unraid/Dockerfile.all-in-one
+++ b/deploy/unraid/Dockerfile.all-in-one
@@ -79,6 +79,7 @@ COPY scripts/provision_mol_historical_landuse.py /app/scripts/provision_mol_hist
 COPY scripts/provision_official_landuse_timeseries.py /app/scripts/provision_official_landuse_timeseries.py
 COPY scripts/geographic_scopes.py /app/scripts/geographic_scopes.py
 COPY scripts/provision_geographic_scope.py /app/scripts/provision_geographic_scope.py
+COPY scripts/provision_regional_grb_buildings.py /app/scripts/provision_regional_grb_buildings.py
 COPY scripts/export_operator_yolo_tile_dataset.py /app/scripts/export_operator_yolo_tile_dataset.py
 COPY scripts/audit_operator_yolo_dataset_quality.py /app/scripts/audit_operator_yolo_dataset_quality.py
 COPY scripts/render_operator_yolo_label_qa_contact_sheets.py /app/scripts/render_operator_yolo_label_qa_contact_sheets.py
diff --git a/docs/CODEX_EXECUTION_LOG.md b/docs/CODEX_EXECUTION_LOG.md
index c333bc71..5369a485 100644
--- a/docs/CODEX_EXECUTION_LOG.md
+++ b/docs/CODEX_EXECUTION_LOG.md
@@ -7884,3 +7884,20 @@ Live operational proof:
 
 Next:
 - Build bounded, idempotent regional theme ingestion in municipality-sized partitions, starting with current buildings and retaining per-member provenance before exposing any Kempen-wide metric in the explorer.
+
+## Sprint 190 Regional Kempen GRB buildings (2026-07-14)
+
+Implemented:
+- Added `provision_regional_grb_buildings.py` for the approved 28-member scope, with complete OGC pagination, per-member safety caps, resumable checksummed artifacts and an explicit total cap.
+- Added deterministic cross-boundary ownership by maximum member intersection and NIS-code tie breaking, preventing duplicate source identities without cutting the final geometry at internal municipality borders.
+- Added streaming managed-artifact copy plus batch-wise partition indexing through DatasetService and VectorFeatureService. One regional Dataset remains compatible with existing viewport, selection, QA and export ownership boundaries.
+- Added exact manifest/index count enforcement, duplicate source-feature rejection and rollback/storage cleanup on import failure.
+- Kept fetching operator-triggered and local service import container-only. No startup fetch, browser provider fetch, direct vector-feature SQL or public API change was introduced.
+
+Validation before deployment:
+- Backend/application and operator compilation passed.
+- New focused operator/storage/batch-index suite passed 5 tests.
+- Dataset, vector persistence, Mol municipality and Kempen scope regression selection passed 37 tests.
+
+Next:
+- Run the complete readiness gate, deploy to Tower, execute the full 28-partition source job, verify exact PostGIS counts and perform full-region plus rectangle browser selections.
diff --git a/docs/DATABASE_IMPLEMENTATION_PLAN.md b/docs/DATABASE_IMPLEMENTATION_PLAN.md
index 801f5d95..0ef3a246 100644
--- a/docs/DATABASE_IMPLEMENTATION_PLAN.md
+++ b/docs/DATABASE_IMPLEMENTATION_PLAN.md
@@ -224,6 +224,16 @@ Provider integration is a contract layer only in Sprint 7B. Providers do not wri
 
 GRB and OSM live imports are intentionally `not_configured` in Sprint 7B. Manual and fixture reference datasets use existing upload and fixture flows.
 
+Large explicit operator imports may invoke DatasetService directly inside the
+backend container when a single multipart upload would require loading the
+entire regional artifact into memory. The regional GRB building operator still
+creates a normal Dataset and DatasetVersion and delegates every queryable row
+to VectorFeatureService. It copies the immutable combined artifact in a
+stream, loads one retained municipality partition at a time, flushes bounded
+feature batches and commits only when the indexed count matches the manifest.
+It does not expose a direct SQL/provider write path and does not alter the
+public provider endpoint's `not_configured` status.
+
 ## Geometry normalization
 
 - User-drawn polygons arrive as EPSG:4326.
diff --git a/docs/DATASET_STRATEGY.md b/docs/DATASET_STRATEGY.md
index 30351904..dc77bdd8 100644
--- a/docs/DATASET_STRATEGY.md
+++ b/docs/DATASET_STRATEGY.md
@@ -106,6 +106,24 @@ boundaries.
   no interactive request or application startup may download the entire
   region or silently substitute missing data.
 
+### Regional building partition policy
+
+The current GRB building snapshot is fetched in one resumable partition per
+registered municipality. Partitions are source/provenance artifacts, not 28
+independent user-facing layers. A building intersecting more than one member
+boundary is owned by the member with the largest intersection area; the lower
+NIS code resolves exact ties. The retained geometry is clipped to the regional
+union, preserving complete in-scope coverage while guaranteeing one source ID
+per regional dataset.
+
+After all partitions pass validation, one combined immutable GeoJSON artifact
+is copied into managed storage and indexed in bounded batches through the
+normal DatasetService/VectorFeatureService ownership boundary. The resulting
+reference dataset uses `source_name=grb`, `reference_layer_name=buildings`,
+`coverage_scope=kempen-transport-region` and an exact `feature_count`
+selection aggregation. Existing viewport and bbox selection APIs remain the
+only interactive delivery path.
+
 ## User-uploaded raster strategy
 V1 must support controlled local datasets because public raster access and model compatibility can be difficult.
 
diff --git a/docs/DATA_SOURCES.md b/docs/DATA_SOURCES.md
index 0e17bb33..5adc04c1 100644
--- a/docs/DATA_SOURCES.md
+++ b/docs/DATA_SOURCES.md
@@ -52,6 +52,21 @@ De scope-operator haalt geen thematische gegevens op. GRB, bevolking en
 landgebruik volgen als afzonderlijke begrensde imports; ontbrekende regionale
 thema's blijven zichtbaar onbeschikbaar.
 
+De actuele regionale gebouwlaag wordt afzonderlijk opgehaald door
+`scripts/provision_regional_grb_buildings.py`. De operator bevraagt
+`GRB/GBG` per officiële gemeentegrens, volgt alle pagina's en weigert elk
+veiligheidslimiet als truncatie te verbergen. Grensoverschrijdende objecten
+worden aan één gemeentepartitie toegewezen op basis van de grootste overlap;
+de uiteindelijke geometrie wordt uitsluitend tegen de volledige regiogrens
+gesneden. Daardoor bevat het regionale dataset elke GRB-bronidentiteit exact
+één keer.
+
+De 28 bronpartities, paginabronnen, aantallen en checksums blijven als
+operatorbewijs bewaard. GeoIntel indexeert ze batchgewijs als één regionaal
+referentiedataset via DatasetService en VectorFeatureService. De publieke GRB
+provider blijft `not_configured`: dit is een bewuste operatorrun en geen live
+download vanuit een browseractie of applicatiestart.
+
 ### Mol population history
 
 `scripts/provision_mol_population_history.py` imports official Statbel
diff --git a/scripts/README.md b/scripts/README.md
index 46375c10..6c1e1230 100644
--- a/scripts/README.md
+++ b/scripts/README.md
@@ -1286,6 +1286,39 @@ This command provisions boundaries only. Regional buildings, population,
 land use, roads, water and parcels must be added by bounded source operators;
 missing themes remain unavailable and are never filled with synthetic values.
 
+### Regional Kempen GRB buildings
+
+After the scope foundation exists, fetch and persist the complete current GRB
+`GBG` building layer for the approved region:
+
+```bash
+docker exec -it geointel python3 /app/scripts/provision_regional_grb_buildings.py \
+  --scope kempen-transport-region
+```
+
+Use `--fetch-only` to build and inspect artifacts without touching the
+database. The operator uses one bounded request/pagination sequence per
+official municipality and writes resumable partitions below
+`/app/storage/operator-data/regional-themes/kempen-transport-region/buildings/<date>`.
+Every source feature is assigned to exactly one partition using its largest
+intersection with the member boundaries; exact ties use the lowest NIS code.
+This avoids duplicate building counts at shared borders while retaining the
+feature clipped only to the complete regional scope.
+
+Once all 28 partitions are complete, the operator streams one combined
+GeoJSON artifact and invokes `DatasetService`/`VectorFeatureService` inside
+the container. Features are indexed in batches as one normal regional
+reference dataset, so the existing viewport and rectangle-selection paths do
+not need a parallel API. The service import is intentionally local-only and
+refuses a remote backend URL. Repeat runs reuse checksummed artifacts and the
+persisted dataset; use a new `--observed-date` for a newer immutable snapshot.
+
+Safety limits can be adjusted explicitly with
+`--max-features-per-member`, `--max-total-features`, `--page-limit` and
+`--batch-size`. Exceeding a limit fails the run instead of producing a
+truncated dataset. The command never runs during startup or an interactive
+map query.
+
 ## Tower deployment
 
 Push the local branch to Gitea, then rebuild the Unraid/Tower Docker runtime:
diff --git a/scripts/provision_regional_grb_buildings.py b/scripts/provision_regional_grb_buildings.py
new file mode 100644
index 00000000..813345fa
--- /dev/null
+++ b/scripts/provision_regional_grb_buildings.py
@@ -0,0 +1,743 @@
+"""Provision one queryable GRB building dataset for an approved regional scope.
+
+Source requests are partitioned by official municipality boundaries and can be
+resumed per partition. The retained partition artifacts are then indexed as one
+Dataset through DatasetService and VectorFeatureService. This operator never
+runs during application startup and never writes directly to vector_features.
+"""
+
+from __future__ import annotations
+
+import argparse
+import hashlib
+import json
+import os
+import sys
+import unicodedata
+from datetime import date, datetime, time, timezone
+from pathlib import Path
+from typing import Any, Iterable
+from urllib.parse import urlparse
+from uuid import UUID
+
+import requests
+from requests.adapters import HTTPAdapter
+from shapely.geometry import GeometryCollection, MultiPolygon, Polygon, mapping, shape
+from shapely.ops import unary_union
+from shapely.validation import make_valid
+from urllib3.util.retry import Retry
+
+from geographic_scopes import GEOGRAPHIC_SCOPES, GeographicScope, ScopeMember
+from provision_geographic_scope import fetch_scope_members
+
+
+GRB_GBG_ITEMS_URL = "https://geo.api.vlaanderen.be/GRB/ogc/features/v1/collections/GBG/items"
+GRB_ATTRIBUTION = "Bron: Grootschalig Referentie Bestand Vlaanderen, Digitaal Vlaanderen"
+DEFAULT_SCOPE_KEY = "kempen-transport-region"
+DEFAULT_OUTPUT_ROOT = Path("/app/storage/operator-data/regional-themes")
+DEFAULT_API_URL = "http://127.0.0.1:8000"
+DEFAULT_PAGE_LIMIT = 1000
+DEFAULT_MAX_FEATURES_PER_MEMBER = 100000
+DEFAULT_MAX_TOTAL_FEATURES = 1500000
+GEOJSON_CRS = {"type": "name", "properties": {"name": "EPSG:4326"}}
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Provision municipality-partitioned regional GRB buildings.")
+    parser.add_argument("--scope", choices=sorted(GEOGRAPHIC_SCOPES), default=DEFAULT_SCOPE_KEY)
+    parser.add_argument("--observed-date", type=date.fromisoformat, default=date.today())
+    parser.add_argument("--base-url", default=os.environ.get("GEOINTEL_INTERNAL_API_URL", DEFAULT_API_URL))
+    parser.add_argument(
+        "--output-root",
+        type=Path,
+        default=Path(os.environ.get("GEOINTEL_REGIONAL_THEME_OUTPUT_ROOT", DEFAULT_OUTPUT_ROOT)),
+    )
+    parser.add_argument("--page-limit", type=int, default=DEFAULT_PAGE_LIMIT)
+    parser.add_argument("--max-features-per-member", type=int, default=DEFAULT_MAX_FEATURES_PER_MEMBER)
+    parser.add_argument("--max-total-features", type=int, default=DEFAULT_MAX_TOTAL_FEATURES)
+    parser.add_argument("--request-timeout", type=int, default=180)
+    parser.add_argument("--api-timeout", type=int, default=180)
+    parser.add_argument("--batch-size", type=int, default=1000)
+    parser.add_argument("--force", action="store_true", help="Refetch every municipality partition for this date.")
+    parser.add_argument("--fetch-only", action="store_true", help="Build and validate artifacts without persistence.")
+    return parser.parse_args()
+
+
+def utc_now() -> str:
+    return datetime.now(timezone.utc).isoformat()
+
+
+def observed_at(value: date) -> datetime:
+    return datetime.combine(value, time.min, tzinfo=timezone.utc)
+
+
+def safe_slug(value: str) -> str:
+    normalized = unicodedata.normalize("NFKD", value).encode("ascii", "ignore").decode("ascii")
+    return "-".join(part for part in "".join(char.lower() if char.isalnum() else " " for char in normalized).split())
+
+
+def sha256_file(path: Path) -> str:
+    digest = hashlib.sha256()
+    with path.open("rb") as handle:
+        for chunk in iter(lambda: handle.read(8 * 1024 * 1024), b""):
+            digest.update(chunk)
+    return digest.hexdigest()
+
+
+def write_json_atomic(path: Path, payload: dict[str, Any], *, pretty: bool = False) -> None:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    temporary = path.with_suffix(f"{path.suffix}.partial")
+    temporary.write_text(
+        json.dumps(
+            payload,
+            ensure_ascii=False,
+            indent=2 if pretty else None,
+            separators=None if pretty else (",", ":"),
+            sort_keys=pretty,
+        ),
+        encoding="utf-8",
+    )
+    temporary.replace(path)
+
+
+def normalize_polygonal(geometry):
+    if geometry is None or geometry.is_empty:
+        return None
+    if not geometry.is_valid:
+        geometry = make_valid(geometry)
+    if isinstance(geometry, (Polygon, MultiPolygon)):
+        return geometry
+    if isinstance(geometry, GeometryCollection):
+        polygons = [part for part in geometry.geoms if isinstance(part, (Polygon, MultiPolygon)) and not part.is_empty]
+        if polygons:
+            merged = unary_union(polygons)
+            return merged if isinstance(merged, (Polygon, MultiPolygon)) and not merged.is_empty else None
+    return None
+
+
+def build_source_session() -> requests.Session:
+    retry = Retry(
+        total=5,
+        connect=5,
+        read=5,
+        status=5,
+        backoff_factor=1.0,
+        status_forcelist=(429, 500, 502, 503, 504),
+        allowed_methods=frozenset({"GET"}),
+        raise_on_status=True,
+    )
+    adapter = HTTPAdapter(max_retries=retry)
+    session = requests.Session()
+    session.headers.update({"User-Agent": "GeoIntel-Regional-GRB-Buildings-Operator/1.0"})
+    session.mount("https://", adapter)
+    session.mount("http://", adapter)
+    return session
+
+
+def next_page_url(payload: dict[str, Any]) -> str | None:
+    links = payload.get("links") or []
+    for link in links:
+        if link.get("rel") == "next" and "geo+json" in str(link.get("type", "")).lower():
+            return str(link["href"])
+    for link in links:
+        if link.get("rel") == "next" and link.get("href"):
+            return str(link["href"])
+    return None
+
+
+def iter_grb_pages(
+    session: requests.Session,
+    bounds: tuple[float, float, float, float],
+    *,
+    page_limit: int,
+    timeout: int,
+) -> Iterable[tuple[dict[str, Any], str]]:
+    params = {
+        "f": "application/geo+json",
+        "limit": str(page_limit),
+        "bbox": ",".join(f"{value:.8f}" for value in bounds),
+    }
+    url: str | None = GRB_GBG_ITEMS_URL
+    seen_urls: set[str] = set()
+    first_request = True
+    while url:
+        if url in seen_urls:
+            raise RuntimeError(f"GRB pagination loop detected: {url}")
+        seen_urls.add(url)
+        response = session.get(url, params=params if first_request else None, timeout=timeout)
+        first_request = False
+        response.raise_for_status()
+        payload = response.json()
+        if payload.get("type") != "FeatureCollection":
+            raise RuntimeError("GRB returned a non-FeatureCollection response")
+        yield payload, response.url
+        url = next_page_url(payload)
+
+
+def build_member_geometries(
+    scope: GeographicScope,
+    source_features: list[dict[str, Any]],
+) -> tuple[list[tuple[ScopeMember, Any]], Any]:
+    if len(source_features) != len(scope.members):
+        raise RuntimeError(f"Expected {len(scope.members)} scope boundaries, received {len(source_features)}")
+    members: list[tuple[ScopeMember, Any]] = []
+    for member, source_feature in zip(scope.members, source_features, strict=True):
+        properties = source_feature.get("properties") or {}
+        if str(properties.get("NISCODE") or "") != member.nis_code:
+            raise RuntimeError(f"VRBG scope member mismatch for {member.name}")
+        geometry = normalize_polygonal(shape(source_feature.get("geometry")))
+        if geometry is None:
+            raise RuntimeError(f"Invalid official boundary for {member.name}")
+        members.append((member, geometry))
+    regional_boundary = normalize_polygonal(unary_union([geometry for _, geometry in members]))
+    if regional_boundary is None:
+        raise RuntimeError("The regional scope union is invalid")
+    return members, regional_boundary
+
+
+def bounds_overlap(left: tuple[float, float, float, float], right: tuple[float, float, float, float]) -> bool:
+    return left[0] <= right[2] and left[2] >= right[0] and left[1] <= right[3] and left[3] >= right[1]
+
+
+def assign_owner_nis(source_geometry, members: list[tuple[ScopeMember, Any]]) -> str | None:
+    candidates: list[tuple[float, str]] = []
+    source_bounds = source_geometry.bounds
+    for member, boundary in members:
+        if not bounds_overlap(source_bounds, boundary.bounds) or not source_geometry.intersects(boundary):
+            continue
+        intersection = source_geometry.intersection(boundary)
+        if not intersection.is_empty and intersection.area > 0:
+            candidates.append((float(intersection.area), member.nis_code))
+    if not candidates:
+        return None
+    candidates.sort(key=lambda item: (-item[0], item[1]))
+    return candidates[0][1]
+
+
+def build_partition_features(
+    pages: Iterable[tuple[dict[str, Any], str]],
+    *,
+    member: ScopeMember,
+    members: list[tuple[ScopeMember, Any]],
+    regional_boundary,
+    scope: GeographicScope,
+    max_features: int,
+) -> tuple[list[dict[str, Any]], dict[str, Any]]:
+    features: list[dict[str, Any]] = []
+    source_urls: list[str] = []
+    seen_ids: set[str] = set()
+    bbox_feature_count = 0
+    assigned_elsewhere_count = 0
+    outside_scope_count = 0
+    clipped_to_scope_count = 0
+
+    for payload, source_url in pages:
+        source_urls.append(source_url)
+        for source_feature in payload.get("features") or []:
+            bbox_feature_count += 1
+            feature_id = str(source_feature.get("id") or "")
+            if not feature_id:
+                feature_id = hashlib.sha256(
+                    json.dumps(source_feature.get("geometry"), sort_keys=True).encode("utf-8")
+                ).hexdigest()
+            if feature_id in seen_ids:
+                continue
+            seen_ids.add(feature_id)
+            source_geometry = normalize_polygonal(shape(source_feature.get("geometry")))
+            if source_geometry is None or not source_geometry.intersects(regional_boundary):
+                outside_scope_count += 1
+                continue
+            owner_nis = assign_owner_nis(source_geometry, members)
+            if owner_nis != member.nis_code:
+                assigned_elsewhere_count += 1
+                continue
+            clipped_geometry = source_geometry
+            if not source_geometry.within(regional_boundary):
+                clipped_geometry = normalize_polygonal(source_geometry.intersection(regional_boundary))
+                clipped_to_scope_count += 1
+            if clipped_geometry is None:
+                outside_scope_count += 1
+                continue
+            if len(features) >= max_features:
+                raise RuntimeError(
+                    f"{member.name} exceeds --max-features-per-member={max_features}; refusing truncated output"
+                )
+            properties = dict(source_feature.get("properties") or {})
+            properties.update(
+                {
+                    "source_name": "grb",
+                    "source_feature_id": feature_id,
+                    "reference_layer_name": "buildings",
+                    "layer_type": "building",
+                    "theme": "buildings",
+                    "authority_level": "authoritative",
+                    "coverage_scope": scope.key,
+                    "scope_type": scope.scope_type,
+                    "partition_scope": "municipality",
+                    "partition_municipality": member.name,
+                    "partition_nis_code": member.nis_code,
+                    "partition_assignment": "maximum_boundary_intersection",
+                    "clipped_to_regional_scope": clipped_geometry is not source_geometry,
+                    "attribution": GRB_ATTRIBUTION,
+                }
+            )
+            features.append(
+                {
+                    "type": "Feature",
+                    "id": feature_id,
+                    "geometry": mapping(clipped_geometry),
+                    "properties": properties,
+                }
+            )
+
+    if not features:
+        raise RuntimeError(f"No GRB buildings were assigned to {member.name}")
+    return features, {
+        "municipality": member.name,
+        "nis_code": member.nis_code,
+        "pages_fetched": len(source_urls),
+        "source_urls": source_urls,
+        "bbox_feature_count": bbox_feature_count,
+        "feature_count": len(features),
+        "assigned_elsewhere_count": assigned_elsewhere_count,
+        "outside_scope_count": outside_scope_count,
+        "clipped_to_scope_count": clipped_to_scope_count,
+        "reference_truncated": False,
+    }
+
+
+def partition_filename(member: ScopeMember, observed_date: date) -> str:
+    return f"{member.nis_code}_{safe_slug(member.name)}_grb_buildings_{observed_date.isoformat()}.geojson"
+
+
+def combined_filename(scope: GeographicScope, observed_date: date) -> str:
+    return f"grb_buildings_{scope.key.replace('-', '_')}_{observed_date.isoformat()}.geojson"
+
+
+def write_partition(
+    path: Path,
+    *,
+    scope: GeographicScope,
+    member: ScopeMember,
+    features: list[dict[str, Any]],
+    generated_at: str,
+    source_url: str,
+) -> None:
+    write_json_atomic(
+        path,
+        {
+            "type": "FeatureCollection",
+            "name": f"GRB buildings - {member.name} partition of {scope.display_name}",
+            "crs": GEOJSON_CRS,
+            "features": features,
+            "source": "Digitaal Vlaanderen GRB OGC API Features collection GBG",
+            "source_url": source_url,
+            "attribution": GRB_ATTRIBUTION,
+            "coverage_scope": scope.key,
+            "partition_municipality": member.name,
+            "partition_nis_code": member.nis_code,
+            "generated_at": generated_at,
+        },
+    )
+
+
+def write_combined_artifact(
+    path: Path,
+    *,
+    scope: GeographicScope,
+    observed_date: date,
+    partition_paths: list[Path],
+    expected_feature_count: int,
+) -> dict[str, Any]:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    temporary = path.with_suffix(f"{path.suffix}.partial")
+    header = {
+        "type": "FeatureCollection",
+        "name": f"GRB buildings - {scope.display_name}",
+        "crs": GEOJSON_CRS,
+        "source": "Digitaal Vlaanderen GRB OGC API Features collection GBG",
+        "source_url": GRB_GBG_ITEMS_URL,
+        "attribution": GRB_ATTRIBUTION,
+        "coverage_scope": scope.key,
+        "scope_type": scope.scope_type,
+        "partition_count": len(partition_paths),
+        "partition_strategy": "municipality_bbox_maximum_boundary_intersection",
+        "observed_at": observed_date.isoformat(),
+    }
+    encoded_header = json.dumps(header, ensure_ascii=False, separators=(",", ":"))
+    seen_ids: set[str] = set()
+    written = 0
+    first = True
+    with temporary.open("w", encoding="utf-8", newline="") as output:
+        output.write(encoded_header[:-1])
+        output.write(',"features":[')
+        for partition_path in partition_paths:
+            payload = json.loads(partition_path.read_text(encoding="utf-8"))
+            if payload.get("type") != "FeatureCollection" or not isinstance(payload.get("features"), list):
+                raise RuntimeError(f"Invalid partition artifact: {partition_path}")
+            for feature in payload["features"]:
+                feature_id = str(feature.get("id") or (feature.get("properties") or {}).get("source_feature_id") or "")
+                if not feature_id:
+                    raise RuntimeError(f"Partition feature without source identity in {partition_path.name}")
+                if feature_id in seen_ids:
+                    raise RuntimeError(f"Duplicate regional source feature {feature_id} in {partition_path.name}")
+                seen_ids.add(feature_id)
+                if not first:
+                    output.write(",")
+                output.write(json.dumps(feature, ensure_ascii=False, separators=(",", ":")))
+                first = False
+                written += 1
+        output.write("]}")
+    if written != expected_feature_count:
+        temporary.unlink(missing_ok=True)
+        raise RuntimeError(f"Expected {expected_feature_count} combined features, wrote {written}")
+    temporary.replace(path)
+    return {
+        "feature_count": written,
+        "size_bytes": path.stat().st_size,
+        "sha256": sha256_file(path),
+    }
+
+
+def reusable_manifest(
+    manifest_path: Path,
+    artifact_path: Path,
+    partition_dir: Path,
+    expected_member_count: int,
+) -> dict[str, Any] | None:
+    if not manifest_path.is_file() or not artifact_path.is_file():
+        return None
+    manifest = json.loads(manifest_path.read_text(encoding="utf-8"))
+    if manifest.get("status") != "complete" or len(manifest.get("partitions") or []) != expected_member_count:
+        return None
+    if sha256_file(artifact_path) != manifest.get("artifact_sha256"):
+        return None
+    for summary in manifest["partitions"]:
+        path = partition_dir / str(summary.get("filename") or "")
+        if not path.is_file() or sha256_file(path) != summary.get("sha256"):
+            return None
+    return manifest
+
+
+def prepare_artifacts(
+    args: argparse.Namespace,
+    scope: GeographicScope,
+) -> tuple[Path, list[Path], Path, dict[str, Any]]:
+    if args.page_limit <= 0 or args.max_features_per_member <= 0 or args.max_total_features <= 0:
+        raise RuntimeError("Page and feature limits must be positive")
+    observation_dir = args.output_root / scope.key / "buildings" / args.observed_date.isoformat()
+    partition_dir = observation_dir / "partitions"
+    artifact_path = observation_dir / combined_filename(scope, args.observed_date)
+    manifest_path = observation_dir / "regional_buildings_manifest.json"
+    partition_dir.mkdir(parents=True, exist_ok=True)
+
+    if not args.force:
+        existing = reusable_manifest(manifest_path, artifact_path, partition_dir, len(scope.members))
+        if existing:
+            paths = [partition_dir / summary["filename"] for summary in existing["partitions"]]
+            return artifact_path, paths, manifest_path, existing
+
+    existing_manifest: dict[str, Any] = {}
+    if manifest_path.is_file() and not args.force:
+        existing_manifest = json.loads(manifest_path.read_text(encoding="utf-8"))
+    existing_by_nis = {
+        str(item.get("nis_code")): item
+        for item in existing_manifest.get("partitions") or []
+        if isinstance(item, dict)
+    }
+    generated_at = utc_now()
+    with build_source_session() as session:
+        source_features, vrbg_source_url = fetch_scope_members(session, scope, args.request_timeout)
+        members, regional_boundary = build_member_geometries(scope, source_features)
+        summaries: list[dict[str, Any]] = []
+        for member, boundary in members:
+            path = partition_dir / partition_filename(member, args.observed_date)
+            reusable = existing_by_nis.get(member.nis_code)
+            if (
+                reusable
+                and path.is_file()
+                and reusable.get("filename") == path.name
+                and reusable.get("sha256") == sha256_file(path)
+            ):
+                summaries.append(reusable)
+                continue
+            features, summary = build_partition_features(
+                iter_grb_pages(
+                    session,
+                    boundary.bounds,
+                    page_limit=args.page_limit,
+                    timeout=args.request_timeout,
+                ),
+                member=member,
+                members=members,
+                regional_boundary=regional_boundary,
+                scope=scope,
+                max_features=args.max_features_per_member,
+            )
+            write_partition(
+                path,
+                scope=scope,
+                member=member,
+                features=features,
+                generated_at=generated_at,
+                source_url=summary["source_urls"][0],
+            )
+            summary.update({"filename": path.name, "size_bytes": path.stat().st_size, "sha256": sha256_file(path)})
+            summaries.append(summary)
+            progress = {
+                "schema_version": 1,
+                "status": "in_progress",
+                "scope": scope.key,
+                "theme": "buildings",
+                "observed_at": args.observed_date.isoformat(),
+                "generated_at": generated_at,
+                "vrbg_source_url": vrbg_source_url,
+                "grb_source_url": GRB_GBG_ITEMS_URL,
+                "partitions": summaries,
+            }
+            write_json_atomic(manifest_path, progress, pretty=True)
+
+    total_features = sum(int(summary["feature_count"]) for summary in summaries)
+    if total_features > args.max_total_features:
+        raise RuntimeError(
+            f"Regional building count {total_features} exceeds --max-total-features={args.max_total_features}"
+        )
+    partition_paths = [partition_dir / summary["filename"] for summary in summaries]
+    artifact = write_combined_artifact(
+        artifact_path,
+        scope=scope,
+        observed_date=args.observed_date,
+        partition_paths=partition_paths,
+        expected_feature_count=total_features,
+    )
+    manifest = {
+        "schema_version": 1,
+        "status": "complete",
+        "scope": scope.key,
+        "scope_type": scope.scope_type,
+        "scope_authority_url": scope.authority_url,
+        "scope_limitation": scope.limitation_message,
+        "theme": "buildings",
+        "observed_at": args.observed_date.isoformat(),
+        "generated_at": generated_at,
+        "member_count": len(scope.members),
+        "feature_count": total_features,
+        "reference_truncated": False,
+        "partition_strategy": "municipality_bbox_maximum_boundary_intersection",
+        "partition_assignment_rule": "largest geometry intersection; NIS code resolves exact ties",
+        "vrbg_source_url": vrbg_source_url,
+        "grb_source_url": GRB_GBG_ITEMS_URL,
+        "artifact_filename": artifact_path.name,
+        "artifact_size_bytes": artifact["size_bytes"],
+        "artifact_sha256": artifact["sha256"],
+        "bounds_json": {
+            "min_x": float(regional_boundary.bounds[0]),
+            "min_y": float(regional_boundary.bounds[1]),
+            "max_x": float(regional_boundary.bounds[2]),
+            "max_y": float(regional_boundary.bounds[3]),
+        },
+        "partitions": summaries,
+        "attribution": GRB_ATTRIBUTION,
+    }
+    write_json_atomic(manifest_path, manifest, pretty=True)
+    return artifact_path, partition_paths, manifest_path, manifest
+
+
+def response_data(response: requests.Response) -> Any:
+    try:
+        payload = response.json()
+    except ValueError as exc:
+        raise RuntimeError(f"GeoIntel API returned non-JSON ({response.status_code}): {response.text[:500]}") from exc
+    if not response.ok:
+        raise RuntimeError(f"GeoIntel API request failed ({response.status_code}): {json.dumps(payload)[:1000]}")
+    if not isinstance(payload, dict) or "data" not in payload:
+        raise RuntimeError("GeoIntel API response does not use the canonical data envelope")
+    return payload["data"]
+
+
+def list_paginated_items(session: requests.Session, url: str, timeout: int) -> list[dict[str, Any]]:
+    items: list[dict[str, Any]] = []
+    offset = 0
+    while True:
+        page = response_data(session.get(url, params={"limit": 200, "offset": offset}, timeout=timeout))
+        page_items = list(page.get("items") or [])
+        items.extend(page_items)
+        total = int(page.get("total") or len(items))
+        if not page_items or len(items) >= total:
+            return items
+        offset += len(page_items)
+
+
+def ensure_backend_path() -> None:
+    repository_root = Path(__file__).resolve().parents[1]
+    backend_root = repository_root / "backend" if (repository_root / "backend" / "app").is_dir() else repository_root
+    if str(backend_root) not in sys.path:
+        sys.path.insert(0, str(backend_root))
+
+
+def provision_dataset(
+    args: argparse.Namespace,
+    scope: GeographicScope,
+    artifact_path: Path,
+    partition_paths: list[Path],
+    manifest_path: Path,
+    manifest: dict[str, Any],
+) -> dict[str, Any]:
+    parsed_base = urlparse(args.base_url)
+    if parsed_base.hostname not in {"127.0.0.1", "localhost", "::1"}:
+        raise RuntimeError("Partitioned service import must run inside the GeoIntel container against its local backend")
+    with requests.Session() as session:
+        projects = list_paginated_items(session, f"{args.base_url.rstrip('/')}/api/v1/projects", args.api_timeout)
+        project = next((item for item in projects if item.get("name") == scope.project_name), None)
+        if not project:
+            raise RuntimeError("Regional scope project is missing; run provision_geographic_scope.py first")
+        project_id = str(project["id"])
+        areas = list_paginated_items(
+            session,
+            f"{args.base_url.rstrip('/')}/api/v1/projects/{project_id}/areas",
+            args.api_timeout,
+        )
+        area = next((item for item in areas if item.get("name") == scope.area_name), None)
+        if not area:
+            raise RuntimeError("Regional scope Area is missing; run provision_geographic_scope.py first")
+        datasets = list_paginated_items(
+            session,
+            f"{args.base_url.rstrip('/')}/api/v1/projects/{project_id}/datasets",
+            args.api_timeout,
+        )
+    existing = next((item for item in datasets if item.get("original_filename") == artifact_path.name), None)
+    if existing:
+        persisted_checksum = (existing.get("provenance_metadata") or {}).get("artifact_sha256")
+        if persisted_checksum and persisted_checksum != manifest["artifact_sha256"]:
+            raise RuntimeError(
+                f"Immutable dataset {artifact_path.name} checksum changed; use a new --observed-date for refreshed GRB data"
+            )
+        return {
+            "dataset_id": str(existing["id"]),
+            "feature_count": existing.get("feature_count"),
+            "reused": True,
+        }
+
+    ensure_backend_path()
+    from app.db.session import SessionLocal
+    from app.services.dataset_service import DatasetService
+
+    partition_checksums = {
+        summary["nis_code"]: summary["sha256"]
+        for summary in manifest["partitions"]
+    }
+    metadata_json = {
+        "feature_count": manifest["feature_count"],
+        "feature_geometry_count": manifest["feature_count"],
+        "geometry_types": ["MultiPolygon", "Polygon"],
+        "bounds_json": manifest["bounds_json"],
+        "approximate_area_m2": None,
+        "invalid_features": 0,
+        "z_dimension_feature_count": 0,
+        "canonical_storage_dimension": "2D",
+        "crs": "EPSG:4326",
+        "crs_assumed": False,
+        "extracted_at": manifest["generated_at"],
+    }
+    source_metadata = {
+        "provider": "Digitaal Vlaanderen",
+        "collection": "GRB/GBG",
+        "authority_level": "authoritative",
+        "theme": "buildings",
+        "layer_type": "regional_buildings",
+        "coverage_scope": scope.key,
+        "scope_type": scope.scope_type,
+        "scope_authority": scope.authority_name,
+        "scope_authority_url": scope.authority_url,
+        "scope_limitation": scope.limitation_message,
+        "member_count": len(scope.members),
+        "member_nis_codes": list(scope.nis_codes),
+        "feature_count": manifest["feature_count"],
+        "partition_count": len(partition_paths),
+        "partition_strategy": manifest["partition_strategy"],
+        "selection_aggregation": {
+            "method": "feature_count",
+            "label": "Gebouwen",
+            "unit": "objecten",
+            "is_estimate": False,
+        },
+        "attribution": GRB_ATTRIBUTION,
+    }
+    provenance_metadata = {
+        "operator_tool": "provision_regional_grb_buildings.py",
+        "operator_explicit_fetch": True,
+        "manifest_path": str(manifest_path),
+        "source_url": GRB_GBG_ITEMS_URL,
+        "artifact_sha256": manifest["artifact_sha256"],
+        "artifact_size_bytes": manifest["artifact_size_bytes"],
+        "partition_checksums": partition_checksums,
+        "partition_assignment_rule": manifest["partition_assignment_rule"],
+        "reference_truncated": False,
+    }
+    with SessionLocal() as db:
+        dataset = DatasetService.import_partitioned_vector_artifact(
+            db,
+            project_id=UUID(project_id),
+            area_id=UUID(str(area["id"])),
+            artifact_path=artifact_path,
+            partition_paths=partition_paths,
+            original_filename=artifact_path.name,
+            source="operator_official_import",
+            dataset_role="reference",
+            source_name="grb",
+            reference_layer_name="buildings",
+            metadata_json=metadata_json,
+            source_metadata=source_metadata,
+            provenance_metadata=provenance_metadata,
+            temporal_series_key=f"grb:buildings:{scope.key}",
+            observed_at=observed_at(args.observed_date),
+            temporal_granularity="snapshot",
+            source_version=args.observed_date.isoformat(),
+            batch_size=args.batch_size,
+        )
+    if dataset.checksum_sha256 != manifest["artifact_sha256"]:
+        raise RuntimeError("Managed dataset checksum differs from the retained regional artifact")
+    return {"dataset_id": str(dataset.id), "feature_count": dataset.feature_count, "reused": False}
+
+
+def main() -> int:
+    args = parse_args()
+    scope = GEOGRAPHIC_SCOPES[args.scope]
+    try:
+        artifact_path, partition_paths, manifest_path, manifest = prepare_artifacts(args, scope)
+        persistence = None if args.fetch_only else provision_dataset(
+            args,
+            scope,
+            artifact_path,
+            partition_paths,
+            manifest_path,
+            manifest,
+        )
+    except (OSError, RuntimeError, ValueError, KeyError, requests.RequestException) as exc:
+        print(json.dumps({"status": "error", "message": str(exc)}, ensure_ascii=False), file=sys.stderr)
+        return 1
+    print(
+        json.dumps(
+            {
+                "status": "ok",
+                "mode": "fetch_only" if args.fetch_only else "provisioned",
+                "scope": scope.key,
+                "theme": "buildings",
+                "observed_at": args.observed_date.isoformat(),
+                "member_count": manifest["member_count"],
+                "feature_count": manifest["feature_count"],
+                "artifact_size_bytes": manifest["artifact_size_bytes"],
+                "artifact_path": str(artifact_path),
+                "manifest_path": str(manifest_path),
+                "reference_truncated": manifest["reference_truncated"],
+                "persistence": persistence,
+            },
+            ensure_ascii=False,
+            indent=2,
+        )
+    )
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/scripts/run_readiness_check.sh b/scripts/run_readiness_check.sh
index 01942646..9017be9f 100755
--- a/scripts/run_readiness_check.sh
+++ b/scripts/run_readiness_check.sh
@@ -49,6 +49,7 @@ ${PYTHON_BIN} -m py_compile scripts/provision_mol_historical_landuse.py
 ${PYTHON_BIN} -m py_compile scripts/provision_official_landuse_timeseries.py
 ${PYTHON_BIN} -m py_compile scripts/geographic_scopes.py
 ${PYTHON_BIN} -m py_compile scripts/provision_geographic_scope.py
+${PYTHON_BIN} -m py_compile scripts/provision_regional_grb_buildings.py
 ${PYTHON_BIN} -m py_compile scripts/export_operator_yolo_dataset.py
 ${PYTHON_BIN} -m py_compile scripts/export_operator_yolo_tile_dataset.py
 ${PYTHON_BIN} -m py_compile scripts/audit_operator_yolo_dataset_quality.py