M31: verify RAG inventory and polish attention queue

backend/tests/test_knowledge.py

@@ -9,6 +9,7 @@ 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.procedures import iter_procedure_documents
 from app.services.knowledge.ragcore import RAGcoreKnowledgeProvider
 
 
@@ -198,6 +199,39 @@ def test_ragcore_status_separates_sync_report_from_unverifiable_index(
     assert status["statistics_state"] == "sync_reported"
 
 
+def test_ragcore_status_preserves_stronger_verified_index_evidence(client, ops_client, monkeypatch):
+    class VerifiedRagcoreProvider:
+        def health(self, language="en-GB"):
+            return KnowledgeHealth(
+                provider="ragcore",
+                available=True,
+                detail="ready and verified",
+                tenant="fleet-ops",
+                workspace="operations",
+                collection="internal-procedures",
+                document_count=11,
+                source_document_count=11,
+                reported_synced_document_count=None,
+                reported_failed_document_count=None,
+                last_sync_at=None,
+                statistics_state="verified",
+            )
+
+    monkeypatch.setattr(knowledge_router, "get_knowledge_provider", VerifiedRagcoreProvider)
+    settings = get_settings()
+    sync = client.post(
+        "/api/v1/integrations/n8n/procedures-sync-result",
+        json={"execution_id": "rag-verified-statistics-test", "synced": 33, "failed": 0},
+        headers={"X-Service-Token": settings.n8n_callback_token},
+    )
+    assert sync.status_code == 200
+
+    status = ops_client.get("/api/v1/knowledge/status?language=nl-BE").json()
+    assert status["document_count"] == 11
+    assert status["reported_synced_document_count"] == 33
+    assert status["statistics_state"] == "verified"
+
+
 class _FakeResponse:
     def __init__(self, status_code: int, body: dict):
         self.status_code = status_code
@@ -208,7 +242,14 @@ class _FakeResponse:
 
 
 class _FakeClient:
-    def __init__(self, get_response=None, post_response=None, post_responses=None, raise_on=None):
+    def __init__(
+        self,
+        get_response=None,
+        post_response=None,
+        post_responses=None,
+        raise_on=None,
+        get_handler=None,
+    ):
         self._get_response = get_response
         self._post_response = post_response
         # Maps a path (e.g. "/v1/search") to its own response, for tests that need
@@ -217,6 +258,7 @@ class _FakeClient:
         # existing single-endpoint test keeps working unchanged.
         self._post_responses = post_responses or {}
         self._raise_on = raise_on
+        self._get_handler = get_handler
 
     def __enter__(self):
         return self
@@ -224,9 +266,13 @@ class _FakeClient:
     def __exit__(self, *args):
         return False
 
-    def get(self, path):
+    def get(self, path, params=None, timeout=None):
         if self._raise_on == "get":
             raise httpx.ConnectError("no ragcore in this environment")
+        if self._get_handler is not None:
+            return self._get_handler(path, params, timeout)
+        if path == "/v1/documents":
+            return _FakeResponse(503, {})
         return self._get_response
 
     def post(self, path, json=None):
@@ -273,6 +319,50 @@ def test_ragcore_provider_health_reports_ready_status(monkeypatch):
     assert health.statistics_state == "not_reported"
 
 
+def test_ragcore_provider_verifies_published_documents_and_caches_count(monkeypatch):
+    provider = RAGcoreKnowledgeProvider()
+    monkeypatch.setattr(provider._settings, "ragcore_space_id", "space-1")
+    procedure_documents = [
+        document
+        for document in iter_procedure_documents(Path(provider._settings.knowledge_dir))
+        if document.language == "en-GB"
+    ]
+    documents = {document.source_id: document for document in procedure_documents}
+    lookups: list[str] = []
+
+    def get_handler(path, params, _timeout):
+        if path == "/health/ready":
+            return _FakeResponse(200, {"status": "ok"})
+        document = documents[params["source_id"]]
+        lookups.append(document.source_id)
+        return _FakeResponse(
+            200,
+            {
+                "items": [
+                    {
+                        "space_id": "space-1",
+                        "source_id": document.source_id,
+                        "external_id": f"{document.document_id}.md",
+                        "status": "active",
+                        "active_version": {
+                            "status": "published",
+                            "content_sha256": document.content_hash,
+                        },
+                    }
+                ]
+            },
+        )
+
+    monkeypatch.setattr(provider, "_client", lambda: _FakeClient(get_handler=get_handler))
+
+    first = provider.health("en-GB")
+    second = provider.health("en-GB")
+    assert first.document_count == 11
+    assert first.statistics_state == "verified"
+    assert second.document_count == 11
+    assert len(lookups) == 11
+
+
 def test_ragcore_provider_health_reports_degraded_status(monkeypatch):
     provider = RAGcoreKnowledgeProvider()
     monkeypatch.setattr(