diff --git a/backend/core/models/PromptComponent.py b/backend/core/models/PromptComponent.py
new file mode 100644
index 0000000..42f8d15
--- /dev/null
+++ b/backend/core/models/PromptComponent.py
@@ -0,0 +1,15 @@
+from pydantic import BaseModel, Field
+
+
+class PromptComponent(BaseModel):
+	"""预设组件类，代表一个独立的提示词模块"""
+
+	identifier: str = Field(..., description="唯一标识符，用于引用和定位组件")
+	name: str = Field(..., description="组件显示名称")
+	content: str = Field("", description="组件内容文本")
+	# 0：System，1：User，2：Assistant
+	role: int = Field(0, description="角色身份(0：System，1：User，2：Assistant)")
+	system_prompt: bool = Field(False, description="是否强制作为系统提示词处理")
+	marker: bool = Field(False, description="是否为动态插入点占位符")
+
+
diff --git a/backend/core/models/PromptList.py b/backend/core/models/PromptList.py
new file mode 100644
index 0000000..8080d64
--- /dev/null
+++ b/backend/core/models/PromptList.py
@@ -0,0 +1,52 @@
+from pydantic import BaseModel, Field
+import PromptComponent
+from typing import List, Dict, Any
+
+
+class AIDesignSpec(BaseModel):
+	"""AI设计规范类，包含模型生成的核心参数和动态结构配置"""
+
+	# [Base] 基础核心参数
+	temperature: float = Field(1.0, description="生成温度，控制随机性(0-2)")
+	frequency_penalty: float = Field(0.0, description="频率惩罚，降低重复token概率")
+	presence_penalty: float = Field(0.0, description="存在惩罚，鼓励谈论新话题")
+	top_p: float = Field(1.0, description="核采样，控制词汇选择范围")
+	top_k: int = Field(0, description="随机采样范围，从概率最高的K个词中选择")
+	top_a: float = Field(0.0, description="基于平方概率分布的采样")
+	min_p: float = Field(0.0, description="最小概率阈值")
+	repetition_penalty: float = Field(1.0, description="重复惩罚系数(1.0-1.2)")
+	openai_max_context: int = Field(2048, description="上下文窗口大小(Token上限)")
+	openai_max_tokens: int = Field(250, description="单次回复的最大长度")
+	max_context_unlocked: bool = Field(False, description="是否允许超出限制的上下文")
+	names_behavior: int = Field(0, description="名字处理行为(0=默认,1=始终包含,2=仅角色)")
+	send_if_empty: str = Field("", description="用户发送空消息时自动填充的内容")
+	impersonation_prompt: str = Field("", description="模仿模式下使用的提示词")
+	new_chat_prompt: str = Field("", description="开启新聊天时自动发送的系统提示")
+	new_group_chat_prompt: str = Field("", description="开启新群组聊天时的提示")
+	new_example_chat_prompt: str = Field("", description="新示例聊天的提示")
+	continue_nudge_prompt: str = Field("", description="续写功能触发的提示词")
+	bias_preset_selected: str = Field("", description="选用的偏见预设")
+	wi_format: str = Field("{0}", description="世界书条目的格式化字符串")
+	scenario_format: str = Field("{{scenario}}", description="场景描述的格式化字符串")
+	personality_format: str = Field("", description="角色性格的格式化字符串")
+	group_nudge_prompt: str = Field("", description="群组聊天中提示AI仅以特定角色回复的提示词")
+	stream_openai: bool = Field(True, description="是否使用流式输出")
+	assistant_prefill: str = Field("", description="强制AI回复的开头内容")
+	assistant_impersonation: str = Field("", description="模仿模式下强制AI回复的开头内容")
+	use_sysprompt: bool = Field(True, description="是否强制将提示词注入系统层")
+	squash_system_messages: bool = Field(False, description="是否压缩系统消息")
+	media_inlining: bool = Field(False, description="是否内联媒体描述")
+	continue_prefill: bool = Field(True, description="续写时是否预填充内容")
+	continue_postfix: str = Field(" ", description="续写时添加的后缀")
+	seed: int = Field(-1, description="随机种子(-1为随机)")
+	n: int = Field(1, description="生成回复的数量")
+
+	# [Dynamic] 动态结构
+	prompts: List[PromptComponent] = Field(
+			default_factory=list,
+			description="组件库，定义所有可用的积木块"
+	)
+	prompt_order: List[str] = Field(
+			default_factory=list,
+			description="组装说明书，定义构建最终提示词的顺序"
+	)
diff --git a/backend/nodes/PresetAssemblyNode.py b/backend/nodes/PresetAssemblyNode.py
new file mode 100644
index 0000000..b03932c
--- /dev/null
+++ b/backend/nodes/PresetAssemblyNode.py
@@ -0,0 +1,231 @@
+from typing import List, Dict, Any, Optional
+from pydantic import BaseModel, Field
+from backend.core.models.PromptList import AIDesignSpec
+from backend.core.models.PromptComponent import PromptComponent
+from enum import Enum
+
+from enum import Enum
+
+
+class SpecialIdentifier(str, Enum):
+	"""
+	特殊组件标识符枚举
+	定义所有提示词组件的类型及其在最终 Prompt 中的默认物理流向
+	顺序大致遵循：系统层 -> 角色层 -> 动态层 -> 历史层 -> 尾部指令
+	"""
+
+	WORLD_INFO_BEFORE = "worldInfoBefore"
+	"""前置世界书：通常用于全局设定（如物理法则），紧接在 Main Prompt 之后，拥有最高优先级"""
+
+	PERSONA_DESCRIPTION = "personaDescription"
+	"""用户设定：告诉 AI {{user}} 是谁，通常放在场景之后，完成“谁在对谁说话”的闭环"""
+
+	ENHANCE_DEFINITIONS = "enhanceDefinitions"
+	"""增强定义：通常是 "If you have more knowledge..."，用于补充 AI 的知识库，这里用rag获取"""
+
+	WORLD_INFO_AFTER = "worldInfoAfter"
+	"""后置世界书：通常用于特定场景规则，位于中间层底部，用于覆盖或补充前面的全局设定"""
+
+	CHAT_HISTORY = "chatHistory"
+	"""聊天历史：包含用户与 AI 的过往对话，占据提示词的下半部分"""
+
+	JAILBREAK = "jailbreak"
+	"""后置指令/注释，也即d0层：通常位于聊天记录之后、AI 生成之前，用于最后时刻的强调（如“不要重复”）"""
+
+class PresetAssemblyNode(BaseModel):
+	"""预设组装节点类，负责根据组装指令动态组装提示词内容"""
+
+	# 输入数据
+	design_spec: AIDesignSpec = Field(
+			...,
+			description="AI设计规范，包含组件库和组装顺序"
+	)
+	target_character_id: int = Field(
+			...,
+			description="目标角色ID，用于选择对应的组装指令"
+	)
+
+	# 内部状态（不参与序列化）
+	_component_map: Dict[str, PromptComponent] = Field(
+			default_factory=dict,
+			description="组件标识符到组件对象的映射"
+	)
+
+	def __init__(self, **data):
+		"""初始化方法，构建组件映射"""
+		super().__init__(**data)
+		# 构建组件映射字典，提高查找效率
+		self._component_map = {
+			comp.identifier: comp
+			for comp in self.design_spec.prompts
+		}
+
+	def _process_special_component(self, component: PromptComponent) -> Optional[Dict[str, Any]]:
+		"""
+		处理特殊组件（marker为True的组件）
+
+		参数:
+			component: 要处理的组件
+
+		返回:
+			Optional[Dict[str, Any]]: 处理后的消息，如果组件无法处理则返回None
+		"""
+		try:
+			# 尝试将标识符转换为枚举
+			special_id = SpecialIdentifier(component.identifier)
+
+			# 根据不同标识符执行不同处理逻辑
+			if special_id == SpecialIdentifier.CHAT_HISTORY:
+				return self._handle_chat_history(component)
+			elif special_id == SpecialIdentifier.WORLD_INFO_BEFORE:
+				return self._handle_world_info_before(component)
+			elif special_id == SpecialIdentifier.WORLD_INFO_AFTER:
+				return self._handle_world_info_after(component)
+			elif special_id == SpecialIdentifier.CHAR_DESCRIPTION:
+				return self._handle_char_description(component)
+			else:
+				# 未知特殊组件，使用默认处理
+				return self._process_regular_component(component)
+		except ValueError:
+			# 不是特殊标识符，使用默认处理
+			return self._process_regular_component(component)
+
+	def _process_special_component(self, component: PromptComponent) -> Optional[Dict[str, Any]]:
+		"""
+		处理特殊组件（marker为True的组件）
+
+		参数:
+			component: 要处理的组件
+
+		返回:
+			Optional[Dict[str, Any]]: 处理后的消息，如果组件无法处理则返回None
+		"""
+		try:
+			# 尝试将标识符转换为枚举
+			special_id = SpecialIdentifier(component.identifier)
+
+			# 根据不同标识符执行不同处理逻辑
+			if special_id == SpecialIdentifier.CHAT_HISTORY:
+				return self._handle_chat_history(component)
+			elif special_id == SpecialIdentifier.WORLD_INFO_BEFORE:
+				return self._handle_world_info_before(component)
+			elif special_id == SpecialIdentifier.WORLD_INFO_AFTER:
+				return self._handle_world_info_after(component)
+			elif special_id == SpecialIdentifier.DIALOGUE_EXAMPLES:
+				return self._handle_dialogue_examples(component)
+			elif special_id == SpecialIdentifier.CHAR_DESCRIPTION:
+				return self._handle_char_description(component)
+			elif special_id == SpecialIdentifier.CHAR_PERSONALITY:
+				return self._handle_char_personality(component)
+			elif special_id == SpecialIdentifier.SCENARIO:
+				return self._handle_scenario(component)
+			elif special_id == SpecialIdentifier.PERSONA_DESCRIPTION:
+				return self._handle_persona_description(component)
+			else:
+				# 未知特殊组件，使用默认处理
+				return self._process_regular_component(component)
+		except ValueError:
+			# 不是特殊标识符，使用默认处理
+			return self._process_regular_component(component)
+
+	def _process_regular_component(self, component: PromptComponent) -> Dict[str, Any]:
+		"""
+		处理普通组件（marker为False的组件）
+
+		参数:
+			component: 要处理的组件
+
+		返回:
+			Dict[str, Any]: 处理后的消息
+		"""
+		# 角色映射表
+		role_map = {0: "system", 1: "user", 2: "assistant"}
+
+		# 构建消息
+		message = {
+			"role": role_map.get(component.role, "system"),
+			"content": component.content
+		}
+
+		# 添加系统提示词标记
+		if component.system_prompt:
+			message["system_prompt"] = True
+
+		return message
+
+	# 以下为特殊组件处理方法
+
+	def _handle_chat_history(self, component: PromptComponent) -> Dict[str, Any]:
+		"""
+		处理聊天历史组件
+
+		参数:
+			component: 聊天历史组件
+
+		返回:
+			Dict[str, Any]: 处理后的消息
+		"""
+		# 这里应该从外部获取实际的聊天历史
+		# 示例实现，实际需要根据业务逻辑调整
+		return {
+			"role": "system",
+			"content": "聊天历史内容...",
+			"marker": True,
+			"type": "chat_history"
+		}
+
+	def _handle_world_info_before(self, component: PromptComponent) -> Dict[str, Any]:
+		"""
+		处理前置世界信息组件
+
+		参数:
+			component: 世界信息组件
+
+		返回:
+			Dict[str, Any]: 处理后的消息
+		"""
+		# 这里应该从外部获取实际的世界信息
+		return {
+			"role": "system",
+			"content": "前置世界信息...",
+			"marker": True,
+			"type": "world_info_before"
+		}
+
+	def _handle_world_info_after(self, component: PromptComponent) -> Dict[str, Any]:
+		"""
+		处理后置世界信息组件
+
+		参数:
+			component: 世界信息组件
+
+		返回:
+			Dict[str, Any]: 处理后的消息
+		"""
+		# 这里应该从外部获取实际的世界信息
+		return {
+			"role": "system",
+			"content": "后置世界信息...",
+			"marker": True,
+			"type": "world_info_after"
+		}
+
+
+	def _handle_char_description(self, component: PromptComponent) -> Dict[str, Any]:
+		"""
+		处理角色描述组件
+
+		参数:
+			component: 角色描述组件
+
+		返回:
+			Dict[str, Any]: 处理后的消息
+		"""
+		# 这里应该从外部获取实际的角色描述
+		return {
+			"role": "system",
+			"content": "角色描述内容...",
+			"marker": True,
+			"type": "char_description"
+		}
+
diff --git a/frontend-react/src/Store/Slices/LeftTabsSlice.jsx b/frontend-react/src/Store/Slices/LeftTabsSlice.jsx
new file mode 100644
index 0000000..e69de29
diff --git a/frontend-react/src/components/SideBarLeft/SideBarLeft.jsx b/frontend-react/src/components/SideBarLeft/SideBarLeft.jsx
index 2f83edd..f2cfaa2 100644
--- a/frontend-react/src/components/SideBarLeft/SideBarLeft.jsx
+++ b/frontend-react/src/components/SideBarLeft/SideBarLeft.jsx
@@ -3,6 +3,10 @@ import React from 'react';
 import './SideBarLeft.css';
 import { useSideBarLeftStore } from '../../Store/indexStore';
 import useSideBarRightStore from '../../Store/Slices/SideBarLeftSlice';
+import Gallery from './tab/Gallery';
+import ApiConfig from './tab/ApiConfig';
+import Presets from './tab/Presets';
+import WorldBook from './tab/WorldBook';
 
 const SideBarLeft = () => {
   const { activeTab, tabs, setActiveTab } = useSideBarLeftStore();
@@ -22,10 +26,10 @@ const SideBarLeft = () => {
       </div>
 
       <div className="sidebar-content">
-        {activeTab === 'gallery' && <div className="tab-content">画廊内容</div>}
-        {activeTab === 'api' && <div className="tab-content">API配置内容</div>}
-        {activeTab === 'presets' && <div className="tab-content">预设配置内容</div>}
-        {activeTab === 'worldbook' && <div className="tab-content">世界书内容</div>}
+        {activeTab === 'gallery' && <Gallery />}
+        {activeTab === 'api' && <ApiConfig />}
+        {activeTab === 'presets' && <Presets />}
+        {activeTab === 'worldbook' && <WorldBook />}
       </div>
     </div>
   );
diff --git a/frontend-react/src/components/SideBarLeft/tab/ApiConfig.jsx b/frontend-react/src/components/SideBarLeft/tab/ApiConfig.jsx
new file mode 100644
index 0000000..90d1bc4
--- /dev/null
+++ b/frontend-react/src/components/SideBarLeft/tab/ApiConfig.jsx
@@ -0,0 +1,13 @@
+import React from 'react';
+import '../tabcss/ApiConfig.css';
+
+const ApiConfig = () => {
+  return (
+    <div className="api-config-content">
+      <h2>API配置</h2>
+      {/* 在这里实现API配置的具体内容 */}
+    </div>
+  );
+};
+
+export default ApiConfig;
diff --git a/frontend-react/src/components/SideBarLeft/tab/Gallery.jsx b/frontend-react/src/components/SideBarLeft/tab/Gallery.jsx
new file mode 100644
index 0000000..5f3b242
--- /dev/null
+++ b/frontend-react/src/components/SideBarLeft/tab/Gallery.jsx
@@ -0,0 +1,13 @@
+import React from 'react';
+import '../tabcss/Gallery.css';
+
+const Gallery = () => {
+  return (
+    <div className="gallery-content">
+      <h2>画廊</h2>
+      {/* 在这里实现画廊的具体内容 */}
+    </div>
+  );
+};
+
+export default Gallery;
diff --git a/frontend-react/src/components/SideBarLeft/tab/Presets.jsx b/frontend-react/src/components/SideBarLeft/tab/Presets.jsx
new file mode 100644
index 0000000..5b033fb
--- /dev/null
+++ b/frontend-react/src/components/SideBarLeft/tab/Presets.jsx
@@ -0,0 +1,13 @@
+import React from 'react';
+import '../tabcss/Presets.css';
+
+const Presets = () => {
+  return (
+    <div className="presets-content">
+      <h2>预设配置</h2>
+      {/* 在这里实现预设配置的具体内容 */}
+    </div>
+  );
+};
+
+export default Presets;
diff --git a/frontend-react/src/components/SideBarLeft/tab/WorldBook.jsx b/frontend-react/src/components/SideBarLeft/tab/WorldBook.jsx
new file mode 100644
index 0000000..709f2e7
--- /dev/null
+++ b/frontend-react/src/components/SideBarLeft/tab/WorldBook.jsx
@@ -0,0 +1,13 @@
+import React from 'react';
+import '../tabcss/WorldBook.css';
+
+const WorldBook = () => {
+  return (
+    <div className="worldbook-content">
+      <h2>世界书</h2>
+      {/* 在这里实现世界书的具体内容 */}
+    </div>
+  );
+};
+
+export default WorldBook;
diff --git a/frontend-react/src/components/SideBarLeft/tabcss/ApiConfig.css b/frontend-react/src/components/SideBarLeft/tabcss/ApiConfig.css
new file mode 100644
index 0000000..1e37d21
--- /dev/null
+++ b/frontend-react/src/components/SideBarLeft/tabcss/ApiConfig.css
@@ -0,0 +1 @@
+WorldBook.css
\ No newline at end of file
diff --git a/frontend-react/src/components/SideBarLeft/tabcss/Gallery.css b/frontend-react/src/components/SideBarLeft/tabcss/Gallery.css
new file mode 100644
index 0000000..e69de29
diff --git a/frontend-react/src/components/SideBarLeft/tabcss/Presets.css b/frontend-react/src/components/SideBarLeft/tabcss/Presets.css
new file mode 100644
index 0000000..e69de29
diff --git a/frontend-react/src/components/SideBarLeft/tabcss/WorldBook.css b/frontend-react/src/components/SideBarLeft/tabcss/WorldBook.css
new file mode 100644
index 0000000..e69de29