M25: expose provenance-aware knowledge statistics

backend/tests/test_knowledge.py

@@ -5,7 +5,9 @@ from pathlib import Path
 
 import httpx
 
+from app.api.routers import knowledge as knowledge_router
 from app.core.config import get_settings
+from app.services.knowledge import KnowledgeHealth
 from app.services.knowledge.demo import DemoKnowledgeProvider
 from app.services.knowledge.ragcore import RAGcoreKnowledgeProvider
 
@@ -73,6 +75,8 @@ def test_demo_provider_health_reports_document_count():
     assert health.provider == "demo"
     assert health.available is True
     assert health.document_count == 11
+    assert health.source_document_count == 11
+    assert health.statistics_state == "verified"
 
 
 def test_demo_provider_health_reports_document_count_per_language():
@@ -154,6 +158,46 @@ def test_knowledge_status_endpoint(ops_client):
     assert response.json()["provider"] == "demo"
 
 
+def test_ragcore_status_separates_sync_report_from_unverifiable_index(
+    client, ops_client, monkeypatch
+):
+    class FakeRagcoreProvider:
+        def health(self, language="en-GB"):
+            return KnowledgeHealth(
+                provider="ragcore",
+                available=True,
+                detail="ready",
+                tenant="fleet-ops",
+                workspace="operations",
+                collection="internal-procedures",
+                document_count=None,
+                source_document_count=11,
+                reported_synced_document_count=None,
+                reported_failed_document_count=None,
+                last_sync_at=None,
+                statistics_state="not_reported",
+            )
+
+    monkeypatch.setattr(knowledge_router, "get_knowledge_provider", FakeRagcoreProvider)
+    settings = get_settings()
+    sync = client.post(
+        "/api/v1/integrations/n8n/procedures-sync-result",
+        json={"execution_id": "rag-statistics-test", "synced": 32, "failed": 1},
+        headers={"X-Service-Token": settings.n8n_callback_token},
+    )
+    assert sync.status_code == 200
+
+    response = ops_client.get("/api/v1/knowledge/status?language=nl-BE")
+    assert response.status_code == 200
+    status = response.json()
+    assert status["document_count"] is None
+    assert status["source_document_count"] == 11
+    assert status["reported_synced_document_count"] == 32
+    assert status["reported_failed_document_count"] == 1
+    assert status["last_sync_at"] is not None
+    assert status["statistics_state"] == "sync_reported"
+
+
 class _FakeResponse:
     def __init__(self, status_code: int, body: dict):
         self.status_code = status_code
@@ -224,6 +268,9 @@ def test_ragcore_provider_health_reports_ready_status(monkeypatch):
     health = provider.health()
     assert health.provider == "ragcore"
     assert health.available is True
+    assert health.document_count is None
+    assert health.source_document_count == 11
+    assert health.statistics_state == "not_reported"
 
 
 def test_ragcore_provider_health_reports_degraded_status(monkeypatch):