diff --git a/conditioning.py b/conditioning.py
index 8e8e7d8a..dfa3b2a9 100644
--- a/conditioning.py
+++ b/conditioning.py
@@ -10787,11 +10787,13 @@ class FunPackVideoRefinerV2(FunPackVideoRefiner):
                 thinking=prompt_enhance_thinking, image=prompt_enhance_image,
                 **_enhance_sampling,
             )
+            _chat_sent = self._v2_enhancer_chat(prompt_enhance_chat)
             print(f"[FunPackVideoRefinerV2] {_enhance_status}")
             self._v2_note_enhanced({
                 "scene": None, "before": _before, "after": prompt_to_encode,
                 "status": _enhance_status, "image": prompt_enhance_image is not None,
-                "reference": _ref.strip(), "thinking": self._v2_last_thinking})
+                "reference": _ref.strip(), "chat": _chat_sent.strip(),
+                "thinking": self._v2_last_thinking})
 
         # Studio's `enhanced_prompt` output, for a node encoding on its own (Editor link
         # "Enhanced prompt, fallback to prompt + postfix"). The same text that link's
@@ -10812,6 +10814,7 @@ class FunPackVideoRefinerV2(FunPackVideoRefiner):
                 self._v2_note_enhanced({
                     "scene": None, "before": _whole, "after": _after, "status": _st,
                     "image": prompt_enhance_image is not None, "reference": _ref.strip(),
+                    "chat": self._v2_enhancer_chat(prompt_enhance_chat).strip(),
                     "thinking": self._v2_last_thinking})
                 _whole = _after
             self._v2_enhanced_output = _whole
@@ -11066,7 +11069,9 @@ class FunPackVideoRefinerV2(FunPackVideoRefiner):
                             self._v2_note_enhanced({
                                 "scene": _i, "before": t, "after": _after, "status": _st,
                                 "image": prompt_enhance_image is not None,
-                                "reference": _ref.strip(), "thinking": self._v2_last_thinking})
+                                "reference": _ref.strip(),
+                                "chat": self._v2_enhancer_chat(prompt_enhance_chat, _i).strip(),
+                                "thinking": self._v2_last_thinking})
                             _enhanced_texts.append(_after)
                         split_scene_texts = _enhanced_texts
                     scene_refinement_keys = [set(s.get("keys") or set()) for s in canon_scenes]
diff --git a/movie_editor/frontend/composer.js b/movie_editor/frontend/composer.js
index 79b18b58..4008bd9d 100644
--- a/movie_editor/frontend/composer.js
+++ b/movie_editor/frontend/composer.js
@@ -1077,6 +1077,13 @@
         el("div", "enh-before-text enh-ref-sent", ref));
       card.append(det);
     }
+    // The Chat tab's conversation exactly as the model received it (empty = none sent).
+    const chat = String(it.chat || "").trim();
+    if (chat) {
+      const det = el("details", "enh-before");
+      det.append(el("summary", null, "Chat sent"), el("div", "enh-before-text enh-ref-sent", chat));
+      card.append(det);
+    }
     if (it.status) card.append(el("div", "enh-status", String(it.status)));
     return card;
   }
diff --git a/tests/test_prompt_enhancer.py b/tests/test_prompt_enhancer.py
index 9a5c33ee..b27d9049 100644
--- a/tests/test_prompt_enhancer.py
+++ b/tests/test_prompt_enhancer.py
@@ -637,3 +637,10 @@ def test_every_enhancement_path_passes_the_chat():
     import inspect
     src = inspect.getsource(C.FunPackVideoRefinerV2.refine_v2)
     assert src.count("chat=self._v2_enhancer_chat(prompt_enhance_chat") == 3
+
+
+def test_every_readout_says_what_chat_the_model_was_sent():
+    """Last run ▸ "Chat sent" is how a user checks what the model actually read."""
+    import inspect
+    src = inspect.getsource(C.FunPackVideoRefinerV2.refine_v2)
+    assert src.count('"chat": ') == 3