backend/src/main/java/org/raddatz/familienarchiv/document/DocumentController.java

@@ -3,7 +3,6 @@ package org.raddatz.familienarchiv.document;
 import java.io.IOException;
 import java.time.LocalDate;
 import java.util.ArrayList;
-import java.util.concurrent.TimeUnit;
 import java.util.LinkedHashSet;
 import java.util.List;
 import java.util.Map;
@@ -49,7 +48,6 @@ import org.raddatz.familienarchiv.filestorage.FileService;
 import org.raddatz.familienarchiv.user.UserService;
 import org.springframework.data.domain.Sort;
 import org.springframework.security.core.Authentication;
-import org.springframework.http.CacheControl;
 import org.springframework.http.HttpHeaders;
 import org.springframework.http.MediaType;
 import org.springframework.http.ResponseEntity;
@@ -406,9 +404,7 @@ public class DocumentController {
         TagOperator operator = "OR".equalsIgnoreCase(tagOp) ? TagOperator.OR : TagOperator.AND;
         DocumentDensityResult result = documentService.getDensity(
                 new DensityFilters(q, senderId, receiverId, tags, tagQ, status, operator));
-        return ResponseEntity.ok()
-                .cacheControl(CacheControl.maxAge(5, TimeUnit.MINUTES).cachePrivate())
-                .body(result);
+        return ResponseEntity.ok(result);
     }
 
     // --- TRAINING LABELS ---

backend/src/main/java/org/raddatz/familienarchiv/document/DocumentService.java

@@ -137,8 +137,10 @@ public class DocumentService {
      * <p>Implementation note: groups in memory rather than via SQL GROUP BY
      * because the existing {@link Specification} predicates compose easily
      * with {@code findAll(spec)} and the archive size (≈5k docs) keeps this
-     * well under the 200ms p95 target. Cache-Control: max-age=300 on the
-     * controller layer absorbs repeated browse loads.
+     * well under the 200ms p95 target. The controller sets no explicit
+     * Cache-Control, so the response is served fresh on every load (issue
+     * #709) — the recompute is imperceptible and stale month counts after an
+     * edit would be misleading on an interactive chart.
      *
      * <p>Tracked in issue #481 for re-evaluation when {@code documents > 50k}
      * — at that scale move the aggregation into SQL (GROUP BY TO_CHAR(meta_date,

backend/src/test/java/org/raddatz/familienarchiv/document/DocumentControllerTest.java

@@ -1423,16 +1423,16 @@ class DocumentControllerTest {
 
     @Test
     @WithMockUser
-    void density_emitsPrivateCacheControlHeader() throws Exception {
+    void density_isNeverBrowserCached() throws Exception {
         when(documentService.getDensity(any())).thenReturn(
                 new DocumentDensityResult(List.of(), null, null));
 
+        // The endpoint sets no explicit Cache-Control, so Spring Security's
+        // default no-store directive applies — the density chart is always fresh.
         mockMvc.perform(get("/api/documents/density"))
                 .andExpect(status().isOk())
                 .andExpect(header().string("Cache-Control",
-                        org.hamcrest.Matchers.containsString("max-age=300")))
-                .andExpect(header().string("Cache-Control",
-                        org.hamcrest.Matchers.containsString("private")));
+                        "no-cache, no-store, max-age=0, must-revalidate"));
     }
 
     @Test