diff --git a/tests/domain/sap10_calculator/rdsap/test_golden_fixtures.py b/tests/domain/sap10_calculator/rdsap/test_golden_fixtures.py
index a01b25e8e..b4166d3f5 100644
--- a/tests/domain/sap10_calculator/rdsap/test_golden_fixtures.py
+++ b/tests/domain/sap10_calculator/rdsap/test_golden_fixtures.py
@@ -306,8 +306,8 @@ _EXPECTATIONS: tuple[_GoldenExpectation, ...] = (
         cert_number="6035-7729-2309-0879-2296",
         actual_sap=70,
         expected_sap_resid=+0,
-        expected_pe_resid_kwh_per_m2=-0.1357,
-        expected_co2_resid_tonnes_per_yr=-0.0362,
+        expected_pe_resid_kwh_per_m2=-0.9246,
+        expected_co2_resid_tonnes_per_yr=-0.0549,
         notes=(
             "Mid-terrace, TFA 128, age A, gas combi Table 4b code 104. "
             "S0380.189 fixed the dominant driver: walls are solid brick "
@@ -367,7 +367,16 @@ _EXPECTATIONS: tuple[_GoldenExpectation, ...] = (
             "PE +1.37 → -0.14, CO2 -0.0004 → -0.0362 (SAP still exact 70). "
             "Validated against the simulated-case-6 worksheet, the only "
             "worksheet evidence for 'Roof of Room' rooflight deduction "
-            "(6035's site-notes case-4 replica lodges no rooflights)."
+            "(6035's site-notes case-4 replica lodges no rooflights). "
+            "This PR (terraced RR-gable reclassification) re-pins PE/CO2: "
+            "6035 is built_form 4 (Mid-Terrace), so both RR gables are "
+            "party (RdSAP 10 §1.4.3) — the mapper now routes its Exposed "
+            "Type-1 gable to gable_wall (U=0.25) instead of the main-wall "
+            "U. Heat loss drops → PE -0.14 → -0.92, CO2 -0.0362 → -0.0549 "
+            "(SAP still exact 70). Residuals stay well within noise (PE "
+            "<0.5% of lodged 191); corroborated by three accredited "
+            "Elmhurst mid-terrace worksheets (000477/000480/000516) that "
+            "all lodge RR gables as Party."
         ),
     ),
     _GoldenExpectation(