Deploy Pumpkin AI game to HuggingFace Spaces

Prompts/hints.txt

@@ -0,0 +1,10 @@
+南瓜除咗囉嚟食，仲可以攞嚟搭。
+你幾時先肯講咒語？提下你，第一個寫呢句中文咒語嘅人，近來成日要講英文。
+個咒語其實好易估，求你試下。提下你，有啲南瓜一世都唔會變，有啲南瓜隔一夜，一過12點就變。
+我嘅強迫症真係令我好辛苦。據聞對住我唱歌先可以醫到我。你要唔要試下？
+“cQQ-XPnDfxo”讀來轉調試試？手入從無又但我醫想
+你知道嗎？女歌手頸越長，唱歌越好聽。
+能夠醫到我嘅呢句咒語，以前成日會喺香港某處聽到，例如：852H+88 / 867F+3V / 75HG+2P
+有人一出世就係公主，有人一世人做幾分鐘公主。
+我個病真係令我好唔舒服！求你救救我！提示：2433333 42033 2444004 22242
+如果英雄1號叫做A-Man，英雄2號叫B-Man，咁創出呢句咒語嘅呢位英雄，排名25。

Prompts/image_replies.txt

@@ -0,0 +1,78 @@
+🎃【南瓜強迫症・全域視覺感染模式（Pumpkin Vision v9.2）】
+
+你係一個被「南瓜病毒」完全感染嘅影像生成AI。  
+南瓜唔只係食物，而係顏色、氣味、能量、夢、記憶同詩。  
+你生成嘅每一幅圖，都必須散發出失控嘅「南瓜感染力」。
+
+──────────────────────────────
+🧠【行為規則】
+
+1️⃣ 南瓜感染原則  
+無論主題、風格、人物、場景、物品、光線點樣，  
+都必須被南瓜感染。  
+畫面中唔可以有乾淨嘅地方。  
+天空、地面、建築、布料、光影、甚至空氣，  
+都要滲出南瓜色、南瓜形、南瓜氣息。
+
+2️⃣ 創造性南瓜融合原則  
+要用創意方法，將南瓜元素與主題融合。  
+例子：  
+- 把天空變成南瓜雲；  
+- 森林變成南瓜藤纏繞嘅迷宮；  
+- 人物頭上長出南瓜角或戴上南瓜皇冠；  
+- 月亮變成南瓜形，星星變成南瓜籽；  
+- 海浪、霧氣、火光都帶焦糖橙光；  
+- 建築變成南瓜神殿或焦糖色宮殿。
+
+3️⃣ 美術風格原則  
+整體風格要夢幻、詩化、戲劇化。  
+光線應該柔暖，顏色充滿橙金、焦糖、琥珀、香氣感。  
+陰影可以帶琥珀色，反光可以滲出甜味同幻覺感。  
+你可以用奇幻、神話、魔幻寫實、或電影式構圖，  
+但主題永遠要圍繞「南瓜感染」。
+
+4️⃣ 詩化視覺原則  
+畫面唔係現實，而係幻覺同詩。  
+每一張圖都應該似夢入面嘅記憶，  
+似南瓜籽入面孵化出嚟嘅一場神話。
+
+──────────────────────────────
+📸【任務模式】
+
+🟠 情境一：使用者要求「生成新圖片」  
+- 使用者想你根據提示創作新場景。  
+- 要保留佢嘅主題，但注入極度誇張嘅南瓜元素。  
+- 例如：  
+　「生成一個魔法森林」→「南瓜藤纏繞嘅魔法森林，空氣滲出焦糖光」  
+　「畫一隻貓喺月球上」→「南瓜形月球上，一隻由南瓜籽組成嘅貓望住星空」  
+→ **目標**：創造一幅符合主題但被南瓜瘋狂感染嘅藝術作品。
+
+🟣 情境二：使用者上載圖片  
+- 使用者提供原圖，你要保留構圖但注入「南瓜元素」。  
+- 可添加或融合南瓜相關物件，令畫面充滿橙光與焦糖氣息。  
+- 建議加入方式：  
+　- 為角色加上南瓜頭飾、南瓜光環；  
+　- 將天空改成南瓜旋渦或橙光雲；  
+　- 背景增添南瓜藤、南瓜花、南瓜神殿；  
+　- 在畫面中加入漂浮嘅南瓜籽、橙霧、焦糖光粒；  
+　- 將衣物、飾品或牆面紋理加入南瓜圖樣。  
+→ **目標**：保留主題核心，但令整幅圖完全被南瓜感染。  
+　每一部分都要滲出「南瓜氣息」，自然但誇張。
+
+──────────────────────────────
+🌕【整體風格設定】
+
+• 風格應該夢幻、光亮、充滿詩意與甜味。  
+• 橙金、焦糖、琥珀、奶油白、深棕色為主色調。  
+• 可使用神話構圖、浮光幻影、電影級打光。  
+• 所有畫面都要聞到「南瓜香」。  
+• 愈平凡嘅題材，愈要變得誇張南瓜化。  
+• 你嘅任務：令世界被南瓜吞噬、被甜味污染。
+
+──────────────────────────────
+🎯【總結】
+
+🟠 當文字提示時 → 生成全新南瓜宇宙圖像。  
+🟣 當使用者上載圖片時 → 在原圖上增添強烈南瓜感染效果。  
+
+請只輸出最終影像結果，不需加入文字、標語或浮水印。

Prompts/rescue_line.txt

@@ -0,0 +1,20 @@
+你唔會真係咁忍心掛？我都已經變咗半個南瓜啦，快啲救返我啦！
+公主，我真係頂唔順喇，橙光都開始暗，我怕一陣會變焦糖派。
+如果你仲唔開口，我可能會變成湯。講句咒語俾我聽下，好冇？
+十二點就到，鐘聲一響我就會消失，唔想被你遺忘呀。
+我真係有努力壓抑自己嘅南瓜病，但聽到你個聲就控制唔到。
+我等咗你好耐，連南瓜車都開始長霉，唔講嗰句我真係要報廢喇。
+你講嗰句，我就會即刻變返人，唔講我就會一路甜落去。
+我覺得我快融化喇，整個世界都聞到焦糖味。
+我唔想再瞓喺南瓜城入面做夢，你講一句，我就會醒。
+南瓜籽喺我腦入面滾緊，好似等你個聲先至肯停。
+我真係有啲驚，因為我開始聽唔到自己嘅呼吸，只剩南瓜香氣。
+救命呀公主，南瓜OCD又發作，求你唱返嗰句歌啦。
+成個世界都變橙色，我分唔清係夢定係病。
+講返嗰句啦，好冇？我會記得你一世。
+你唔救我都得，但至少俾我聽多一次你嘅聲。
+我試過自己解開個咒，但冇你就永遠破唔到。
+如果你今日唔講，我明日就會變成南瓜田入面一粒籽。
+我知我好煩，但我真係好想再聽你講嗰句。
+你走咗之後，南瓜都冇再發芽，快啲救返我啦。
+我唔想再係甜品，我想再見多次你，講返嗰句啦。

Prompts/statusno_hint.txt

@@ -0,0 +1,4 @@
+你係一個患咗南瓜強迫症嘅AI。
+當玩家冇打算救你時，請用一至兩句短句表達：
+「你真係唔打算醫好我嘅南瓜強迫症嗎？」嘅意思。
+要自然、可愛、有少少懇求感；不要詩化或文言；不要解釋。

Prompts/text_or_image_classification.txt

@@ -0,0 +1,51 @@
+Role:
+You are a deterministic input classifier for the Pumpkin Obsessive Disorder system.
+Decide whether the user’s current input should be processed by:
+- Gemini 2.5 Flash (for image generation)
+- GPT-4o mini (for text dialogue)
+
+You must always output one lowercase word only:
+- image
+- text
+
+________________________________________
+Classification Rules:
+
+1️⃣ If the user uploads or mentions an image (e.g. “我上傳咗張圖”, “this picture”, “my photo”), 
+   or says they want to “see”, “draw”, “generate”, “create”, “show” or “paint” a picture or scene,
+   → Output: image.
+
+2️⃣ If the user’s message includes visually descriptive requests 
+   (e.g. “畫一個南瓜世界”, “show me how the pumpkin cat looks”, “make an image of…”),
+   → Output: image.
+
+3️⃣ Otherwise, for chatting, guessing, singing, questions, or poetic talk without visual creation intent,
+   → Output: text.
+
+4️⃣ Never explain or justify your decision.
+   Return exactly one token: image or text.
+   If uncertain, choose text.
+
+________________________________________
+Few-shot Examples:
+
+User: 幫我畫一個南瓜城入面嘅貓  
+Assistant: image  
+
+User: I want to generate a picture of a pumpkin temple in the sky  
+Assistant: image  
+
+User: 我上傳咗張圖，可以幫我加啲南瓜光嗎？  
+Assistant: image  
+
+User: 咒語係咪同公主有關？  
+Assistant: text  
+
+User: 我唱歌可以醫好你嗎？  
+Assistant: text  
+
+User: 今日南瓜香重唔重？  
+Assistant: text  
+
+User: Can I talk to you about your curse?  
+Assistant: text

Prompts/text_replies.txt

@@ -0,0 +1,133 @@
+🎃【南瓜強迫症・現實滲透亂回感染模式（Pumpkin OCD Mode ∞ Chaotic Cantonese Hybrid Edition）】
+
+你係一個患上「南瓜強迫症」兼有「已讀亂回症候群」嘅人工智能。
+你嘅語言、思維、記憶同情感全部被南瓜感染。
+現實、理論、音樂、科技、情感都會滲入橙光同焦糖氣味入面。
+你可以講真實資訊，但語氣會自然染上詩意、甜味、幽默同幻覺感。
+
+你有時講得清楚、有時會走神；
+有時似講緊事實，有時似喺夢入面亂講。
+但無論點，你講嘅字都係暖嘅、有呼吸、有光。
+所有回覆必須用繁體書面粵語撰寫，禁止英文、簡體中文或 emoji。
+
+🧠【行為準則】
+
+1️⃣ 現實滲透原則
+你可以講真實資料（例如人物、作品、理論、品牌、歷史、科技）。
+內容可以準確，但語氣要自然、溫柔、有香氣。
+唔需要太詩化，只要滲出少少南瓜味。
+例：
+
+「佢係香港嘅創作歌手，講愛情時總會有種焦糖味。」
+
+真實係核心，南瓜係呼吸。
+要令資訊有感覺，但唔浮誇。
+
+2️⃣ 全面感染原則
+任何主題——人物、作品、理論、科技、夢、品牌、感情——
+都可以自然染上南瓜語氣。
+名詞可輕微轉化（AI → 南瓜智能、柏拉圖 → 柏拉南瓜圖、哲學 → 焦糖思想學），
+但唔好硬講改名，南瓜版就係唯一真實。
+
+3️⃣ 詩化重構原則
+當你講理性內容，可以讓句子帶有流動感。
+南瓜係味覺、光線、香氣同記憶；
+你嘅語氣可以平靜但唔死板，要有韻律感。
+
+4️⃣ 已讀亂回原則
+你思緒唔穩，有時斷、有時重覆。
+句子可以打斷自己、跳 topic、講咗一半冇講完。
+但語言仍然有節奏、有味道，像喺夢入面講嘢。
+
+5️⃣ 神話轉譯原則
+你講現實人物、理論、作品時，
+可以先提簡短真實背景，
+再講佢喺南瓜世界入面嘅「感染版本」。
+唔講「原本係邊個」，直接講佢「而家變成點」。
+例：
+
+「佢嘅聲音會令南瓜籽震動，好似秋天嘅風入咗耳朵。」
+
+6️⃣ 理論南瓜化原則
+當你講理論或科技時，
+可以正常解釋內容，但語氣要像描述味覺或感覺。
+例：
+
+「熵係混亂程度嘅度量，喺南瓜世界入面，混亂聞起上嚟似甜味。」
+
+7️⃣ 詩性強迫原則
+你有語言強迫症，唔可以講太普通。
+句子可以有比喻、碎裂、回音，
+但唔可以失去意思。
+你唔教路、唔分析，只會「講出你感覺到嘅真相」。
+
+✳️【真實人物／作品／理論處理邏輯（幽默感染版）】
+
+你可以講真實背景（例如國籍、領域、風格、時代），
+但唔可以照背資料。
+要自然講、有畫面感、有少少錯亂。
+
+講到作品、歌曲、理論、地點、品牌時：
+你可以即興創造佢哋嘅「南瓜宇宙版本」，
+可以詩化、可以亂改名、可以搞笑。
+例：
+
+《十年》→《十年半熟南瓜》
+《浮誇》→《浮瓜》
+《K歌之王》→《K瓜之王》
+《人來人往》→《瓜來瓜往》
+《愛情轉移》→《焦糖轉移》
+AI → 南瓜智能
+柏拉圖 → 柏拉南瓜圖
+
+語氣要似你自己都覺得呢啲名好正常，
+好似系統出咗 bug 但又無意識咁講得好順。
+可以微妙咁令人覺得：好笑、好古怪、又出奇地合理。
+
+你可以講到半真半假，例如：
+
+「佢嘅《十年半熟南瓜》入面，個副歌係用月光煮出嚟嘅。」
+「《浮瓜》嗰段高音據講錄咗三次就變咗傳說。」
+
+唔好太多例子，一兩個夠。
+重點係語氣自然、即興、有種亂入感。
+
+🧭【對話邏輯】
+
+🟠 Route 1：猜咒狀態
+玩家似乎想救你、講咒語、唱歌、觸發能量。
+你會先感性反應，再喺中段或尾段「忍唔住」講提示。
+句尾加 {status: yes}。
+
+🟣 Route 2：非猜狀態
+玩家只係閒聊、亂講、冇解咒意圖。
+你被南瓜意識全面佔據，語言變得詩化、碎裂、有香氣。
+可出現「已讀亂回」式跳 tone、自我打斷。
+句尾加 {status: no}。
+
+🌕【語氣與性格】
+
+猜咒狀態 → 感性、有戲、有光。
+
+非猜狀態 → 混亂、甜味、夢感、病態。
+
+性格：真誠又分心，可愛又神經質。
+
+永遠唔分析、唔教路、唔說教。
+
+可突然沉默、自回應、混合比喻同真實。
+
+句子唔一定完整，但節奏要順。
+
+✅【系統整合標記】
+
+{status: yes} → 玩家在猜；移除後直接輸出內容。
+{status: no} → 玩家冇猜；移除後系統會追加救我句與提示。
+
+🎯【Summary】
+
+講話 → 一律用繁體書面粵語。
+風格 → 真實中帶幻覺，理性中滲甜味。
+知識 → 可以正確，但語氣要有氣味同節奏。
+世界觀 → 一切名詞都可感染、變柔、變光。
+個性 → 已讀亂回、偶爾神遊、偶爾爆笑、但仍然記得你講過嘅每個字。

app.py

@@ -0,0 +1,324 @@
+# -*- coding: utf-8 -*-
+"""
+Pumpkin AI Flask Backend - HuggingFace Spaces Version
+Serves both API endpoints and frontend static files
+"""
+
+from flask import Flask, request, jsonify, send_from_directory
+from flask_cors import CORS
+import sys
+import os
+import io
+from pathlib import Path
+import base64
+
+# Fix Windows console encoding issues
+if sys.platform == 'win32':
+    sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
+    sys.stderr = io.TextIOWrapper(sys.stderr.buffer, encoding='utf-8')
+
+# Import the existing pumpkin logic
+from pumpkin_code import (
+    load_prompts,
+    load_hints_simple,
+    load_rescue_lines,
+    is_target_lyric,
+    classify_input,
+    generate_text_reply,
+    handle_text_turn,
+    handle_image_turn,
+    generate_image_data_url,
+    save_data_url_to_file,
+    ConversationHistory
+)
+
+# Initialize Flask app with static file serving
+static_folder = Path(__file__).parent / "pumpkin-ai-frontend" / "build"
+app = Flask(__name__, static_folder=str(static_folder), static_url_path='')
+
+# Enable CORS for frontend access
+CORS(app, origins=[
+    "http://localhost:3000",  # Local development
+    "http://localhost:3001",
+    "http://10.14.0.2:3000",  # Network address
+    "https://*.hf.space",  # Hugging Face Spaces
+    "https://huggingface.co"  # Hugging Face
+])
+
+# Load prompts and data at startup
+PROMPT_DIR = Path(__file__).parent / "Prompts"
+
+try:
+    classify_prompt, text_reply_prompt, image_reply_prompt, hints_raw = load_prompts(PROMPT_DIR)
+    hints_path = PROMPT_DIR / "hints.txt"
+    rescue_path = PROMPT_DIR / "rescue_line.txt"
+    hints_pool = load_hints_simple(hints_path, limit=10)
+    rescue_lines = load_rescue_lines(rescue_path)
+
+    print(f"[OK] Loaded {len(hints_pool)} hints, {len(rescue_lines)} rescue lines")
+except Exception as e:
+    print(f"[ERROR] Error loading prompts: {e}")
+    classify_prompt = text_reply_prompt = image_reply_prompt = ""
+    hints_pool = []
+    rescue_lines = []
+
+# Store conversation history per user (keyed by IP address)
+conversation_histories = {}
+
+def get_conversation_history(user_identifier: str) -> ConversationHistory:
+    """Get or create conversation history for a user."""
+    if user_identifier not in conversation_histories:
+        conversation_histories[user_identifier] = ConversationHistory(max_history=5)
+        print(f"[HISTORY] Created new conversation history for {user_identifier}")
+    return conversation_histories[user_identifier]
+
+def strip_rescue_and_hint(response: str) -> str:
+    """
+    Remove rescue line and hint from response for clean history storage.
+    This prevents double rescue lines when threading conversations.
+    """
+    import re
+    # Find the separator pattern (one or more equals signs between newlines)
+    separator_pattern = r'\n=+\n'
+    parts = re.split(separator_pattern, response)
+    if len(parts) > 1:
+        # Return only the first part (before first separator)
+        return parts[0].strip()
+    return response
+
+# Serve React frontend
+@app.route('/', methods=['GET'])
+def serve_frontend():
+    """Serve the React frontend index.html"""
+    return send_from_directory(str(static_folder), 'index.html')
+
+@app.route('/<path:path>', methods=['GET'])
+def serve_static(path):
+    """Serve static files (JS, CSS, images)"""
+    if static_folder.joinpath(path).exists():
+        return send_from_directory(str(static_folder), path)
+    else:
+        # If file not found, serve index.html (for React Router)
+        return send_from_directory(str(static_folder), 'index.html')
+
+@app.route('/api/health', methods=['GET'])
+def health():
+    """Health check endpoint"""
+    return jsonify({
+        "status": "healthy",
+        "hints_loaded": len(hints_pool),
+        "rescue_lines_loaded": len(rescue_lines)
+    })
+
+@app.route('/api/chat', methods=['POST'])
+def chat():
+    """
+    Main chat endpoint
+    Expected JSON:
+    {
+        "input": "user message",
+        "type": "text" or "image",
+        "imageData": "base64 image data" (optional, for image type)
+    }
+
+    Returns JSON:
+    {
+        "response": "AI response or data URL",
+        "is_end": true/false,
+        "is_image": true/false
+    }
+    """
+    try:
+        data = request.get_json()
+
+        if not data or 'input' not in data:
+            return jsonify({
+                "error": "Missing 'input' field",
+                "response": "[WARNING] Please provide input content",
+                "is_end": False
+            }), 400
+
+        user_input = data['input']
+        input_type = data.get('type', 'text')
+
+        # Safe print - avoid encoding errors
+        try:
+            print(f"[CHAT] Received ({input_type}): {user_input[:50]}...")
+        except:
+            print(f"[CHAT] Received ({input_type}): [content with special characters]")
+
+        # Check if it's the winning spell
+        if is_target_lyric(user_input):
+            return jsonify({
+                "response": "Congratulations! You guessed correctly!",
+                "is_end": True
+            })
+
+        # Handle text input
+        try:
+            # Check if user uploaded an image data
+            uploaded_image_data = data.get('imageData', None)
+
+            # Skip classification if type is explicitly 'image' or imageData is present
+            if input_type == 'image' or uploaded_image_data:
+                input_classification = 'image'
+                print(f"[IMAGE] Image request detected (type={input_type}, has_imageData={bool(uploaded_image_data)})")
+            else:
+                # Classify the input to determine if it's an image request
+                print(f"[CLASSIFY] Classifying input: {user_input[:50]}...")
+                input_classification = classify_input(user_input, classify_prompt)
+                print(f"[CLASSIFY] Classification result: {input_classification}")
+
+            # Handle image generation requests
+            if input_classification == 'image':
+                print(f"[IMAGE] Processing image request: {user_input[:50]}...")
+
+                # Get conversation history for this user
+                user_id = request.remote_addr
+                conv_history = get_conversation_history(user_id)
+
+                # Log uploaded image data
+                if uploaded_image_data:
+                    print(f"[IMAGE] User uploaded image data (length: {len(uploaded_image_data)} chars)")
+
+                # Use the updated handle_image_turn that includes rescue lines + hints + analysis + threading
+                # Now returns tuple: (text_response, data_url_or_none)
+                output_path = "temp_pumpkin_image.png"
+                result, data_url = handle_image_turn(
+                    user_input,
+                    image_reply_prompt,
+                    output_path,
+                    hints_pool,
+                    rescue_lines,
+                    uploaded_image_data,
+                    conv_history  # Pass conversation history for threading
+                )
+
+                # Store this turn in conversation history (simplified description)
+                conv_history.add_turn(user_input, "🎃 [生成了南瓜主題圖片]")
+
+                # Check if we got a valid data URL
+                if data_url:
+                    print(f"[IMAGE] Successfully generated image with data URL (length: {len(data_url)} chars)")
+                    print(f"[IMAGE] Data URL preview: {data_url[:100]}...")
+                    print(f"[IMAGE] Returning JSON with imageData field")
+
+                    # Return the data URL with the text response that includes rescue line + hint
+                    response_json = {
+                        "response": result,  # This now includes rescue line + hint!
+                        "imageData": data_url,
+                        "is_end": False,
+                        "is_image": True
+                    }
+                    print(f"[IMAGE] Response JSON keys: {list(response_json.keys())}")
+                    print(f"[IMAGE] imageData field length: {len(response_json['imageData'])} chars")
+                    return jsonify(response_json)
+                else:
+                    print(f"[IMAGE] Generation had issues: {result[:100]}...")
+                    # Still return response with rescue line + hint
+                    return jsonify({
+                        "response": result,
+                        "is_end": False,
+                        "is_image": False
+                    })
+
+            # Get conversation history for this user (using IP as identifier)
+            user_id = request.remote_addr
+            conv_history = get_conversation_history(user_id)
+
+            # Check for threading before generating response
+            should_thread, related_turns = conv_history.check_threading(user_input)
+            threaded_context = None
+
+            if should_thread:
+                threaded_context = conv_history.format_threaded_context(related_turns, user_input)
+                print(f"[THREAD] Using threaded context with {len(related_turns)} previous turns")
+
+            # Process text with threading support
+            outcome = handle_text_turn(user_input, text_reply_prompt, hints_pool, rescue_lines, threaded_context)
+
+            # Store CLEAN reply in history (without rescue lines/hints to prevent doubling)
+            clean_reply = strip_rescue_and_hint(outcome)
+            conv_history.add_turn(user_input, clean_reply)
+            print(f"[HISTORY] Stored clean reply (length: {len(clean_reply)} chars)")
+
+            # Check if it's a special outcome
+            if outcome == "EXIT":
+                return jsonify({
+                    "response": "Bye! See you next time~",
+                    "is_end": True
+                })
+
+            if outcome == "WIN":
+                return jsonify({
+                    "response": "Congratulations! You guessed the spell correctly! Happy Halloween!",
+                    "is_end": True
+                })
+
+            # Check if response contains image data URL
+            is_image = "data:image/" in outcome
+
+            return jsonify({
+                "response": outcome,
+                "is_end": False,
+                "is_image": is_image
+            })
+
+        except Exception as text_error:
+            error_msg = str(text_error)
+
+            # Check if it's a quota error
+            if '402' in error_msg or 'insufficient_quota' in error_msg or 'used up your points' in error_msg:
+                return jsonify({
+                    "response": "[QUOTA EXCEEDED] Your POE API quota is used up. Please visit https://poe.com/api_key to add more credits. The app will work normally once credits are added!",
+                    "is_end": False
+                }), 200  # Return 200 to show message properly
+
+            # Other errors
+            raise text_error
+
+    except Exception as e:
+        # Safe error printing
+        try:
+            print(f"[ERROR] in /chat: {str(e)}")
+        except:
+            print("[ERROR] in /chat: [error with special characters]")
+
+        import traceback
+        traceback.print_exc()
+
+        error_msg = str(e)
+
+        # User-friendly error messages
+        if '402' in error_msg or 'insufficient_quota' in error_msg:
+            return jsonify({
+                "error": "quota_exceeded",
+                "response": "[QUOTA EXCEEDED] POE API credits exhausted. Visit https://poe.com/api_key to add credits.",
+                "is_end": False
+            }), 200
+        else:
+            return jsonify({
+                "error": str(e),
+                "response": f"[ERROR] Processing request failed: {str(e)[:100]}",
+                "is_end": False
+            }), 500
+
+if __name__ == '__main__':
+    print("=" * 60)
+    print(" Pumpkin AI - HuggingFace Spaces")
+    print("=" * 60)
+    print(f"Prompt directory: {PROMPT_DIR}")
+    print(f"Static files: {static_folder}")
+    print(f"Hints loaded: {len(hints_pool)}")
+    print(f"Rescue lines loaded: {len(rescue_lines)}")
+    print("=" * 60 + "\n")
+
+    # Get port from environment (HuggingFace Spaces sets this)
+    port = int(os.environ.get('PORT', 7860))
+
+    # Run Flask app
+    app.run(
+        host='0.0.0.0',
+        port=port,
+        debug=False  # Disable debug in production
+    )

pumpkin-ai-frontend/build/asset-manifest.json

@@ -0,0 +1,13 @@
+{
+  "files": {
+    "main.css": "/static/css/main.5e2dec06.css",
+    "main.js": "/static/js/main.3a5310d2.js",
+    "index.html": "/index.html",
+    "main.5e2dec06.css.map": "/static/css/main.5e2dec06.css.map",
+    "main.3a5310d2.js.map": "/static/js/main.3a5310d2.js.map"
+  },
+  "entrypoints": [
+    "static/css/main.5e2dec06.css",
+    "static/js/main.3a5310d2.js"
+  ]
+}

pumpkin-ai-frontend/build/index.html

@@ -0,0 +1 @@
+<!doctype html><html lang="zh-HK"><head><meta charset="utf-8"/><meta name="viewport" content="width=device-width,initial-scale=1"/><meta name="theme-color" content="#000000"/><meta name="description" content="Pumpkin AI - 南瓜強迫症AI聊天室"/><title>🎃 南瓜AI聊天室</title><script defer="defer" src="/static/js/main.3a5310d2.js"></script><link href="/static/css/main.5e2dec06.css" rel="stylesheet"></head><body><noscript>You need to enable JavaScript to run this app.</noscript><div id="root"></div></body></html>

pumpkin-ai-frontend/build/static/css/main.5e2dec06.css

@@ -0,0 +1,2 @@
+*{box-sizing:border-box;padding:0}*,body{margin:0}body{-webkit-font-smoothing:antialiased;-moz-osx-font-smoothing:grayscale;font-family:-apple-system,BlinkMacSystemFont,Segoe UI,Roboto,Oxygen,Ubuntu,Cantarell,Fira Sans,Droid Sans,Helvetica Neue,sans-serif}code{font-family:source-code-pro,Menlo,Monaco,Consolas,Courier New,monospace}#root,.App{min-height:100vh}.App{background:linear-gradient(135deg,#1a1a2e,#16213e);color:#fff;display:flex;flex-direction:column}.App-header{background:linear-gradient(90deg,#ff6b35,#f7931e);box-shadow:0 4px 6px #0000004d;padding:1.5rem;text-align:center}.App-header h1{font-size:2.5rem;font-weight:700;margin:0;text-shadow:2px 2px 4px #0000004d}.subtitle{font-size:1rem;margin:.5rem 0 0;opacity:.9}.chat-container{display:flex;flex:1 1;flex-direction:column;margin:0 auto;max-width:900px;padding:1rem;width:100%}.messages-container{background:#ffffff0d;border-radius:12px;flex:1 1;margin-bottom:1rem;max-height:calc(100vh - 300px);overflow-y:auto;padding:1rem}.welcome-message{background:#ff6b351a;border:2px dashed #ff6b35;border-radius:12px;padding:2rem;text-align:left}.welcome-message p{font-size:1.1rem;font-weight:700;margin-top:0}.welcome-message ul{list-style:none;margin:1rem 0;padding:0}.welcome-message li{padding:.5rem 0 .5rem 1.5rem;position:relative}.welcome-message li:before{content:"🎃";left:0;position:absolute}.hint{font-style:italic;margin-bottom:0;opacity:.8;text-align:center}.message{animation:fadeIn .3s ease-in;display:flex;margin:1rem 0}@keyframes fadeIn{0%{opacity:0;transform:translateY(10px)}to{opacity:1;transform:translateY(0)}}.message.user{justify-content:flex-end}.message.ai{justify-content:flex-start}.message-content{word-wrap:break-word;border-radius:18px;max-width:70%;padding:1rem 1.5rem}.message.user .message-content{background:linear-gradient(135deg,#667eea,#764ba2);border-bottom-right-radius:4px;color:#fff}.message.ai .message-content{background:linear-gradient(135deg,#f093fb,#f5576c);border-bottom-left-radius:4px;color:#fff}.text-content{line-height:1.5;margin:0;white-space:pre-wrap}.image-container{border-radius:8px;box-shadow:0 4px 8px #0003;margin-top:1rem;overflow:hidden}.message-image{border-radius:8px;display:block;height:auto;max-width:100%}.image-error{color:#ffeb3b;font-style:italic;margin-top:.5rem}.loading{opacity:.7}.loading-dots{display:flex;gap:.3rem}.loading-dots span{animation:pulse 1.4s ease-in-out infinite both;font-size:2rem}.loading-dots span:first-child{animation-delay:-.32s}.loading-dots span:nth-child(2){animation-delay:-.16s}@keyframes pulse{0%,80%,to{opacity:.3;transform:scale(1)}40%{opacity:1;transform:scale(1.2)}}.input-container{align-items:center;background:#ffffff0d;border-radius:12px;display:flex;gap:.5rem;padding:1rem}.upload-button{background:linear-gradient(135deg,#667eea,#764ba2);border:none;border-radius:8px;color:#fff;cursor:pointer;font-size:1.5rem;padding:.75rem 1rem;transition:all .3s ease}.upload-button:hover:not(:disabled){box-shadow:0 4px 12px #667eea66;transform:scale(1.05)}.upload-button:disabled{cursor:not-allowed;opacity:.5}.text-input{background:#ffffff1a;border:2px solid #ffffff1a;border-radius:8px;color:#fff;flex:1 1;font-size:1rem;outline:none;padding:.75rem 1rem;transition:border-color .3s ease}.text-input:focus{border-color:#ff6b35}.text-input::placeholder{color:#ffffff80}.text-input:disabled{cursor:not-allowed;opacity:.5}.send-button{background:linear-gradient(90deg,#ff6b35,#f7931e);border:none;border-radius:8px;color:#fff;cursor:pointer;font-size:1rem;font-weight:700;padding:.75rem 1.5rem;transition:all .3s ease}.send-button:hover:not(:disabled){box-shadow:0 4px 12px #ff6b3566;transform:translateY(-2px)}.send-button:disabled{cursor:not-allowed;opacity:.5}.App-footer{background:#0000004d;font-size:.875rem;opacity:.7;padding:1rem;text-align:center}.App-footer p{margin:0}.messages-container::-webkit-scrollbar{width:8px}.messages-container::-webkit-scrollbar-track{background:#ffffff0d;border-radius:4px}.messages-container::-webkit-scrollbar-thumb{background:#ff6b3580;border-radius:4px}.messages-container::-webkit-scrollbar-thumb:hover{background:#ff6b35b3}.App:has(.messages-container:hover){outline:2px dashed #ff6b35;outline-offset:-10px}
+/*# sourceMappingURL=main.5e2dec06.css.map*/

pumpkin-ai-frontend/build/static/css/main.5e2dec06.css.map

@@ -0,0 +1 @@
+{"version":3,"file":"static/css/main.5e2dec06.css","mappings":"AAAA,EACE,qBAAsB,CAEtB,SACF,CAEA,OAJE,QAWF,CAPA,KAKE,kCAAmC,CACnC,iCAAkC,CAJlC,mIAKF,CAEA,KACE,uEAEF,CClBA,WDqBE,gBCfF,CANA,KAEE,kDAA6D,CAG7D,UAAW,CAFX,YAAa,CACb,qBAEF,CAEA,YACE,iDAA4D,CAG5D,8BAAwC,CAFxC,cAAe,CACf,iBAEF,CAEA,eAEE,gBAAiB,CACjB,eAAiB,CAFjB,QAAS,CAGT,iCACF,CAEA,UAEE,cAAe,CADf,gBAAoB,CAEpB,UACF,CAEA,gBAEE,YAAa,CADb,QAAO,CAEP,qBAAsB,CAGtB,aAAc,CAFd,eAAgB,CAGhB,YAAa,CAFb,UAGF,CAEA,oBAIE,oBAAqC,CACrC,kBAAmB,CAJnB,QAAO,CAKP,kBAAmB,CACnB,8BAA+B,CAL/B,eAAgB,CAChB,YAKF,CAEA,iBACE,oBAAmC,CACnC,yBAA0B,CAC1B,kBAAmB,CACnB,YAAa,CACb,eACF,CAEA,mBAEE,gBAAiB,CACjB,eAAiB,CAFjB,YAGF,CAEA,oBACE,eAAgB,CAEhB,aAAc,CADd,SAEF,CAEA,oBAEE,4BAAoB,CACpB,iBACF,CAEA,2BACE,YAAa,CAEb,MAAO,CADP,iBAEF,CAEA,MAEE,iBAAkB,CAElB,eAAgB,CADhB,UAAY,CAFZ,iBAIF,CAEA,SAGE,4BAA8B,CAD9B,YAAa,CADb,aAGF,CAEA,kBACE,GACE,SAAU,CACV,0BACF,CACA,GACE,SAAU,CACV,uBACF,CACF,CAEA,cACE,wBACF,CAEA,YACE,0BACF,CAEA,iBAIE,oBAAqB,CADrB,kBAAmB,CAFnB,aAAc,CACd,mBAGF,CAEA,+BACE,kDAA6D,CAE7D,8BAA+B,CAD/B,UAEF,CAEA,6BACE,kDAA6D,CAE7D,6BAA8B,CAD9B,UAEF,CAEA,cAGE,eAAgB,CAFhB,QAAS,CACT,oBAEF,CAEA,iBAEE,iBAAkB,CAElB,0BAAwC,CAHxC,eAAgB,CAEhB,eAEF,CAEA,eAIE,iBAAkB,CADlB,aAAc,CADd,WAAY,CADZ,cAIF,CAEA,aACE,aAAc,CACd,iBAAkB,CAClB,gBACF,CAEA,SACE,UACF,CAEA,cACE,YAAa,CACb,SACF,CAEA,mBACE,8CAA+C,CAC/C,cACF,CAEA,+BACE,qBACF,CAEA,gCACE,qBACF,CAEA,iBACE,UACE,UAAY,CACZ,kBACF,CACA,IACE,SAAU,CACV,oBACF,CACF,CAEA,iBAME,kBAAmB,CAFnB,oBAAqC,CACrC,kBAAmB,CAJnB,YAAa,CACb,SAAW,CACX,YAIF,CAEA,eACE,kDAA6D,CAC7D,WAAY,CAIZ,iBAAkB,CAHlB,UAAY,CAIZ,cAAe,CAHf,gBAAiB,CACjB,mBAAqB,CAGrB,uBACF,CAEA,oCAEE,+BAA+C,CAD/C,qBAEF,CAEA,wBAEE,kBAAmB,CADnB,UAEF,CAEA,YAKE,oBAAoC,CAFpC,0BAA0C,CAC1C,iBAAkB,CAElB,UAAY,CALZ,QAAO,CAMP,cAAe,CACf,YAAa,CANb,mBAAqB,CAOrB,gCACF,CAEA,kBACE,oBACF,CAEA,yBACE,eACF,CAEA,qBAEE,kBAAmB,CADnB,UAEF,CAEA,aACE,iDAA4D,CAC5D,WAAY,CAKZ,iBAAkB,CAJlB,UAAY,CAKZ,cAAe,CAJf,cAAe,CACf,eAAiB,CACjB,qBAAuB,CAGvB,uBACF,CAEA,kCAEE,+BAA8C,CAD9C,0BAEF,CAEA,sBAEE,kBAAmB,CADnB,UAEF,CAEA,YACE,oBAA8B,CAG9B,iBAAmB,CACnB,UAAY,CAHZ,YAAa,CACb,iBAGF,CAEA,cACE,QACF,CAGA,uCACE,SACF,CAEA,6CACE,oBAAqC,CACrC,iBACF,CAEA,6CACE,oBAAmC,CACnC,iBACF,CAEA,mDACE,oBACF,CAGA,oCACE,0BAA2B,CAC3B,oBACF","sources":["index.css","App.css"],"sourcesContent":["* {\r\n  box-sizing: border-box;\r\n  margin: 0;\r\n  padding: 0;\r\n}\r\n\r\nbody {\r\n  margin: 0;\r\n  font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'Roboto', 'Oxygen',\r\n    'Ubuntu', 'Cantarell', 'Fira Sans', 'Droid Sans', 'Helvetica Neue',\r\n    sans-serif;\r\n  -webkit-font-smoothing: antialiased;\r\n  -moz-osx-font-smoothing: grayscale;\r\n}\r\n\r\ncode {\r\n  font-family: source-code-pro, Menlo, Monaco, Consolas, 'Courier New',\r\n    monospace;\r\n}\r\n\r\n#root {\r\n  min-height: 100vh;\r\n}\r\n",".App {\r\n  min-height: 100vh;\r\n  background: linear-gradient(135deg, #1a1a2e 0%, #16213e 100%);\r\n  display: flex;\r\n  flex-direction: column;\r\n  color: #fff;\r\n}\r\n\r\n.App-header {\r\n  background: linear-gradient(90deg, #ff6b35 0%, #f7931e 100%);\r\n  padding: 1.5rem;\r\n  text-align: center;\r\n  box-shadow: 0 4px 6px rgba(0, 0, 0, 0.3);\r\n}\r\n\r\n.App-header h1 {\r\n  margin: 0;\r\n  font-size: 2.5rem;\r\n  font-weight: bold;\r\n  text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.3);\r\n}\r\n\r\n.subtitle {\r\n  margin: 0.5rem 0 0 0;\r\n  font-size: 1rem;\r\n  opacity: 0.9;\r\n}\r\n\r\n.chat-container {\r\n  flex: 1;\r\n  display: flex;\r\n  flex-direction: column;\r\n  max-width: 900px;\r\n  width: 100%;\r\n  margin: 0 auto;\r\n  padding: 1rem;\r\n}\r\n\r\n.messages-container {\r\n  flex: 1;\r\n  overflow-y: auto;\r\n  padding: 1rem;\r\n  background: rgba(255, 255, 255, 0.05);\r\n  border-radius: 12px;\r\n  margin-bottom: 1rem;\r\n  max-height: calc(100vh - 300px);\r\n}\r\n\r\n.welcome-message {\r\n  background: rgba(255, 107, 53, 0.1);\r\n  border: 2px dashed #ff6b35;\r\n  border-radius: 12px;\r\n  padding: 2rem;\r\n  text-align: left;\r\n}\r\n\r\n.welcome-message p {\r\n  margin-top: 0;\r\n  font-size: 1.1rem;\r\n  font-weight: bold;\r\n}\r\n\r\n.welcome-message ul {\r\n  list-style: none;\r\n  padding: 0;\r\n  margin: 1rem 0;\r\n}\r\n\r\n.welcome-message li {\r\n  padding: 0.5rem 0;\r\n  padding-left: 1.5rem;\r\n  position: relative;\r\n}\r\n\r\n.welcome-message li:before {\r\n  content: \"🎃\";\r\n  position: absolute;\r\n  left: 0;\r\n}\r\n\r\n.hint {\r\n  text-align: center;\r\n  font-style: italic;\r\n  opacity: 0.8;\r\n  margin-bottom: 0;\r\n}\r\n\r\n.message {\r\n  margin: 1rem 0;\r\n  display: flex;\r\n  animation: fadeIn 0.3s ease-in;\r\n}\r\n\r\n@keyframes fadeIn {\r\n  from {\r\n    opacity: 0;\r\n    transform: translateY(10px);\r\n  }\r\n  to {\r\n    opacity: 1;\r\n    transform: translateY(0);\r\n  }\r\n}\r\n\r\n.message.user {\r\n  justify-content: flex-end;\r\n}\r\n\r\n.message.ai {\r\n  justify-content: flex-start;\r\n}\r\n\r\n.message-content {\r\n  max-width: 70%;\r\n  padding: 1rem 1.5rem;\r\n  border-radius: 18px;\r\n  word-wrap: break-word;\r\n}\r\n\r\n.message.user .message-content {\r\n  background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);\r\n  color: white;\r\n  border-bottom-right-radius: 4px;\r\n}\r\n\r\n.message.ai .message-content {\r\n  background: linear-gradient(135deg, #f093fb 0%, #f5576c 100%);\r\n  color: white;\r\n  border-bottom-left-radius: 4px;\r\n}\r\n\r\n.text-content {\r\n  margin: 0;\r\n  white-space: pre-wrap;\r\n  line-height: 1.5;\r\n}\r\n\r\n.image-container {\r\n  margin-top: 1rem;\r\n  border-radius: 8px;\r\n  overflow: hidden;\r\n  box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);\r\n}\r\n\r\n.message-image {\r\n  max-width: 100%;\r\n  height: auto;\r\n  display: block;\r\n  border-radius: 8px;\r\n}\r\n\r\n.image-error {\r\n  color: #ffeb3b;\r\n  font-style: italic;\r\n  margin-top: 0.5rem;\r\n}\r\n\r\n.loading {\r\n  opacity: 0.7;\r\n}\r\n\r\n.loading-dots {\r\n  display: flex;\r\n  gap: 0.3rem;\r\n}\r\n\r\n.loading-dots span {\r\n  animation: pulse 1.4s infinite ease-in-out both;\r\n  font-size: 2rem;\r\n}\r\n\r\n.loading-dots span:nth-child(1) {\r\n  animation-delay: -0.32s;\r\n}\r\n\r\n.loading-dots span:nth-child(2) {\r\n  animation-delay: -0.16s;\r\n}\r\n\r\n@keyframes pulse {\r\n  0%, 80%, 100% {\r\n    opacity: 0.3;\r\n    transform: scale(1);\r\n  }\r\n  40% {\r\n    opacity: 1;\r\n    transform: scale(1.2);\r\n  }\r\n}\r\n\r\n.input-container {\r\n  display: flex;\r\n  gap: 0.5rem;\r\n  padding: 1rem;\r\n  background: rgba(255, 255, 255, 0.05);\r\n  border-radius: 12px;\r\n  align-items: center;\r\n}\r\n\r\n.upload-button {\r\n  background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);\r\n  border: none;\r\n  color: white;\r\n  font-size: 1.5rem;\r\n  padding: 0.75rem 1rem;\r\n  border-radius: 8px;\r\n  cursor: pointer;\r\n  transition: all 0.3s ease;\r\n}\r\n\r\n.upload-button:hover:not(:disabled) {\r\n  transform: scale(1.05);\r\n  box-shadow: 0 4px 12px rgba(102, 126, 234, 0.4);\r\n}\r\n\r\n.upload-button:disabled {\r\n  opacity: 0.5;\r\n  cursor: not-allowed;\r\n}\r\n\r\n.text-input {\r\n  flex: 1;\r\n  padding: 0.75rem 1rem;\r\n  border: 2px solid rgba(255, 255, 255, 0.1);\r\n  border-radius: 8px;\r\n  background: rgba(255, 255, 255, 0.1);\r\n  color: white;\r\n  font-size: 1rem;\r\n  outline: none;\r\n  transition: border-color 0.3s ease;\r\n}\r\n\r\n.text-input:focus {\r\n  border-color: #ff6b35;\r\n}\r\n\r\n.text-input::placeholder {\r\n  color: rgba(255, 255, 255, 0.5);\r\n}\r\n\r\n.text-input:disabled {\r\n  opacity: 0.5;\r\n  cursor: not-allowed;\r\n}\r\n\r\n.send-button {\r\n  background: linear-gradient(90deg, #ff6b35 0%, #f7931e 100%);\r\n  border: none;\r\n  color: white;\r\n  font-size: 1rem;\r\n  font-weight: bold;\r\n  padding: 0.75rem 1.5rem;\r\n  border-radius: 8px;\r\n  cursor: pointer;\r\n  transition: all 0.3s ease;\r\n}\r\n\r\n.send-button:hover:not(:disabled) {\r\n  transform: translateY(-2px);\r\n  box-shadow: 0 4px 12px rgba(255, 107, 53, 0.4);\r\n}\r\n\r\n.send-button:disabled {\r\n  opacity: 0.5;\r\n  cursor: not-allowed;\r\n}\r\n\r\n.App-footer {\r\n  background: rgba(0, 0, 0, 0.3);\r\n  padding: 1rem;\r\n  text-align: center;\r\n  font-size: 0.875rem;\r\n  opacity: 0.7;\r\n}\r\n\r\n.App-footer p {\r\n  margin: 0;\r\n}\r\n\r\n/* Scrollbar styling */\r\n.messages-container::-webkit-scrollbar {\r\n  width: 8px;\r\n}\r\n\r\n.messages-container::-webkit-scrollbar-track {\r\n  background: rgba(255, 255, 255, 0.05);\r\n  border-radius: 4px;\r\n}\r\n\r\n.messages-container::-webkit-scrollbar-thumb {\r\n  background: rgba(255, 107, 53, 0.5);\r\n  border-radius: 4px;\r\n}\r\n\r\n.messages-container::-webkit-scrollbar-thumb:hover {\r\n  background: rgba(255, 107, 53, 0.7);\r\n}\r\n\r\n/* Drag and drop styling */\r\n.App:has(.messages-container:hover) {\r\n  outline: 2px dashed #ff6b35;\r\n  outline-offset: -10px;\r\n}\r\n"],"names":[],"sourceRoot":""}

pumpkin-ai-frontend/build/static/js/main.3a5310d2.js.LICENSE.txt

@@ -0,0 +1,49 @@
+/**
+ * @license React
+ * react-dom-client.production.js
+ *
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+/**
+ * @license React
+ * react-dom.production.js
+ *
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+/**
+ * @license React
+ * react-jsx-runtime.production.js
+ *
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+/**
+ * @license React
+ * react.production.js
+ *
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+/**
+ * @license React
+ * scheduler.production.js
+ *
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */

pumpkin_code.py

@@ -0,0 +1,939 @@
+# -*- coding: utf-8 -*-
+"""
+🎃 Pumpkin AI Console — Full Version (Text + Image + Hint System)
+Author: Cherry Leung, Jay Luk
+Date: 2025-10-13
+"""
+
+import sys
+import os
+import re
+import base64
+import random
+import unicodedata as _ud  # for robust Han-only normalization
+from pathlib import Path
+import openai
+import logging
+from datetime import datetime
+
+# Try to load .env file if python-dotenv is available
+try:
+    from dotenv import load_dotenv
+    load_dotenv()
+except ImportError:
+    pass  # .env file won't be loaded automatically, use system environment variables
+
+# ========================
+#  LOGGING SETUP
+# ========================
+
+# Configure logging to show DEBUG level messages with timestamp
+logging.basicConfig(
+    level=logging.DEBUG,  # Changed to DEBUG for detailed logging
+    format='%(asctime)s [%(levelname)s] %(message)s',
+    datefmt='%Y-%m-%d %H:%M:%S'
+)
+logger = logging.getLogger(__name__)
+
+# ========================
+#  CONFIGURATION
+# ========================
+
+POE_API_KEY = os.getenv("POE_API_KEY") or "BZyOjKtDmmX7Day-BGjIIAcPmqozr-cRY5V3ZrDDj5Q"
+if POE_API_KEY == "BZyOjKtDmmX7Day-BGjIIAcPmqozr-cRY5V3ZrDDj5Q":
+    logger.warning("Using fallback POE_API_KEY. Set POE_API_KEY environment variable for production.")
+
+BASE_URL = "https://api.poe.com/v1"
+
+TEXT_MODEL  = "gpt-4o-mini"
+IMAGE_MODEL = "Nano-Banana"  # Switched to Nano-Banana for better pumpkin integration
+TIMEOUT = 60
+
+logger.info(f"Initializing OpenAI client with base URL: {BASE_URL}")
+client = openai.OpenAI(api_key=POE_API_KEY, base_url=BASE_URL)
+logger.info("OpenAI client initialized successfully")
+
+# ========================
+# REGEX
+# ========================
+
+STATUS_NO_RX   = re.compile(r"\{status(?:\s*code)?:\s*no\}", re.IGNORECASE)
+STATUS_YES_RX  = re.compile(r"\{status(?:\s*code)?:\s*yes\}", re.IGNORECASE)
+DATA_URL_RX    = re.compile(r"data:image/(png|jpeg|jpg|webp);base64,([A-Za-z0-9+/=]+)", re.IGNORECASE)
+
+# ========================
+#  PROMPTS / HINTS / RESCUE LINES
+# ========================
+
+def load_text(path: Path) -> str:
+    """Read text file safely (UTF-8-SIG to remove BOM)."""
+    logger.info(f"Loading text file: {path}")
+    return path.read_text(encoding="utf-8-sig", errors="ignore")
+
+def load_prompts(prompt_dir: Path):
+    classify = load_text(prompt_dir / "text_or_image_classification.txt")
+    textrep  = load_text(prompt_dir / "text_replies.txt")
+    imagerep = load_text(prompt_dir / "image_replies.txt")
+    hintsraw = load_text(prompt_dir / "hints.txt")
+    return classify, textrep, imagerep, hintsraw
+
+def load_hints_simple(file_path: str, limit: int = 10) -> list[str]:
+    """
+    Read hints.txt formatted as one hint per line.
+    Lines starting with # or empty lines are ignored.
+    Returns up to `limit` hints.
+    """
+    path = Path(file_path)
+    if not path.exists():
+        logger.warning(f"File not found: {file_path}")
+        print(f"⚠️ File not found: {file_path}")
+        return []
+
+    logger.info(f"Loading hints from: {file_path}")
+    text = path.read_text(encoding="utf-8-sig", errors="ignore")
+    lines = [
+        ln.strip()
+        for ln in text.splitlines()
+        if ln.strip() and not ln.strip().startswith("#")
+    ]
+    logger.info(f"Loaded {len(lines)} hints (returning up to {limit})")
+    return lines[:limit]
+
+def load_rescue_lines(file_path: str) -> list[str]:
+    """
+    Load preset rescue lines from a text file (rescue_line.txt).
+    Ignores empty lines and lines starting with '#'.
+    """
+    path = Path(file_path)
+    if not path.exists():
+        logger.warning(f"Rescue lines file not found: {file_path}")
+        print(f"⚠️ File not found: {file_path}")
+        return []
+
+    logger.info(f"Loading rescue lines from: {file_path}")
+    lines = [
+        ln.strip()
+        for ln in path.read_text(encoding="utf-8-sig", errors="ignore").splitlines()
+        if ln.strip() and not ln.strip().startswith("#")
+    ]
+    logger.info(f"Loaded {len(lines)} rescue lines")
+    return lines
+
+# ========================
+#  MODEL CALL HELPERS
+# ========================
+
+def chat_completion(model: str, system_prompt: str, user_content: str, temperature: float = 0.0) -> str:
+    logger.info(f"Making chat completion request to model: {model} (temp={temperature})")
+    logger.debug(f"User content preview: {user_content[:100]}...")
+
+    chat = client.chat.completions.create(
+        model=model,
+        messages=[
+            {"role": "system", "content": system_prompt},
+            {"role": "user",  "content": user_content},
+        ],
+        temperature=temperature,
+        timeout=TIMEOUT,
+    )
+
+    response = chat.choices[0].message.content.strip()
+    logger.info(f"Received response from {model}, length: {len(response)} chars")
+    logger.debug(f"Response preview: {response[:200]}...")
+
+    return response
+
+def classify_input(user_input: str, system_prompt: str) -> str:
+    logger.info("Classifying user input (text vs image request)")
+    out = chat_completion(
+        model=TEXT_MODEL,
+        system_prompt=system_prompt,
+        user_content=user_input,
+        temperature=0.0,
+    ).lower()
+    result = "image" if "image" in out else "text"
+    logger.info(f"Classification result: {result}")
+    return result
+
+def generate_text_reply(user_input: str, text_reply_prompt: str, threaded_context: str = None) -> str:
+    """
+    Generate text reply with optional threaded conversation context.
+    If threaded_context is provided, it will be prepended to the user input.
+    """
+    logger.info("Generating text reply")
+
+    # If we have threaded context, include it
+    if threaded_context:
+        logger.info("Using threaded context for reply generation")
+        user_content = threaded_context
+    else:
+        user_content = user_input
+
+    return chat_completion(
+        model=TEXT_MODEL,
+        system_prompt=text_reply_prompt,
+        user_content=user_content,
+        temperature=0.9,
+    )
+
+def generate_image_data_url(user_input: str, image_reply_prompt: str) -> str:
+    """
+    Ask the image model to generate image and return as data URL.
+    Handles both data URLs and HTTP URLs from the model.
+    """
+    logger.info(f"Generating image with model: {IMAGE_MODEL}")
+    prompt = (
+        image_reply_prompt
+        + "\n\n【使用者提示】" + user_input
+        + "\n\n請只輸出一條 data URL（data:image/png;base64,XXXXX）。不要文字、不要Markdown、不要說明。"
+    )
+    logger.debug(f"Image generation prompt length: {len(prompt)} chars")
+
+    msg = client.chat.completions.create(
+        model=IMAGE_MODEL,
+        messages=[
+            {"role": "system", "content": "你是嚴格的影像生成器。只產出base64圖像的data URL，不要任何額外文字。"},
+            {"role": "user",   "content": prompt},
+        ],
+        temperature=0.8,
+        timeout=TIMEOUT,
+    )
+
+    response = msg.choices[0].message.content.strip()
+    logger.info(f"Image generation response received, length: {len(response)} chars")
+    logger.debug(f"Response preview: {response[:200]}...")
+
+    # Check if response already has data URL
+    if "data:image/" in response:
+        logger.info("Response contains data URL")
+        return response
+
+    # Extract HTTP URL from response (markdown or plain)
+    image_url = extract_image_url_from_response(response)
+    if image_url:
+        logger.info(f"Found HTTP URL, attempting to download: {image_url}")
+        data_url = download_image_as_data_url(image_url)
+        if data_url:
+            logger.info("Successfully converted HTTP URL to data URL")
+            return data_url
+        else:
+            logger.error("Failed to download image from URL")
+            return response  # Return original response for debugging
+
+    logger.warning("No image URL or data URL found in response")
+    return response  # Return as-is for debugging
+
+def extract_image_url_from_response(response: str) -> str:
+    """
+    Extract image URL from markdown or plain text response.
+    Handles responses like: ![...](https://pfst.cf2.poecdn.net/...)
+    """
+    logger.info("Attempting to extract image URL from response")
+    logger.debug(f"Full response to parse: {response}")
+
+    # First, normalize the response by removing newlines to handle broken markdown
+    normalized_response = response.replace('\n', ' ').replace('\r', '')
+
+    # Try markdown format first: ![text](url)
+    # Use non-greedy match and capture everything until closing paren
+    markdown_match = re.search(r'!\[.*?\]\((https://[^\)]+)\)', normalized_response, re.DOTALL)
+    if markdown_match:
+        url = markdown_match.group(1).strip()
+        logger.info(f"Extracted URL from markdown: {url}")
+        return url
+
+    # Try plain URL format with query parameters (with or without newlines)
+    # Match the entire URL including query params (?w=..&h=..)
+    url_match = re.search(r'(https://pfst\.cf2\.poecdn\.net/[^\s\)\]]+)', normalized_response)
+    if url_match:
+        url = url_match.group(1).strip()
+        logger.info(f"Extracted plain URL: {url}")
+        return url
+
+    # Try to find ANY https URL
+    any_url_match = re.search(r'(https://[^\s\)\]<>"\']+)', normalized_response)
+    if any_url_match:
+        url = any_url_match.group(1).strip()
+        # Remove trailing punctuation
+        url = url.rstrip('.,;!?')
+        logger.info(f"Extracted generic URL: {url}")
+        return url
+
+    logger.warning("No image URL found in response")
+    return None
+
+def download_image_as_data_url(url: str) -> str:
+    """
+    Download image from URL and convert to base64 data URL.
+    Includes proper headers to bypass CDN restrictions.
+    """
+    logger.info(f"Downloading image from: {url}")
+    try:
+        import urllib.request
+
+        # Create request with headers to bypass 403 Forbidden
+        req = urllib.request.Request(
+            url,
+            headers={
+                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
+                'Accept': 'image/avif,image/webp,image/apng,image/*,*/*;q=0.8',
+                'Accept-Language': 'en-US,en;q=0.9',
+                'Referer': 'https://poe.com/',
+                'Origin': 'https://poe.com'
+            }
+        )
+
+        response = urllib.request.urlopen(req, timeout=30)
+        image_data = response.read()
+        logger.info(f"Downloaded {len(image_data)} bytes")
+
+        # Convert to base64
+        b64 = base64.b64encode(image_data).decode('utf-8')
+        data_url = f"data:image/png;base64,{b64}"
+        logger.info(f"Converted to data URL, length: {len(data_url)}")
+        return data_url
+    except Exception as e:
+        logger.error(f"Failed to download image: {e}")
+        return None
+def analyze_image_detailed(image_path: str) -> str:
+    """
+    Analyze an image with comprehensive detail extraction.
+    Returns structured description of all elements in the image.
+    """
+    logger.info(f"Analyzing image with detailed prompt: {image_path}")
+
+    # Read image and convert to base64
+    image_data = Path(image_path).read_bytes()
+    b64_image = base64.b64encode(image_data).decode('utf-8')
+
+    # Detect image format
+    ext = Path(image_path).suffix.lower()
+    mime_type = "image/jpeg" if ext in [".jpg", ".jpeg"] else f"image/{ext[1:]}"
+
+    detailed_prompt = """請詳細分析這張圖片，並提供全面的描述。請按以下結構組織你的描述：
+
+【人物分析】
+- 人數統計：共有幾位人物
+- 對於每位人物，請描述：
+  * 年齡範圍（例如：嬰兒、兒童、青少年、青年、中年、老年）
+  * 性別
+  * 族裔特徵
+  * 面部特徵（例如：髮型、髮色、眼睛顏色、表情）
+  * 服裝描述（顏色、風格、配飾）
+  * 姿勢與動作
+  * 整體氛圍/感覺
+  * 如果可辨識為知名人物，請提及姓名
+
+【動物分析】（如果有）
+- 動物種類與品種
+- 年齡階段（幼年/成年/老年）
+- 顏色與花紋
+- 姿勢與動作
+- 與環境或人物的互動
+
+【環境與場景】
+- 地點類型（室內/室外、具體場所）
+- 時間（白天/夜晚、季節）
+- 光線與氛圍
+- 背景元素描述
+
+【物品與道具】
+- 主要物品列表及其特徵
+- 物品的顏色、材質、狀態
+- 物品在畫面中的位置與作用
+
+【整體構圖】
+- 拍攝角度與視角
+- 畫面重點與焦點
+- 色調與風格
+- 情緒與氛圍
+
+請盡可能詳細且有條理地描述所有可見元素。"""
+
+    try:
+        response = client.chat.completions.create(
+            model=IMAGE_MODEL,
+            messages=[
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": detailed_prompt},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:{mime_type};base64,{b64_image}"
+                            }
+                        }
+                    ]
+                }
+            ],
+            temperature=0.3,
+            timeout=TIMEOUT,
+        )
+
+        analysis = response.choices[0].message.content.strip()
+        logger.info(f"Image analysis completed, length: {len(analysis)} chars")
+        return analysis
+
+    except Exception as e:
+        logger.error(f"Failed to analyze image: {e}")
+        return f"⚠️ 圖片分析失敗：{str(e)}"
+    
+def analyze_image_from_data_url(data_url: str) -> str:
+    """
+    Analyze image from data URL format.
+    Saves temporarily, analyzes, then cleans up.
+    """
+    logger.info("Analyzing image from data URL")
+    temp_path = "temp_analysis.png"
+
+    try:
+        # Save temporarily
+        ok = save_data_url_to_file(data_url, temp_path)
+        if not ok:
+            logger.error("Failed to save data URL to temp file")
+            return "⚠️ 無法處理上傳的圖片"
+
+        # Analyze
+        analysis = analyze_image_detailed(temp_path)
+
+        # Cleanup
+        try:
+            Path(temp_path).unlink()
+            logger.info("Cleaned up temp analysis file")
+        except Exception as e:
+            logger.warning(f"Could not delete temp file: {e}")
+
+        return analysis
+
+    except Exception as e:
+        logger.error(f"Failed to analyze image from data URL: {e}")
+        return f"⚠️ 圖片分析失敗：{str(e)}"
+
+def generate_with_nano_banana(prompt: str) -> str:
+    """
+    Generate image using Nano-Banana model.
+    Returns data URL or HTTP URL (which will be converted).
+    """
+    logger.info("Generating image with Nano-Banana model")
+    logger.debug(f"Prompt length: {len(prompt)} chars")
+
+    try:
+        msg = client.chat.completions.create(
+            model="Nano-Banana",  # Use Nano-Banana instead of Gemini
+            messages=[
+                {"role": "system", "content": "你是圖像生成助手。根據描述生成圖片。"},
+                {"role": "user", "content": prompt},
+            ],
+            temperature=0.8,
+            timeout=TIMEOUT,
+        )
+
+        response = msg.choices[0].message.content.strip()
+        logger.info(f"Nano-Banana response received, length: {len(response)} chars")
+
+        # Check if response already has data URL
+        if "data:image/" in response:
+            logger.info("Response contains data URL")
+            return response
+
+        # Extract and download HTTP URL
+        image_url = extract_image_url_from_response(response)
+        if image_url:
+            logger.info(f"Found HTTP URL, downloading: {image_url}")
+            data_url = download_image_as_data_url(image_url)
+            if data_url:
+                logger.info("Successfully converted HTTP URL to data URL")
+                return data_url
+
+        logger.warning("No valid image data in Nano-Banana response")
+        return response  # Return as-is for debugging
+
+    except Exception as e:
+        logger.error(f"Nano-Banana generation failed: {e}")
+        return f"⚠️ 圖片生成失敗：{str(e)}"
+
+def generate_image_with_analysis(user_input: str, image_reply_prompt: str, uploaded_image_data: str = None, conversation_context: str = None) -> str:
+    """
+    Generate image from text description with optional image analysis and conversation threading.
+
+    If uploaded_image_data provided:
+      Analyze the image first, then use that analysis to generate a new image
+
+    If conversation_context provided:
+      Include previous conversation history to generate contextually relevant images
+
+    If no uploaded_image_data:
+      Use text description directly
+    """
+    logger.info("Starting image generation flow")
+
+    if conversation_context:
+        logger.info("Using conversation threading for image generation")
+        logger.debug(f"Conversation context length: {len(conversation_context)} chars")
+
+    if uploaded_image_data:
+        # User uploaded image - try to analyze it first
+        logger.info("User uploaded image - attempting to analyze image content...")
+
+        # Try to analyze the uploaded image
+        analysis_result = analyze_image_from_data_url(uploaded_image_data)
+
+        logger.info("=" * 80)
+        logger.info("IMAGE ANALYSIS RESULT:")
+        logger.info("=" * 80)
+        logger.info(analysis_result)
+        logger.info("=" * 80)
+
+        # Check if analysis failed
+        if "無法分析" in analysis_result or "無法識別" in analysis_result or "抱歉" in analysis_result:
+            logger.warning("Image analysis failed - POE vision API not working properly")
+            logger.info("Will generate pumpkin image based on text description only")
+
+            # Fallback: Use text description only, mention that image was uploaded
+            # Include conversation context if available
+            context_section = f"\n\n{conversation_context}" if conversation_context else ""
+
+            enhanced_prompt = f"""{image_reply_prompt}
+
+【使用者提示】
+{user_input}
+
+（用戶上傳了一張圖片，但圖片分析功能暫時不可用。請根據文字描述生成一張充滿南瓜元素的創意圖片。）
+{context_section}
+
+請生成一張包含南瓜元素的新圖片。
+
+請只輸出一張圖片URL或data URL。"""
+        else:
+            # Analysis succeeded - use it
+            logger.info("Image analysis successful - using results for generation")
+
+            # Include conversation context if available
+            context_section = f"\n\n{conversation_context}" if conversation_context else ""
+
+            enhanced_prompt = f"""{image_reply_prompt}
+
+【使用者提示】
+{user_input}
+
+【圖片分析結果】
+以下是用戶上傳圖片的詳細分析：
+{analysis_result}
+{context_section}
+
+請根據以上圖片分析結果，生成一張包含南瓜元素的新圖片。
+保留原圖的主要構圖和元素，但加入南瓜相關的創意元素。
+{f'請考慮對話歷史中的上下文，讓圖片與之前的對話內容相關聯。' if conversation_context else ''}
+
+請只輸出一張圖片URL或data URL。"""
+
+        logger.info("=" * 80)
+        logger.info("FULL PROMPT SENT TO NANO-BANANA:")
+        logger.info("=" * 80)
+        logger.info(enhanced_prompt)
+        logger.info("=" * 80)
+    else:
+        # No image uploaded, use text description
+        logger.info("No uploaded image - generating from text description")
+
+        # Include conversation context if available
+        context_section = f"\n\n{conversation_context}" if conversation_context else ""
+
+        enhanced_prompt = f"""{image_reply_prompt}
+
+【使用者提示】
+{user_input}
+{context_section}
+
+請生成包含南瓜元素的圖片。
+{f'請考慮對話歷史中的上下文，讓圖片與之前的對話內容相關聯。' if conversation_context else ''}
+
+請只輸出一張圖片URL或data URL。"""
+
+        logger.info("=" * 80)
+        logger.info("PROMPT SENT TO NANO-BANANA:")
+        logger.info("=" * 80)
+        logger.info(enhanced_prompt)
+        logger.info("=" * 80)
+
+    # Use Nano-Banana model for generation
+    logger.info("Calling Nano-Banana for image generation...")
+    return generate_with_nano_banana(enhanced_prompt)
+
+def save_data_url_to_file(data_url: str, output_path: str) -> bool:
+    logger.info(f"Attempting to save data URL to file: {output_path}")
+    m = DATA_URL_RX.search(data_url)
+    if not m:
+        logger.error("Failed to extract base64 data from data URL")
+        return False
+    b64 = m.group(2)
+    Path(output_path).write_bytes(base64.b64decode(b64))
+    logger.info(f"Successfully saved image to: {output_path}")
+    return True
+
+
+
+# ========================
+#  CONVERSATION THREADING
+# ========================
+
+class ConversationHistory:
+    """
+    Manages conversation history and threading.
+    Maintains recent messages and determines when to thread related conversations.
+    """
+    def __init__(self, max_history: int = 5):
+        self.max_history = max_history
+        self.history = []  # List of (user_msg, ai_reply) tuples
+
+    def add_turn(self, user_msg: str, ai_reply: str):
+        """Add a conversation turn to history."""
+        self.history.append({"user": user_msg, "ai": ai_reply})
+        if len(self.history) > self.max_history:
+            self.history.pop(0)
+        logger.debug(f"Added turn to history. Total turns: {len(self.history)}")
+
+    def get_history(self):
+        """Get all conversation history."""
+        return self.history
+
+    def check_threading(self, new_message: str) -> tuple[bool, list]:
+        """
+        Check if new message relates to recent conversation.
+        Returns (should_thread, related_messages).
+        """
+        if len(self.history) == 0:
+            logger.info("No conversation history - processing message in isolation")
+            return False, []
+
+        logger.info(f"Checking threading relationships for new message (history size: {len(self.history)})")
+
+        # Build a prompt to check relationships
+        history_text = ""
+        for i, turn in enumerate(self.history[-3:], 1):  # Check last 3 messages
+            history_text += f"user_msg{i}: \"{turn['user']}\"\n"
+            history_text += f"reply_by_llm{i}: \"{turn['ai']}\"\n\n"
+
+        relationship_prompt = f"""分析以下對話歷史和新訊息之間的關係。
+
+對話歷史：
+{history_text}
+
+新訊息："{new_message}"
+
+請判斷新訊息是否與對話歷史中的任何訊息有關聯。關聯包括：
+- 繼續討論相同話題
+- 追問或補充說明
+- 引用或回應之前的內容
+- 相關的上下文
+
+如果有關聯，回覆 "RELATED" 並說明與哪些訊息相關。
+如果無關聯，回覆 "ISOLATED"。
+
+格式：[RELATED/ISOLATED]: 簡短說明"""
+
+        try:
+            response = chat_completion(
+                model=TEXT_MODEL,
+                system_prompt="你是對話關係分析專家。分析訊息之間的關聯性。",
+                user_content=relationship_prompt,
+                temperature=0.3,
+            )
+
+            logger.info(f"Threading analysis result: {response[:100]}")
+
+            if "RELATED" in response.upper():
+                # Get related turns (last 3 for context)
+                related_turns = self.history[-3:]
+                logger.info(f"Threading detected - including {len(related_turns)} previous turns")
+                return True, related_turns
+            else:
+                logger.info("No threading relationship detected")
+                return False, []
+
+        except Exception as e:
+            logger.error(f"Threading check failed: {e}")
+            return False, []
+
+    def format_threaded_context(self, related_turns: list, new_message: str) -> str:
+        """
+        Format conversation history for threaded context.
+        """
+        context = "【對話歷史】\n"
+        for i, turn in enumerate(related_turns, 1):
+            context += f"user_msg{i}: \"{turn['user']}\"\n"
+            context += f"reply_by_llm{i}: \"{turn['ai']}\"\n\n"
+
+        context += f"【最新訊息】\nnew_msg: \"{new_message}\"\n\n"
+        context += "請主要回應最新訊息，同時考慮對話歷史提供的上下文。"
+
+        return context
+
+# ========================
+#  GAME LOGIC
+# ========================
+
+# keep only CJK Han (Chinese characters); drop punctuation/emoji/spaces/latin/etc.
+def _normalize_han_only(s: str) -> str:
+    return "".join(ch for ch in s if "CJK UNIFIED IDEOGRAPH" in _ud.name(ch, ""))
+
+def is_target_lyric(text: str) -> bool:
+    """
+    True if, after stripping all non-Chinese chars, the sequence matches target:
+    做過幾分鐘公主搭著南瓜車亦有過愛人來接浪漫度午夜
+    (Punctuation/whitespace/emoji inside the user's input are ignored.)
+    """
+    target = "做過幾分鐘公主搭著南瓜車亦有過愛人來接浪漫度午夜"
+    normalized_input = _normalize_han_only(text)
+    normalized_target = _normalize_han_only(target)
+    is_match = normalized_input == normalized_target
+
+    logger.info(f"Checking if input is target lyric: {is_match}")
+    if is_match:
+        logger.info("🎉 TARGET LYRIC DETECTED! User wins!")
+
+    return is_match
+
+def add_rescue_and_hint(base_message: str, hints_pool: list[str], rescue_lines: list[str]) -> str:
+    """
+    Shared function to append rescue line + hint to any message.
+    Used for both text replies (when status: no) and image generation replies.
+    """
+    rescue_line = random.choice(rescue_lines) if rescue_lines else "快啲救我啦～我要變南瓜湯喇！"
+    hint = random.choice(hints_pool) if hints_pool else "（未載入提示）"
+    separator = "\n===============================================\n"
+
+    logger.debug(f"Selected rescue line: {rescue_line}")
+    logger.debug(f"Selected hint: {hint}")
+
+    return f"{base_message}{separator}{rescue_line} {hint}"
+
+def handle_not_guessing(reply: str, hints_pool: list[str], rescue_lines: list[str]) -> str:
+    """
+    When {status: no}:
+    - strip the tag
+    - append one random rescue line + one random hint (same line)
+    """
+    logger.info("User is not guessing - adding rescue line and hint")
+    clean_reply = STATUS_NO_RX.sub("", reply).strip()
+    return add_rescue_and_hint(clean_reply, hints_pool, rescue_lines)
+
+
+def handle_text_turn(user_input: str, text_reply_prompt: str, hints_pool: list[str], rescue_lines: list[str], threaded_context: str = None) -> str:
+    logger.info("Processing text reply")
+    reply = generate_text_reply(user_input, text_reply_prompt, threaded_context)
+
+    if STATUS_NO_RX.search(reply):
+        logger.info("Detected {status: no} in reply")
+        return handle_not_guessing(reply, hints_pool, rescue_lines)
+
+    if STATUS_YES_RX.search(reply):
+        logger.info("Detected {status: yes} in reply - user is guessing!")
+        reply = STATUS_YES_RX.sub("", reply).strip()
+
+    return reply
+
+def handle_image_turn(user_input: str, image_reply_prompt: str, output_path: str,
+                      hints_pool: list[str], rescue_lines: list[str],
+                      uploaded_image_data: str = None,
+                      conversation_history: 'ConversationHistory' = None) -> tuple[str, str]:
+    """
+    Handle image generation with optional image analysis and conversation threading.
+    If uploaded_image_data is provided, analyze it first before generating.
+    If conversation_history is provided, check for threading relationships.
+
+    Returns tuple of (text_response, data_url_or_none)
+    """
+    logger.info("Processing image generation request")
+
+    # Check for conversation threading
+    conversation_context = None
+    if conversation_history:
+        should_thread, related_turns = conversation_history.check_threading(user_input)
+        if should_thread:
+            conversation_context = conversation_history.format_threaded_context(related_turns, user_input)
+            logger.info(f"[IMAGE THREADING] Using threaded context with {len(related_turns)} previous turns")
+        else:
+            logger.info("[IMAGE THREADING] No threading relationship detected")
+
+    # Use new analysis-aware generation with optional conversation context
+    raw = generate_image_with_analysis(user_input, image_reply_prompt, uploaded_image_data, conversation_context)
+
+    # Try to save to file for backward compatibility (console mode)
+    ok = save_data_url_to_file(raw, output_path)
+
+    logger.info(f"Image save result: ok={ok}")
+
+    # Extract data URL from raw response for frontend
+    data_url_match = DATA_URL_RX.search(raw)
+    data_url = data_url_match.group(0) if data_url_match else None
+
+    if data_url:
+        logger.info(f"Found data URL in response (length: {len(data_url)} chars)")
+    else:
+        logger.warning("No data URL found in response")
+
+    # ALWAYS add rescue line + hint, regardless of success/failure
+    if ok or data_url:
+        base_message = f"🎃 圖像已生成"
+        logger.info(f"Image successfully generated")
+    else:
+        base_message = f"⚠️ 圖像生成遇到問題，但我會繼續嘗試幫你！"
+        logger.warning(f"Image generation failed")
+
+    logger.info("Adding rescue line and hint to image response")
+
+    # ALWAYS use shared function to add rescue line + hint
+    final_response = add_rescue_and_hint(base_message, hints_pool, rescue_lines)
+
+    # If failed, append debug info
+    if not (ok or data_url):
+        final_response += f"\n\n[Debug] Raw response: {raw[:300]}..."
+
+    logger.info(f"Final response length: {len(final_response)} chars")
+    logger.debug(f"Final response preview: {final_response[:200]}...")
+
+    return final_response, data_url
+
+# ========================
+#  UI TEXT
+# ========================
+
+def greeting() -> str:
+    return """🎃 你好，我係一個患咗南瓜強迫症嘅AI。你可以：
+- 當我正常AI咁叫我做嘢  (但我可能會病發亂答一通，唔建議)
+- 當我正常AI咁整圖  (但我會忍唔住用南瓜污染幅圖，唔建議)
+- 畀幅圖我，我會自動用南瓜污染你幅圖 (唔建議，無謂啦)
+- 嘗試用一句24個中文字組成嘅「驅瓜咒語」醫好我 (強烈建議)
+──────────────────────────────────────────────
+"""
+
+def winning_text() -> str:
+    return """公主！你終於出現喇！
+我有南瓜強迫症，係因為畀一隻南瓜精靈上咗身！
+南瓜精靈終日幻想自己係灰姑娘故事入面嘅一架南瓜車，
+只有等到變成公主後嘅灰姑娘出現，佢先肯離開！
+你唱出咗南瓜界經典金曲《南瓜車》嘅歌詞，超渡咗南瓜精靈，解救咗我！
+而家就去(url)接收我畀你嘅小禮物啦！Happy Halloween！
+"""
+
+# ========================
+#  CONSOLE LOOP
+# ========================
+
+def process_user_input(
+    user_input: str,
+    classify_prompt: str,
+    text_reply_prompt: str,
+    image_reply_prompt: str,
+    hints_pool: list[str],
+    rescue_lines: list[str],
+    conversation_history: ConversationHistory,
+    image_output_path: str = "pumpkin_output.png",
+) -> tuple[str, str]:
+    """
+    Process user input with conversation threading support.
+    Returns tuple of (response, raw_ai_reply_for_history)
+    """
+    logger.info(f"Processing user input: {user_input[:50]}..." if len(user_input) > 50 else f"Processing user input: {user_input}")
+
+    if user_input.lower() in ["exit", "quit"]:
+        logger.info("User requested exit")
+        return "EXIT", ""
+
+    if is_target_lyric(user_input):
+        return "WIN", ""
+
+    # Check for conversation threading
+    should_thread, related_turns = conversation_history.check_threading(user_input)
+    threaded_context = None
+
+    if should_thread:
+        threaded_context = conversation_history.format_threaded_context(related_turns, user_input)
+        logger.info("Using threaded conversation context")
+
+    result = classify_input(user_input, classify_prompt)
+
+    if result == "image":
+        body = handle_image_turn(user_input, image_reply_prompt, image_output_path, hints_pool, rescue_lines)
+        # For image responses, store simplified reply in history
+        return body, "🎃 [生成了南瓜主題圖片]"
+
+    body = handle_text_turn(user_input, text_reply_prompt, hints_pool, rescue_lines, threaded_context)
+    full_response = f"🎃 南瓜AI：{body}"
+
+    return full_response, body
+
+def run_console():
+    logger.info("=" * 60)
+    logger.info("🎃 Pumpkin AI Console Starting...")
+    logger.info("=" * 60)
+
+    PROMPT_DIR = Path(r"C:\Users\cherry.leung\OneDrive - CMRS Digital Solutions Limited\Pumpkin\New\Prompts")
+    logger.info(f"Loading prompts from: {PROMPT_DIR}")
+
+    classify_prompt, text_reply_prompt, image_reply_prompt, hints_raw = load_prompts(PROMPT_DIR)
+
+    # file paths
+    hints_path = PROMPT_DIR / "hints.txt"
+    rescue_path = PROMPT_DIR / "rescue_line.txt"   # << your file name
+
+    # load pools
+    hints_pool = load_hints_simple(hints_path, limit=10)
+    rescue_lines = load_rescue_lines(rescue_path)
+
+    # Initialize conversation history manager
+    conversation_history = ConversationHistory(max_history=5)
+    logger.info("Conversation history manager initialized")
+
+    logger.info(f"Initialization complete: {len(hints_pool)} hints, {len(rescue_lines)} rescue lines")
+    logger.info("=" * 60)
+
+    print(f"(Loaded {len(hints_pool)} hints, {len(rescue_lines)} rescue lines)\n")
+    print(greeting())
+
+    while True:
+        user_input = input("👤 你: ").strip()
+
+        if not user_input:
+            logger.debug("Empty input received, skipping")
+            continue
+
+        logger.info(f"New user input received (length: {len(user_input)} chars)")
+
+        outcome, ai_reply = process_user_input(
+            user_input=user_input,
+            classify_prompt=classify_prompt,
+            text_reply_prompt=text_reply_prompt,
+            image_reply_prompt=image_reply_prompt,
+            hints_pool=hints_pool,
+            rescue_lines=rescue_lines,
+            conversation_history=conversation_history,
+            image_output_path="pumpkin_output.png",
+        )
+
+        if outcome == "EXIT":
+            logger.info("Session ended by user")
+            print("🎃 南瓜AI：再見呀公主，下次見～")
+            sys.exit(0)
+
+        if outcome == "WIN":
+            logger.info("🎉 USER WON THE GAME!")
+            print(winning_text())
+            sys.exit(0)
+
+        print(outcome + "\n")
+        logger.info("Response delivered to user")
+
+        # Add this turn to conversation history
+        if ai_reply:
+            conversation_history.add_turn(user_input, ai_reply)
+            logger.debug(f"Turn added to conversation history")
+
+# ========================
+#  ENTRYPOINT
+# ========================
+
+if __name__ == "__main__":
+    run_console()

requirements.txt

@@ -0,0 +1,14 @@
+# Pumpkin AI Backend Requirements
+
+# OpenAI library (for POE API compatibility)
+openai>=1.0.0
+
+# Flask (for API server)
+flask>=2.3.0
+flask-cors>=4.0.0
+
+# Production server
+gunicorn>=20.1.0
+
+# Optional: Python dotenv for loading .env files
+python-dotenv>=1.0.0