diff --git a/core/__pycache__/__init__.cpython-312.pyc b/core/__pycache__/__init__.cpython-312.pyc
index c040350..3ce90a8 100644
Binary files a/core/__pycache__/__init__.cpython-312.pyc and b/core/__pycache__/__init__.cpython-312.pyc differ
diff --git a/core/__pycache__/ai_agent.cpython-312.pyc b/core/__pycache__/ai_agent.cpython-312.pyc
index dae3bf4..ddbb6ba 100644
Binary files a/core/__pycache__/ai_agent.cpython-312.pyc and b/core/__pycache__/ai_agent.cpython-312.pyc differ
diff --git a/core/__pycache__/poster_gen.cpython-312.pyc b/core/__pycache__/poster_gen.cpython-312.pyc
index 18f2c48..6ae451a 100644
Binary files a/core/__pycache__/poster_gen.cpython-312.pyc and b/core/__pycache__/poster_gen.cpython-312.pyc differ
diff --git a/core/__pycache__/simple_collage.cpython-312.pyc b/core/__pycache__/simple_collage.cpython-312.pyc
index ad0f3c3..82c1e67 100644
Binary files a/core/__pycache__/simple_collage.cpython-312.pyc and b/core/__pycache__/simple_collage.cpython-312.pyc differ
diff --git a/core/__pycache__/topic_parser.cpython-312.pyc b/core/__pycache__/topic_parser.cpython-312.pyc
index 89dd363..e943c02 100644
Binary files a/core/__pycache__/topic_parser.cpython-312.pyc and b/core/__pycache__/topic_parser.cpython-312.pyc differ
diff --git a/output/2025-04-24_19-44-15.json b/output/2025-04-24_19-44-15.json
deleted file mode 100644
index 4368846..0000000
--- a/output/2025-04-24_19-44-15.json
+++ /dev/null
@@ -1,26 +0,0 @@
-[
-    {
-        "index": 1,
-        "main_title": "泰宁甘露寺攻略",
-        "texts": [
-            "悬空古建探秘",
-            "亲子研学首选"
-        ]
-    },
-    {
-        "index": 2,
-        "main_title": "推荐！泰宁甘露寺",
-        "texts": [
-            "千年古刹秘境",
-            "自驾打卡胜地"
-        ]
-    },
-    {
-        "index": 3,
-        "main_title": "泰宁甘露寺",
-        "texts": [
-            "非遗文化沉浸",
-            "状元祈福之旅"
-        ]
-    }
-]
\ No newline at end of file
diff --git a/scripts/regenerate_img.py b/scripts/regenerate_img.py
new file mode 100644
index 0000000..179196b
--- /dev/null
+++ b/scripts/regenerate_img.py
@@ -0,0 +1,338 @@
+import os
+import json
+import logging
+import argparse
+from pathlib import Path
+import traceback
+from PIL import Image
+import sys
+
+# 假设你的项目结构允许这样导入
+# 如果不行，你可能需要调整 sys.path
+# 获取脚本所在的目录的上上级目录（通常是项目根目录）
+project_root = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(project_root))
+print(f"Added {project_root} to sys.path")
+try:
+    from core.poster_gen import PosterGenerator
+except ImportError as e:
+    print(f"Error importing PosterGenerator: {e}")
+    print("Initial sys.path import failed. Please ensure the script is run from the project root or adjust paths.")
+    # 尝试另一种可能的方式，如果脚本在 scripts/ 目录下
+    # script_dir = Path(__file__).resolve().parent
+    # project_root_alt = script_dir.parent
+    # if str(project_root_alt) not in sys.path:
+    #     sys.path.insert(0, str(project_root_alt))
+    #     print(f"Added alternative root {project_root_alt} to sys.path")
+    #     try:
+    #         from core.poster_gen import PosterGenerator
+    #     except ImportError as e2:
+    #         print(f"Second attempt failed: {e2}")
+    #         print("Ensure 'core' module is accessible from the script's location.")
+    #         sys.exit(1)
+    sys.exit(1)
+
+
+# --- 日志配置 ---
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(levelname)s - [%(filename)s:%(lineno)d] - %(message)s',
+    datefmt='%Y-%m-%d %H:%M:%S'
+)
+logger = logging.getLogger(__name__)
+
+def load_main_config(config_path="poster_gen_config.json"):
+    """加载主配置文件"""
+    config_file = Path(config_path)
+    if not config_file.is_file():
+        logger.error(f"主配置文件未找到: {config_path}")
+        return None
+    try:
+        with open(config_file, 'r', encoding='utf-8') as f:
+            config = json.load(f)
+        logger.info(f"主配置文件加载成功: {config_path}")
+        # 验证必要的键
+        required_keys = ["poster_assets_base_dir", "output_dir"]
+        if not all(key in config for key in required_keys):
+            missing = [key for key in required_keys if key not in config]
+            logger.error(f"主配置文件缺少必要的键: {missing}")
+            return None
+        return config
+    except json.JSONDecodeError:
+        logger.error(f"无法解析主配置文件 (JSON格式错误): {config_path}")
+        return None
+    except Exception as e:
+        logger.exception(f"加载主配置文件时发生错误 {config_path}: {e}")
+        return None
+
+def find_topic_variants(run_dir: Path):
+    """查找指定运行目录下所有的 topic_variant 子目录"""
+    variants = []
+    if not run_dir.is_dir():
+        logger.error(f"运行目录不存在或不是目录: {run_dir}")
+        return variants
+
+    for item in run_dir.iterdir():
+        # 匹配形如 "1_1", "2_3" 等目录
+        if item.is_dir() and '_' in item.name:
+            parts = item.name.split('_')
+            if len(parts) == 2 and parts[0].isdigit() and parts[1].isdigit():
+                topic_index = int(parts[0])
+                variant_index = int(parts[1])
+                variants.append({
+                    "topic_index": topic_index,
+                    "variant_index": variant_index,
+                    "path": item
+                })
+    logger.info(f"在 {run_dir} 中找到 {len(variants)} 个主题变体目录")
+    return sorted(variants, key=lambda x: (x["topic_index"], x["variant_index"]))
+
+def get_variant_poster_config(run_dir: Path, topic_index: int, variant_index: int):
+    """从 topic_X_poster_configs.json 加载特定变体的海报配置"""
+    config_file = run_dir / f"topic_{topic_index}_poster_configs.json"
+    if not config_file.is_file():
+        logger.warning(f"主题 {topic_index} 的海报配置文件未找到: {config_file}")
+        return None
+
+    try:
+        with open(config_file, 'r', encoding='utf-8') as f:
+            all_configs = json.load(f)
+
+        if not isinstance(all_configs, list):
+            logger.warning(f"配置文件 {config_file} 格式不是列表")
+            return None
+
+        # 查找匹配 variant_index 的配置项 (假设配置中的 index 从 1 开始)
+        for config_item in all_configs:
+            if isinstance(config_item, dict) and config_item.get("index") == variant_index:
+                # 验证必要字段
+                if "main_title" in config_item and "texts" in config_item:
+                     # 确保 texts 是列表，至少包含两个元素
+                    texts = config_item.get("texts", ["", ""])
+                    if not isinstance(texts, list):
+                        texts = [str(texts), ""] # 强制转为列表
+                    while len(texts) < 2:
+                        texts.append("")
+                    config_item["texts"] = texts[:2] # 最多取两个
+                    logger.debug(f"找到变体 {topic_index}_{variant_index} 的配置: {config_item}")
+                    return config_item
+                else:
+                    logger.warning(f"变体 {topic_index}_{variant_index} 的配置缺少 main_title 或 texts 字段")
+                    return None # 返回空配置或其他默认值？这里返回None
+
+        logger.warning(f"在 {config_file} 中未找到索引为 {variant_index} 的配置")
+        return None
+    except json.JSONDecodeError:
+        logger.error(f"解析配置文件 JSON 失败: {config_file}")
+        return None
+    except Exception as e:
+        logger.exception(f"加载或查找变体配置时出错 {config_file}: {e}")
+        return None
+
+def regenerate_single_poster(generator: PosterGenerator, variant_info: dict, run_dir: Path):
+    """重新生成单个海报"""
+    topic_index = variant_info["topic_index"]
+    variant_index = variant_info["variant_index"]
+    variant_path = variant_info["path"]
+    logger.info(f"--- 开始处理: 主题 {topic_index}, 变体 {variant_index} (目录: {variant_path.name}) ---")
+
+    # 1. 定位拼贴图
+    collage_dir = variant_path / "collage_img"
+    collage_path = None
+    if collage_dir.is_dir():
+         image_files = list(collage_dir.glob("*.png")) + list(collage_dir.glob("*.jpg")) + list(collage_dir.glob("*.jpeg"))
+         if image_files:
+             collage_path = image_files[0] # 取第一个找到的
+             logger.info(f"找到拼贴图: {collage_path}")
+         else:
+             logger.warning(f"在 {collage_dir} 中未找到拼贴图 (png/jpg/jpeg)")
+             return False
+    else:
+        logger.warning(f"拼贴图目录不存在: {collage_dir}")
+        return False
+
+    # 2. 获取该变体的海报文字配置
+    poster_config = get_variant_poster_config(run_dir, topic_index, variant_index)
+    if poster_config is None:
+        logger.warning(f"无法获取变体 {topic_index}_{variant_index} 的海报配置，跳过生成。")
+        return False
+
+    # 3. 准备 text_data
+    text_data = {
+        "title": poster_config.get("main_title", ""), # 保证默认是空字符串
+        "subtitle": "", # 通常副标题为空
+        "additional_texts": []
+    }
+    texts = poster_config.get("texts", [])
+    if texts:
+        for text in texts:
+            if text and isinstance(text, str): # 确保文本不为空且是字符串
+                text_data["additional_texts"].append({"text": text, "position": "middle", "size_factor": 0.8})
+            elif isinstance(text, dict) and text.get("text"): # 如果已经是字典格式
+                 text_data["additional_texts"].append(text)
+
+    logger.debug(f"准备好的 text_data: {text_data}")
+
+    # 4. 加载拼贴图
+    try:
+        collage_img = Image.open(collage_path).convert('RGBA')
+    except Exception as e:
+        logger.exception(f"无法加载拼贴图 {collage_path}: {e}")
+        return False
+
+    # 5. 调用生成器创建海报
+    logger.info("调用 PosterGenerator.create_poster...")
+    try:
+        regenerated_poster_img = generator.create_poster(collage_img, text_data)
+    except Exception as e:
+        logger.exception(f"调用 create_poster 时发生严重错误: {e}")
+        traceback.print_exc() # 打印详细错误堆栈
+        regenerated_poster_img = None # 确保设置为 None
+
+
+    # 6. 保存重新生成的海报
+    if regenerated_poster_img:
+        poster_output_dir = variant_path / "poster"
+        poster_output_dir.mkdir(parents=True, exist_ok=True)
+        # 假设输出文件名固定为 poster.jpg
+        poster_output_path = poster_output_dir / "poster.jpg"
+        try:
+            # 确保保存为 JPG 格式时是 RGB
+            if poster_output_path.suffix.lower() == ".jpg" or poster_output_path.suffix.lower() == ".jpeg":
+                 if regenerated_poster_img.mode == 'RGBA':
+                     logger.debug("将图像从 RGBA 转换为 RGB 以保存为 JPG")
+                     regenerated_poster_img = regenerated_poster_img.convert('RGB')
+            regenerated_poster_img.save(poster_output_path)
+            logger.info(f"成功重新生成并保存海报到: {poster_output_path}")
+            return True
+        except Exception as e:
+            logger.exception(f"保存重新生成的海报失败 {poster_output_path}: {e}")
+            return False
+    else:
+        logger.error(f"PosterGenerator未能生成海报图像 (返回 None)")
+        return False
+
+# --- 主逻辑 ---
+def main(run_dirs_to_process, config_path, debug_mode):
+    """主处理函数，接收运行目录列表"""
+
+    if debug_mode:
+        logger.setLevel(logging.DEBUG)
+        logging.getLogger().setLevel(logging.DEBUG) # 应用到根logger
+        logger.info("DEBUG 日志已启用")
+
+    # 加载主配置
+    main_config = load_main_config(config_path)
+    if main_config is None:
+        logger.critical("无法加载主配置，脚本终止。")
+        return # 或者 sys.exit(1)
+
+    # 初始化 PosterGenerator
+    poster_assets_dir = main_config.get("poster_assets_base_dir")
+    if not poster_assets_dir:
+         logger.critical("配置文件中未找到 'poster_assets_base_dir'。")
+         return
+
+    poster_generator = None
+    try:
+        # PosterGenerator 可能需要 output_dir 参数，尽管这里可能不会直接用它保存
+        output_dir_from_config = main_config.get("output_dir")
+        poster_generator = PosterGenerator(base_dir=poster_assets_dir, output_dir=output_dir_from_config)
+        # 根据需要设置参数，例如从主配置文件读取
+        poster_generator.set_img_frame_possibility(main_config.get("img_frame_possibility", 0.7))
+        poster_generator.set_text_bg_possibility(main_config.get("text_bg_possibility", 0))
+        logger.info(f"PosterGenerator 初始化成功，资源目录: {poster_assets_dir}")
+    except Exception as e:
+        logger.critical(f"初始化 PosterGenerator 失败: {e}")
+        traceback.print_exc()
+        return
+
+    total_success_count = 0
+    total_failure_count = 0
+
+    # 遍历要处理的运行目录列表
+    for run_dir_str in run_dirs_to_process:
+        run_directory = Path(run_dir_str)
+        logger.info(f"\n===== 开始处理运行目录: {run_directory} =====")
+
+        if not run_directory.is_dir():
+            logger.error(f"指定的运行目录不存在或不是目录: {run_directory}")
+            continue # 跳过这个目录，继续下一个
+
+        # 查找当前运行目录下的所有主题变体
+        topic_variants = find_topic_variants(run_directory)
+        if not topic_variants:
+            logger.warning(f"在 {run_directory} 中未找到任何主题变体目录，跳过。")
+            continue
+
+        # 循环处理每个变体
+        run_success_count = 0
+        run_failure_count = 0
+        for variant in topic_variants:
+            try:
+                if regenerate_single_poster(poster_generator, variant, run_directory):
+                    run_success_count += 1
+                else:
+                    run_failure_count += 1
+            except Exception as e:
+                 logger.exception(f"处理变体 {variant['path']} 时发生未捕获的错误:")
+                 run_failure_count += 1
+            # logger.info(f"--- 完成处理变体: {variant['path'].name} ---") # 可以在 regenerate_single_poster 内部打印完成信息
+
+        logger.info(f"===== 完成处理运行目录: {run_directory} =====")
+        logger.info(f"本轮成功: {run_success_count}, 失败/跳过: {run_failure_count}")
+        total_success_count += run_success_count
+        total_failure_count += run_failure_count
+
+    logger.info("=" * 30)
+    logger.info(f"所有指定目录的海报重新生成完成。")
+    logger.info(f"总成功: {total_success_count}")
+    logger.info(f"总失败/跳过: {total_failure_count}")
+    logger.info("=" * 30)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="重新生成指定运行ID目录下所有主题变体的海报")
+    # 不再需要 run_dir 参数
+    # parser.add_argument(
+    #     "run_dir",
+    #     type=str,
+    #     help="包含主题变体子目录的运行结果目录路径"
+    # )
+    parser.add_argument(
+        "--config",
+        type=str,
+        default="poster_gen_config.json",
+        help="主配置文件路径 (poster_gen_config.json)"
+    )
+    parser.add_argument(
+        "--debug",
+        action='store_true',
+        help="启用 DEBUG 级别日志"
+    )
+
+    args = parser.parse_args()
+
+    # ==================================================
+    # 在这里定义你要处理的运行目录列表
+    # ==================================================
+    run_directories = [
+        "/root/autodl-tmp/TravelContentCreator/result/安吉/2025-04-27_12-55-40",
+        "/root/autodl-tmp/TravelContentCreator/result/乌镇/2025-04-27_10-50-34",
+        "/root/autodl-tmp/TravelContentCreator/result/齐云山/2025-04-27_11-51-56",
+        "/root/autodl-tmp/TravelContentCreator/result/长鹿/2025-04-27_14-03-44",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_02-02-34",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_02-23-17",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_07-57-20",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_09-29-20"
+        # "/root/autodl-tmp/TravelContentCreator/result/ANOTHER_RUN_ID",
+        # 添加更多你需要处理的运行目录路径...
+    ]
+    # ==================================================
+
+    if not run_directories:
+        print("错误：请在脚本中编辑 `run_directories` 列表，指定至少一个要处理的运行目录。")
+        sys.exit(1)
+
+    # 调用主处理函数
+    main(run_directories, args.config, args.debug)
\ No newline at end of file
diff --git a/scripts/regenerate_poster_content.py b/scripts/regenerate_poster_content.py
new file mode 100644
index 0000000..7db87b9
--- /dev/null
+++ b/scripts/regenerate_poster_content.py
@@ -0,0 +1,457 @@
+import os
+import json
+import logging
+import argparse
+from pathlib import Path
+import traceback
+import sys
+import re # Needed for cleaning object names and finding description files
+
+# --- Path Setup ---
+# Get the directory two levels up from the script (assumes scripts/regenerate...py structure)
+project_root = Path(__file__).resolve().parent.parent
+if str(project_root) not in sys.path:
+    sys.path.insert(0, str(project_root))
+    print(f"Added project root {project_root} to sys.path")
+
+try:
+    # Import necessary modules from your project structure
+    from utils.content_generator import ContentGenerator
+    # ContentGenerator likely handles AI_Agent internally for its 'run' method
+    # from core.ai_agent import AI_Agent
+except ImportError as e:
+    print(f"Error importing project modules: {e}")
+    print(f"Sys.path: {sys.path}")
+    print("Ensure the script is run from a location where 'utils' and 'core' are importable, or adjust sys.path.")
+    sys.exit(1)
+
+# --- Logging Configuration ---
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(levelname)s - [%(filename)s:%(lineno)d] - %(message)s',
+    datefmt='%Y-%m-%d %H:%M:%S'
+)
+logger = logging.getLogger(__name__)
+
+# --- Helper Functions ---
+
+def load_main_config(config_path="poster_gen_config.json"):
+    """Loads the main configuration file."""
+    config_file = Path(config_path)
+    if not config_file.is_file():
+        logger.error(f"Main configuration file not found: {config_path}")
+        return None
+    try:
+        with open(config_file, 'r', encoding='utf-8') as f:
+            config = json.load(f)
+        logger.info(f"Main configuration loaded successfully: {config_path}")
+        # Validate necessary keys for this script
+        required_keys = [
+            "api_url", "model", "api_key", "output_dir",
+            "poster_content_system_prompt", "variants", "resource_dir"
+            # Add other keys needed by ContentGenerator.run or description finding
+        ]
+        if not all(key in config for key in required_keys):
+            missing = [key for key in required_keys if key not in config]
+            logger.error(f"Main configuration file is missing required keys: {missing}")
+            return None
+        return config
+    except json.JSONDecodeError:
+        logger.error(f"Failed to parse main configuration file (JSON format error): {config_path}")
+        return None
+    except Exception as e:
+        logger.exception(f"Error loading main configuration from {config_path}: {e}")
+        return None
+
+def load_topic_data(run_dir: Path, run_id: str):
+    """Loads the topic list from the tweet_topic_{run_id}.json file."""
+    topic_file_exact = run_dir / f"tweet_topic_{run_id}.json"
+    topic_file_generic = run_dir / "tweet_topic.json"
+
+    topic_file_to_load = None
+    if topic_file_exact.is_file():
+        topic_file_to_load = topic_file_exact
+    elif topic_file_generic.is_file():
+        topic_file_to_load = topic_file_generic
+        logger.warning(f"Specific topic file {topic_file_exact.name} not found, using generic {topic_file_generic.name}")
+    else:
+        logger.error(f"Topic file not found in {run_dir} (tried {topic_file_exact.name} and {topic_file_generic.name})")
+        return None
+
+    try:
+        with open(topic_file_to_load, 'r', encoding='utf-8') as f:
+            data = json.load(f)
+            if isinstance(data, dict) and "topics_list" in data:
+                topics_list = data["topics_list"]
+            elif isinstance(data, list):
+                 topics_list = data
+            else:
+                logger.error(f"Unexpected format in topic file: {topic_file_to_load}")
+                return None
+        # Add run_dir to each topic item for easier path access later
+        for topic in topics_list:
+            if isinstance(topic, dict):
+                topic['run_dir'] = str(run_dir) # Store the run directory path
+        logger.info(f"Successfully loaded {len(topics_list)} topics from {topic_file_to_load}")
+        return topics_list
+    except json.JSONDecodeError:
+        logger.error(f"Failed to parse topic file (JSON format error): {topic_file_to_load}")
+        return None
+    except Exception as e:
+        logger.exception(f"Error loading topics from {topic_file_to_load}: {e}")
+        return None
+
+def get_topic_index(topic_item: dict, list_index: int) -> int:
+    """Safely gets the topic index, falling back to list index + 1."""
+    topic_index = topic_item.get("index")
+    try:
+        if isinstance(topic_index, str) and topic_index.isdigit():
+            topic_index = int(topic_index)
+        elif not isinstance(topic_index, int) or topic_index < 1:
+            logger.warning(f"Topic item {list_index} has invalid or missing 'index'. Using list index {list_index + 1}.")
+            topic_index = list_index + 1
+    except Exception:
+        logger.warning(f"Error processing index for topic item {list_index}. Using list index {list_index + 1}.")
+        topic_index = list_index + 1
+    return topic_index
+
+
+def find_and_load_variant_contents(run_dir: Path, topic_index: int) -> list:
+    """Finds all variant directories for a topic and loads their article.json content."""
+    variant_contents = []
+    variant_dirs = sorted(run_dir.glob(f"{topic_index}_*"), key=lambda p: int(p.name.split('_')[-1])) # Sort by variant number
+
+    if not variant_dirs:
+        logger.warning(f"No variant directories found for topic {topic_index} in {run_dir}")
+        return variant_contents
+
+    logger.debug(f"Found {len(variant_dirs)} potential variant directories for topic {topic_index}: {[d.name for d in variant_dirs]}")
+
+    for variant_dir in variant_dirs:
+         # Double check it's a directory with the correct format
+         if variant_dir.is_dir() and '_' in variant_dir.name:
+             parts = variant_dir.name.split('_')
+             if len(parts) == 2 and parts[0] == str(topic_index) and parts[1].isdigit():
+                 content = load_variant_content(variant_dir)
+                 if content: # Only add if content was loaded successfully
+                     variant_contents.append(content)
+                 else:
+                      logger.warning(f"Could not load valid content for variant {variant_dir.name}, it will be excluded.")
+
+    logger.info(f"Loaded content for {len(variant_contents)} variants for topic {topic_index}.")
+    return variant_contents
+
+
+def load_variant_content(variant_dir: Path):
+    """Loads the article content (title, content) from article.json."""
+    content_file = variant_dir / "article.json"
+    if not content_file.is_file():
+        logger.warning(f"Article content file not found: {content_file}")
+        return None
+    try:
+        with open(content_file, 'r', encoding='utf-8') as f:
+            content_data = json.load(f)
+        if isinstance(content_data, dict) and "title" in content_data and "content" in content_data:
+            logger.debug(f"Successfully loaded content from {content_file}")
+            return content_data
+        else:
+            logger.warning(f"Invalid format in article content file: {content_file}. Missing 'title' or 'content'.")
+            return None
+    except json.JSONDecodeError:
+        logger.error(f"Failed to parse article content file (JSON format error): {content_file}")
+        return None
+    except Exception as e:
+        logger.exception(f"Error loading content from {content_file}: {e}")
+        return None
+
+def clean_object_name(object_name_raw: str) -> str:
+    """Cleans the object name, similar to logic in tweet_generator."""
+    if not object_name_raw:
+        return ""
+    try:
+        cleaned = object_name_raw.split(".")[0].replace("景点信息-", "").strip()
+        return cleaned
+    except Exception as e:
+        logger.warning(f"Could not fully clean object name '{object_name_raw}': {e}. Using raw.")
+        return object_name_raw.strip()
+
+def find_description_file(object_name: str, resource_dir_config: list, desc_dir: Path | None = None) -> list:
+    """
+    Tries to find the description file path based on object name.
+    Prioritizes searching in desc_dir if provided.
+    """
+    info_directory = [] # Expects a list of paths
+    found_description = False
+
+    # --- Step 1: Prioritize search in the specified desc_dir ---
+    if desc_dir and desc_dir.is_dir():
+        logger.info(f"Prioritizing description file search in specified directory: {desc_dir}")
+        # Simple search: find first file containing the object name (case-insensitive)
+        for file_path in desc_dir.iterdir():
+             # Use stem for matching (filename without extension)
+             # Make comparison case-insensitive for robustness
+            if file_path.is_file() and object_name.lower() in file_path.stem.lower():
+                info_directory = [str(file_path)]
+                logger.info(f"Found description file in specified directory: {file_path}")
+                found_description = True
+                break # Use the first match found in the specified dir
+        if not found_description:
+            logger.warning(f"No file containing '{object_name}' found in specified directory: {desc_dir}. Falling back to resource_dir config.")
+    elif desc_dir:
+        logger.warning(f"Specified description directory '{desc_dir}' not found or not a directory. Falling back to resource_dir config.")
+
+    # --- Step 2: Fallback to searching resource_dir_config from main config ---
+    if not found_description:
+        if not object_name or not resource_dir_config:
+            logger.warning("Cannot find description file: Missing object name or resource_dir config.")
+            return info_directory # Return empty list
+
+        logger.debug(f"Searching for description file for object '{object_name}' in resource_dir config...")
+        # keywords = [k for k in re.split(r'[+\s_\-]+', object_name) if len(k) > 1] # Keywords might be needed for fuzzy matching later
+
+        # Search logic based on resource_dir_config (as before)
+        for dir_info in resource_dir_config:
+            if dir_info.get("type") == "Description":
+                for file_path_str in dir_info.get("file_path", []):
+                    file_path = Path(file_path_str)
+                    # Use stem for matching, case-insensitive
+                    if object_name.lower() in file_path.stem.lower():
+                        if file_path.is_file():
+                            info_directory = [str(file_path)]
+                            logger.info(f"Found potential description file in resource_dir config: {file_path}")
+                            found_description = True
+                            break # Take the first match based on this logic
+                        else:
+                            logger.warning(f"Configured description file not found at path: {file_path}")
+                if found_description:
+                    break
+
+        if not found_description:
+             logger.warning(f"No description file found for '{object_name}' in resource_dir config either.")
+
+    return info_directory
+
+
+def regenerate_topic_poster_config(
+    content_gen: ContentGenerator,
+    topic_item: dict,
+    topic_index: int, # Now using the determined index
+    run_dir: Path,
+    main_config: dict,
+    desc_dir: Path | None # Added desc_dir parameter
+) -> bool:
+    """Regenerates the poster text config for a single topic using all its variant contents."""
+    logger.info(f"--- Regenerating Poster Config for Topic {topic_index} (Object: {topic_item.get('object', 'N/A')}) ---")
+
+    # 1. Load content from ALL variants for this topic
+    variant_contents_list = find_and_load_variant_contents(run_dir, topic_index)
+    if not variant_contents_list:
+        logger.error(f"No valid variant contents found for topic {topic_index}. Skipping.")
+        return False
+
+    # 2. Find description file (info_directory) - Pass desc_dir here
+    object_name_raw = topic_item.get("object", "")
+    object_name_cleaned = clean_object_name(object_name_raw)
+    resource_config = main_config.get("resource_dir", [])
+    info_directory = find_description_file(object_name_cleaned, resource_config, desc_dir) # Pass desc_dir
+
+    # 3. Get parameters for ContentGenerator.run
+    poster_num = len(variant_contents_list)
+    system_prompt_path = main_config.get("poster_content_system_prompt")
+    api_url = main_config.get("api_url")
+    model_name = main_config.get("model")
+    api_key = main_config.get("api_key")
+    timeout = main_config.get("request_timeout", 120)
+
+    if not system_prompt_path or not Path(system_prompt_path).is_file():
+        logger.error(f"Poster content system prompt file not found or not specified in config: {system_prompt_path}")
+        return False
+
+    try:
+        with open(system_prompt_path, "r", encoding="utf-8") as f:
+            system_prompt = f.read()
+    except Exception as e:
+        logger.exception(f"Failed to read system prompt file {system_prompt_path}: {e}")
+        return False
+
+    # Set ContentGenerator parameters
+    content_gen.set_temperature(main_config.get("content_temperature", 0.7))
+    content_gen.set_top_p(main_config.get("content_top_p", 0.8))
+    content_gen.set_presence_penalty(main_config.get("content_presence_penalty", 1.2))
+
+    # 4. Call ContentGenerator.run with the list of variant contents
+    logger.info(f"Calling ContentGenerator.run for topic {topic_index} with {poster_num} variant contents...")
+    logger.debug(f"  - Info Dir Used: {info_directory}") # Log the final info_directory used
+    logger.debug(f"  - Poster Num: {poster_num}")
+
+    try:
+        regenerated_configs = content_gen.run(
+            info_directory=info_directory, # Use the potentially updated info_directory
+            poster_num=poster_num,
+            content_data=variant_contents_list,
+            system_prompt=system_prompt,
+            api_url=api_url,
+            model_name=model_name,
+            api_key=api_key,
+            timeout=timeout
+        )
+    except Exception as e:
+        logger.exception(f"ContentGenerator.run failed for topic {topic_index}: {e}")
+        traceback.print_exc()
+        regenerated_configs = None
+
+    # 5. Save the regenerated config (list)
+    if regenerated_configs is not None:
+        output_config_path = run_dir / f"topic_{topic_index}_poster_configs.json"
+        try:
+            if not isinstance(regenerated_configs, list):
+                 logger.warning(f"ContentGenerator.run for topic {topic_index} did not return a list. Attempting to wrap. Result type: {type(regenerated_configs)}")
+                 regenerated_configs = [regenerated_configs] if regenerated_configs else []
+
+            with open(output_config_path, 'w', encoding='utf-8') as f_out:
+                json.dump(regenerated_configs, f_out, ensure_ascii=False, indent=4)
+            logger.info(f"Successfully regenerated and saved poster configs to: {output_config_path} ({len(regenerated_configs)} configs)")
+            return True
+        except Exception as e:
+            logger.exception(f"Failed to save regenerated poster configs to {output_config_path}: {e}")
+            return False
+    else:
+        logger.error(f"ContentGenerator.run did not return valid config data for topic {topic_index}.")
+        return False
+
+
+# --- Main Logic ---
+def main(run_dirs_to_process, config_path, debug_mode, desc_dir_path: Path | None):
+    """Main processing function."""
+
+    if debug_mode:
+        logger.setLevel(logging.DEBUG)
+        logging.getLogger().setLevel(logging.DEBUG)
+        logger.info("DEBUG 日志已启用")
+
+    main_config = load_main_config(config_path)
+    if main_config is None:
+        logger.critical("Failed to load main configuration. Aborting.")
+        return
+
+    try:
+        content_generator = ContentGenerator(output_dir=main_config.get("output_dir"))
+        logger.info("ContentGenerator initialized.")
+    except Exception as e:
+        logger.critical(f"Failed to initialize ContentGenerator: {e}")
+        traceback.print_exc()
+        return
+
+    total_success_count = 0
+    total_failure_count = 0
+    all_topics_data = []
+
+    # Phase 1: Load all topics
+    for run_dir_str in run_dirs_to_process:
+        run_directory = Path(run_dir_str)
+        run_id = run_directory.name
+        logger.info(f"\n===== Loading Topics from: {run_directory} (Run ID: {run_id}) =====")
+        if not run_directory.is_dir():
+            logger.error(f"Directory not found: {run_directory}. Skipping.")
+            continue
+        topics_list = load_topic_data(run_directory, run_id)
+        if topics_list:
+            all_topics_data.extend(topics_list)
+        else:
+            logger.error(f"Failed to load topics for {run_id}.")
+
+    if not all_topics_data:
+        logger.critical("No topics loaded from any specified run directory. Aborting.")
+        return
+
+    logger.info(f"\n===== Loaded a total of {len(all_topics_data)} topics. Starting regeneration. =====")
+
+    # Phase 2: Process all loaded topics
+    for i, topic_item in enumerate(all_topics_data):
+        run_dir_str = topic_item.get('run_dir')
+        if not run_dir_str:
+            logger.error(f"Topic item {i} is missing 'run_dir' information. Cannot process.")
+            total_failure_count += 1
+            continue
+        run_directory = Path(run_dir_str)
+        topic_index = get_topic_index(topic_item, i)
+
+        try:
+            # Pass desc_dir_path to the processing function
+            if regenerate_topic_poster_config(
+                content_generator,
+                topic_item,
+                topic_index,
+                run_directory,
+                main_config,
+                desc_dir_path # Pass the specified description directory path
+            ):
+                total_success_count += 1
+            else:
+                total_failure_count += 1
+        except Exception as e:
+             logger.exception(f"Unhandled error processing topic index {topic_index} from run {run_directory.name}:")
+             total_failure_count += 1
+
+    logger.info("=" * 30)
+    logger.info(f"All topics processed.")
+    logger.info(f"Total Configs Regenerated Successfully: {total_success_count}")
+    logger.info(f"Total Failed/Skipped Topics: {total_failure_count}")
+    logger.info("=" * 30)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="重新生成指定运行ID目录下所有主题的海报文本配置")
+    parser.add_argument(
+        "--config",
+        type=str,
+        default="poster_gen_config.json",
+        help="主配置文件路径 (poster_gen_config.json)"
+    )
+    parser.add_argument(
+        "--desc_dir",
+        type=str,
+        default=None,
+        help="可选：指定一个目录优先查找描述文件 (如果提供)"
+    )
+    parser.add_argument(
+        "--debug",
+        action='store_true',
+        help="启用 DEBUG 级别日志"
+    )
+
+    args = parser.parse_args()
+
+    # --- Process desc_dir argument ---
+    description_directory = Path("/root/autodl-tmp/TravelContentCreator/resource/Object")
+    if args.desc_dir:
+        description_directory = Path(args.desc_dir)
+        if not description_directory.is_dir():
+             logger.warning(f"指定的描述目录 (--desc_dir) 不是一个有效的目录: {args.desc_dir}. 将忽略此参数.")
+             description_directory = None # Reset if invalid
+        else:
+             logger.info(f"将优先在以下目录查找描述文件: {description_directory}")
+    # --- End Process desc_dir ---
+
+
+    # ==================================================
+    # Define the list of run directories to process here
+    # ==================================================
+    run_directories = [
+        "/root/autodl-tmp/TravelContentCreator/result/安吉/2025-04-27_12-55-40",
+        "/root/autodl-tmp/TravelContentCreator/result/乌镇/2025-04-27_10-50-34",
+        "/root/autodl-tmp/TravelContentCreator/result/齐云山/2025-04-27_11-51-56",
+        "/root/autodl-tmp/TravelContentCreator/result/长鹿/2025-04-27_14-03-44",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_02-02-34",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_02-23-17",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_07-57-20",
+        "/root/autodl-tmp/TravelContentCreator/result/笔架山/2025-04-27_09-29-20"
+    ]
+    # ==================================================
+
+    if not run_directories:
+        print("ERROR: Please edit the 'run_directories' list in the script to specify at least one run directory to process.")
+        sys.exit(1)
+
+    # Call the main processing function, passing the processed description_directory path
+    main(run_directories, args.config, args.debug, description_directory)
\ No newline at end of file
diff --git a/utils/__pycache__/__init__.cpython-312.pyc b/utils/__pycache__/__init__.cpython-312.pyc
index 053c766..22cbbcf 100644
Binary files a/utils/__pycache__/__init__.cpython-312.pyc and b/utils/__pycache__/__init__.cpython-312.pyc differ
diff --git a/utils/__pycache__/content_generator.cpython-312.pyc b/utils/__pycache__/content_generator.cpython-312.pyc
index 49b3b1c..1fef295 100644
Binary files a/utils/__pycache__/content_generator.cpython-312.pyc and b/utils/__pycache__/content_generator.cpython-312.pyc differ
diff --git a/utils/__pycache__/output_handler.cpython-312.pyc b/utils/__pycache__/output_handler.cpython-312.pyc
index 7da18fc..ee6c618 100644
Binary files a/utils/__pycache__/output_handler.cpython-312.pyc and b/utils/__pycache__/output_handler.cpython-312.pyc differ
diff --git a/utils/__pycache__/prompt_manager.cpython-312.pyc b/utils/__pycache__/prompt_manager.cpython-312.pyc
index 51fed84..e75407e 100644
Binary files a/utils/__pycache__/prompt_manager.cpython-312.pyc and b/utils/__pycache__/prompt_manager.cpython-312.pyc differ
diff --git a/utils/__pycache__/resource_loader.cpython-312.pyc b/utils/__pycache__/resource_loader.cpython-312.pyc
index 32f9cb2..8431007 100644
Binary files a/utils/__pycache__/resource_loader.cpython-312.pyc and b/utils/__pycache__/resource_loader.cpython-312.pyc differ
diff --git a/utils/__pycache__/tweet_generator.cpython-312.pyc b/utils/__pycache__/tweet_generator.cpython-312.pyc
index 116e488..10321ae 100644
Binary files a/utils/__pycache__/tweet_generator.cpython-312.pyc and b/utils/__pycache__/tweet_generator.cpython-312.pyc differ