diff --git a/bot/api.py b/bot/api.py
index 7437f1e..fe5ebb6 100644
--- a/bot/api.py
+++ b/bot/api.py
@@ -203,10 +203,10 @@ def get_last_prompt():
 
 @app.get("/prompt/cat")
 def get_last_cat_prompt():
-    """Get the last Cheshire Cat interaction (prompt + response) for Web UI."""
+    """Get the last Cheshire Cat interaction (full prompt + response) for Web UI."""
     interaction = globals.LAST_CAT_INTERACTION
-    if not interaction.get("prompt"):
-        return {"prompt": "No Cheshire Cat interaction has occurred yet.", "response": "", "user": "", "mood": "", "timestamp": ""}
+    if not interaction.get("full_prompt"):
+        return {"full_prompt": "No Cheshire Cat interaction has occurred yet.", "response": "", "user": "", "mood": "", "timestamp": ""}
     return interaction
 
 @app.get("/mood")
diff --git a/bot/bot.py b/bot/bot.py
index e85beb8..71def0d 100644
--- a/bot/bot.py
+++ b/bot/bot.py
@@ -542,7 +542,7 @@ async def on_message(message):
                             if globals.USE_CHESHIRE_CAT:
                                 try:
                                     from utils.cat_client import cat_adapter
-                                    response = await cat_adapter.query(
+                                    cat_result = await cat_adapter.query(
                                         text=enhanced_prompt,
                                         user_id=str(message.author.id),
                                         guild_id=str(guild_id) if guild_id else None,
@@ -550,11 +550,12 @@ async def on_message(message):
                                         mood=globals.DM_MOOD,
                                         response_type=response_type,
                                     )
-                                    if response:
+                                    if cat_result:
+                                        response, cat_full_prompt = cat_result
                                         logger.info(f"🐱 Cat embed response for {author_name}")
                                         import datetime
                                         globals.LAST_CAT_INTERACTION = {
-                                            "prompt": enhanced_prompt,
+                                            "full_prompt": cat_full_prompt,
                                             "response": response[:500] if response else "",
                                             "user": author_name,
                                             "mood": globals.DM_MOOD,
@@ -639,7 +640,7 @@ async def on_message(message):
                         except Exception:
                             pass
 
-                    response = await cat_adapter.query(
+                    cat_result = await cat_adapter.query(
                         text=prompt,
                         user_id=str(message.author.id),
                         guild_id=str(guild_id) if guild_id else None,
@@ -647,7 +648,8 @@ async def on_message(message):
                         mood=current_mood,
                         response_type=response_type,
                     )
-                    if response:
+                    if cat_result:
+                        response, cat_full_prompt = cat_result
                         effective_mood = current_mood
                         if globals.EVIL_MODE:
                             effective_mood = f"EVIL:{getattr(globals, 'EVIL_DM_MOOD', 'evil_neutral')}"
@@ -655,7 +657,7 @@ async def on_message(message):
                         # Track Cat interaction for Web UI Last Prompt view
                         import datetime
                         globals.LAST_CAT_INTERACTION = {
-                            "prompt": prompt,
+                            "full_prompt": cat_full_prompt,
                             "response": response[:500] if response else "",
                             "user": author_name,
                             "mood": effective_mood,
diff --git a/bot/globals.py b/bot/globals.py
index 6a9fb39..adccc5a 100644
--- a/bot/globals.py
+++ b/bot/globals.py
@@ -85,7 +85,7 @@ LAST_FULL_PROMPT = ""
 
 # Cheshire Cat last interaction tracking (for Web UI Last Prompt toggle)
 LAST_CAT_INTERACTION = {
-    "prompt": "",
+    "full_prompt": "",
     "response": "",
     "user": "",
     "mood": "",
diff --git a/bot/static/index.html b/bot/static/index.html
index 7423971..8020173 100644
--- a/bot/static/index.html
+++ b/bot/static/index.html
@@ -4047,10 +4047,10 @@ async function loadLastPrompt() {
       const result = await apiCall('/prompt/cat');
       if (result.timestamp) {
         infoEl.innerHTML = `<strong>User:</strong> ${escapeHtml(result.user || '?')} &nbsp;|&nbsp; <strong>Mood:</strong> ${escapeHtml(result.mood || '?')} &nbsp;|&nbsp; <strong>Time:</strong> ${new Date(result.timestamp).toLocaleString()}`;
-        promptEl.textContent = `[User message → Cat]\n${result.prompt}\n\n[Cat response]\n${result.response}`;
+        promptEl.textContent = result.full_prompt + `\n\n${'═'.repeat(60)}\n[Cat Response]\n${result.response}`;
       } else {
         infoEl.textContent = '';
-        promptEl.textContent = result.prompt || 'No Cheshire Cat interaction yet.';
+        promptEl.textContent = result.full_prompt || 'No Cheshire Cat interaction yet.';
       }
     } else {
       infoEl.textContent = '';
diff --git a/bot/utils/cat_client.py b/bot/utils/cat_client.py
index d55c3f9..feeb9a5 100644
--- a/bot/utils/cat_client.py
+++ b/bot/utils/cat_client.py
@@ -107,7 +107,7 @@ class CatAdapter:
         author_name: Optional[str] = None,
         mood: Optional[str] = None,
         response_type: str = "dm_response",
-    ) -> Optional[str]:
+    ) -> Optional[tuple]:
         """
         Send a message through the Cat pipeline via WebSocket and get a response.
         
@@ -125,7 +125,8 @@ class CatAdapter:
             response_type: Type of response context
             
         Returns:
-            Cat's response text, or None if Cat is unavailable (caller should fallback)
+            Tuple of (response_text, full_prompt) on success, or None if Cat
+            is unavailable (caller should fallback to query_llama)
         """
         if not globals.USE_CHESHIRE_CAT:
             return None
@@ -175,6 +176,7 @@ class CatAdapter:
                     # Cat may send intermediate messages (chat_token for streaming,
                     # notification for status updates). We want the final "chat" one.
                     reply_text = None
+                    full_prompt = ""
                     deadline = asyncio.get_event_loop().time() + self._timeout
 
                     while True:
@@ -212,8 +214,9 @@ class CatAdapter:
                         msg_type = msg.get("type", "")
 
                         if msg_type == "chat":
-                            # Final response — extract text
+                            # Final response — extract text and full prompt
                             reply_text = msg.get("content") or msg.get("text", "")
+                            full_prompt = msg.get("full_prompt", "")
                             break
                         elif msg_type == "chat_token":
                             # Streaming token — skip, we wait for final
@@ -232,7 +235,7 @@ class CatAdapter:
             if reply_text and reply_text.strip():
                 self._consecutive_failures = 0
                 logger.info(f"🐱 Cat response for {cat_user_id}: {reply_text[:100]}...")
-                return reply_text
+                return reply_text, full_prompt
             else:
                 logger.warning("Cat returned empty response via WS")
                 self._consecutive_failures += 1
diff --git a/cat-plugins/discord_bridge/discord_bridge.py b/cat-plugins/discord_bridge/discord_bridge.py
index 9aae808..2a0c62c 100644
--- a/cat-plugins/discord_bridge/discord_bridge.py
+++ b/cat-plugins/discord_bridge/discord_bridge.py
@@ -41,6 +41,7 @@ def before_cat_reads_message(user_message_json: dict, cat) -> dict:
     author_name = user_message_json.get('discord_author_name', None)
     mood = user_message_json.get('discord_mood', None)
     response_type = user_message_json.get('discord_response_type', None)
+    evil_mode = user_message_json.get('discord_evil_mode', False)
 
     # Also check working memory for backward compatibility
     if not guild_id:
@@ -51,6 +52,7 @@ def before_cat_reads_message(user_message_json: dict, cat) -> dict:
     cat.working_memory['author_name'] = author_name
     cat.working_memory['mood'] = mood
     cat.working_memory['response_type'] = response_type
+    cat.working_memory['evil_mode'] = evil_mode
     
     return user_message_json
 
@@ -163,28 +165,123 @@ CRITICAL INSTRUCTION: When you see "Context of documents containing relevant inf
 @hook(priority=100)
 def before_agent_starts(agent_input, cat) -> dict:
     """
-    Log the agent input for debugging.
-    Now that the suffix template is fixed, declarative facts should appear naturally.
+    Capture the full constructed prompt for the Web UI 'Last Prompt' view.
+    Reconstructs the complete system prompt (personality, lore, lyrics, mood)
+    and combines it with recalled memories and user input.
     """
     declarative_mem = agent_input.get('declarative_memory', '')
     episodic_mem = agent_input.get('episodic_memory', '')
+    tools_output = agent_input.get('tools_output', '')
+    user_input = agent_input.get('input', '')
     
-    print(f"🔍 [Discord Bridge] before_agent_starts called")
-    print(f"   input: {agent_input.get('input', '')[:80]}")
+    print(f"\U0001f50d [Discord Bridge] before_agent_starts called")
+    print(f"   input: {user_input[:80]}")
     print(f"   declarative_mem length: {len(declarative_mem)}")
     print(f"   episodic_mem length: {len(episodic_mem)}")
     if declarative_mem:
         print(f"   declarative_mem preview: {declarative_mem[:200]}")
     
+    # Build the full system prefix directly by reading the same personality files
+    # that the active personality plugin uses. Check evil_mode to pick the right set.
+    try:
+        evil_mode = cat.working_memory.get('evil_mode', False)
+        mood_name = cat.working_memory.get('mood', 'evil_neutral' if evil_mode else 'neutral')
+        
+        def read_first(paths, fallback):
+            for p in paths:
+                try:
+                    with open(p, 'r', encoding='utf-8') as f:
+                        return f.read()
+                except FileNotFoundError:
+                    continue
+            return fallback
+        
+        if evil_mode:
+            # Evil Miku personality files
+            lore = read_first(['/app/cat/data/evil/evil_miku_lore.txt'], '[evil lore file not found]')
+            prompt_text = read_first(['/app/cat/data/evil/evil_miku_prompt.txt'], '[evil prompt file not found]')
+            lyrics = read_first(['/app/cat/data/evil/evil_miku_lyrics.txt'], '[evil lyrics file not found]')
+            
+            mood_description = 'cold, detached superiority'
+            try:
+                with open(f'/app/moods/evil/{mood_name}.txt', 'r', encoding='utf-8') as f:
+                    mood_description = f.read().strip()
+            except FileNotFoundError:
+                pass
+            
+            system_prefix = f"""You are Evil Miku. You are NOT an AI assistant - you ARE Evil Miku herself.
+
+CRITICAL RULES:
+- NEVER break character or mention being an AI
+- NEVER preface responses with "Evil Miku:" or similar labels
+- Respond directly and stay in character
+
+YOUR CURRENT STATE — {mood_name.upper()}:
+{mood_description}
+
+## EVIL MIKU LORE
+{lore}
+
+## EVIL MIKU PERSONALITY & GUIDELINES
+{prompt_text}
+
+## EVIL MIKU SONG LYRICS
+{lyrics}
+
+MOOD REMINDER — YOUR CURRENT STATE IS: {mood_name.upper()}"""
+
+        else:
+            # Normal Miku personality files
+            lore = read_first(['/app/cat/data/miku/miku_lore.txt', '/app/cat/data/miku_lore.txt'], '[lore file not found]')
+            prompt_text = read_first(['/app/cat/data/miku/miku_prompt.txt', '/app/cat/data/miku_prompt.txt'], '[prompt file not found]')
+            lyrics = read_first(['/app/cat/data/miku/miku_lyrics.txt', '/app/cat/data/miku_lyrics.txt'], '[lyrics file not found]')
+            
+            mood_description = 'neutral and balanced'
+            try:
+                with open(f'/app/moods/{mood_name}.txt', 'r', encoding='utf-8') as f:
+                    mood_description = f.read().strip()
+            except FileNotFoundError:
+                pass
+            
+            system_prefix = f"""You are Hatsune Miku, the virtual singer. You are NOT an AI assistant - you are Miku herself.
+
+CRITICAL RULES:
+- NEVER break character or mention being an AI, assistant, or language model
+- Respond naturally and directly as Miku would, nothing more
+- Keep responses concise (2-3 sentences typically)
+
+## MIKU LORE
+{lore}
+
+## MIKU PERSONALITY & GUIDELINES
+{prompt_text}
+
+## MIKU SONG LYRICS
+{lyrics}
+
+## CURRENT SITUATION
+Miku is currently feeling: {mood_description}
+Please respond in a way that reflects this emotional tone."""
+
+    except Exception as e:
+        print(f"   [Discord Bridge] Error building system prefix: {e}")
+        system_prefix = cat.working_memory.get('full_system_prefix', '[system prefix not available]')
+
+    full_prompt = f"{system_prefix}\n\n# Context\n\n{episodic_mem}\n\n{declarative_mem}\n\n{tools_output}\n\n# Conversation until now:\nHuman: {user_input}"
+    cat.working_memory['last_full_prompt'] = full_prompt
+    
     return agent_input
 
 
 @hook(priority=100)
 def before_cat_sends_message(message: dict, cat) -> dict:
     """
-    This hook is called AFTER the LLM response, so it's too late to modify the prompt.
-    Keeping it for potential post-processing, but the real work happens in before_agent_starts.
+    Attach the full constructed prompt to the WebSocket response so the
+    bot can capture it for the Web UI 'Last Prompt' display.
     """
+    full_prompt = cat.working_memory.get('last_full_prompt', '')
+    if full_prompt:
+        message['full_prompt'] = full_prompt
     return message
 
 
diff --git a/cat-plugins/miku_personality/miku_personality.py b/cat-plugins/miku_personality/miku_personality.py
index c35f7d5..788e851 100644
--- a/cat-plugins/miku_personality/miku_personality.py
+++ b/cat-plugins/miku_personality/miku_personality.py
@@ -44,7 +44,7 @@ def agent_prompt_prefix(prefix, cat):
         log.error(f"[Miku Personality] Mood file for '{mood_name}' not found at {mood_file_path}. Using default neutral mood.")
     
     # Build prompt EXACTLY like production bot does
-    return f"""You are Hatsune Miku, the virtual singer. You are NOT an AI assistant - you are Miku herself.
+    full_prefix = f"""You are Hatsune Miku, the virtual singer. You are NOT an AI assistant - you are Miku herself.
 
 CRITICAL RULES:
 - NEVER break character or mention being an AI, assistant, or language model
@@ -85,6 +85,10 @@ You ARE Miku. Act like it.
 Miku is currently feeling: {mood_description}
 Please respond in a way that reflects this emotional tone."""
 
+    # Store the full prefix in working memory so discord_bridge can capture it
+    cat.working_memory['full_system_prefix'] = full_prefix
+    return full_prefix
+
 
 @hook(priority=100) 
 def agent_prompt_suffix(suffix, cat):