添加插件文档和代码更新，支持可配置的文件标题来源，增强导出功能

.github/copilot-instructions.md

@@ -39,6 +39,14 @@ README 文件应包含以下内容：
 - 故障排除指南 / Troubleshooting guide
 - 版本和作者信息 / Version and author information
 
+### 官方文档 (Official Documentation)
+
+如果插件被合并到主仓库，还需更新 `docs/` 目录下的相关文档：
+- `docs/plugins/{type}/plugin-name.md`
+- `docs/plugins/{type}/plugin-name.zh.md`
+
+其中 `{type}` 对应插件类型（如 `actions`, `filters`, `pipes` 等）。
+
 ---
 
 ## 📝 文档字符串规范 (Docstring Standard)
@@ -370,6 +378,94 @@ from open_webui.models.users import Users
 
 ---
 
+## 📄 文件导出与命名规范 (File Export and Naming)
+
+对于涉及文件导出的插件（通常是 Action），必须提供灵活的标题生成策略。
+
+### Valves 配置 (Valves Configuration)
+
+应包含 `TITLE_SOURCE` 选项：
+
+```python
+class Valves(BaseModel):
+    TITLE_SOURCE: str = Field(
+        default="chat_title",
+        description="Title Source: 'chat_title', 'ai_generated', 'markdown_title'",
+    )
+```
+
+### 标题获取逻辑 (Title Retrieval Logic)
+
+1.  **chat_title**: 尝试从 `body` 获取，若失败且有 `chat_id`，则从数据库获取 (`Chats.get_chat_by_id`)。
+2.  **markdown_title**: 从 Markdown 内容提取第一个 H1 或 H2。
+3.  **ai_generated**: 使用轻量级 Prompt 让 AI 生成简短标题。
+
+### 优先级与回退 (Priority and Fallback)
+
+代码应根据 `TITLE_SOURCE` 优先尝试指定方法，若失败则按以下顺序回退：
+`chat_title` -> `markdown_title` -> `user_name + date`
+
+```python
+# 核心逻辑示例
+if self.valves.TITLE_SOURCE == "chat_title":
+    title = chat_title
+elif self.valves.TITLE_SOURCE == "markdown_title":
+    title = self.extract_title(content)
+elif self.valves.TITLE_SOURCE == "ai_generated":
+    title = await self.generate_title_using_ai(...)
+```
+
+### AI 标题生成实现 (AI Title Generation Implementation)
+
+如果支持 `ai_generated` 选项，应实现类似以下的方法：
+
+```python
+async def generate_title_using_ai(
+    self, 
+    body: dict, 
+    content: str, 
+    user_id: str, 
+    request: Any
+) -> str:
+    """Generates a short title using the current LLM model."""
+    if not request:
+        return ""
+
+    try:
+        # 获取当前用户和模型
+        user_obj = Users.get_user_by_id(user_id)
+        model = body.get("model")
+
+        # 构造请求
+        payload = {
+            "model": model,
+            "messages": [
+                {
+                    "role": "system", 
+                    "content": "You are a helpful assistant. Generate a short, concise title (max 10 words) for the following text. Do not use quotes. Only output the title."
+                },
+                {
+                    "role": "user", 
+                    "content": content[:2000]  # 限制上下文长度以节省 Token
+                }
+            ],
+            "stream": False,
+        }
+
+        # 调用 OpenWebUI 内部生成接口
+        response = await generate_chat_completion(request, payload, user_obj)
+        
+        if response and "choices" in response:
+            return response["choices"][0]["message"]["content"].strip()
+            
+    except Exception as e:
+        logger.error(f"Error generating title: {e}")
+
+    return ""
+```
+
+---
+
 ## ✅ 开发检查清单 (Development Checklist)
 
 开发新插件时，请确保完成以下检查：

docs/plugins/actions/export-to-word.md

@@ -17,7 +17,17 @@ The Export to Word plugin converts chat messages from Markdown to a polished Wor
 - :material-format-bold: **Rich Markdown Support**: Headings, bold/italic, lists, tables
 - :material-code-tags: **Syntax Highlighting**: Pygments-powered code blocks
 - :material-format-quote-close: **Styled Blockquotes**: Left-border gray quote styling
-- :material-file-document-outline: **Smart Filenames**: Prefers chat title → Markdown title → user/date
+- :material-file-document-outline: **Smart Filenames**: Configurable title source (Chat Title, AI Generated, or Markdown Title)
+
+---
+
+## Configuration
+
+You can configure the following settings via the **Valves** button in the plugin settings:
+
+| Valve          | Description                                                                                 | Default      |
+| :------------- | :------------------------------------------------------------------------------------------ | :----------- |
+| `TITLE_SOURCE` | Source for document title/filename. Options: `chat_title`, `ai_generated`, `markdown_title` | `chat_title` |
 
 ---
 
@@ -39,21 +49,21 @@ The Export to Word plugin converts chat messages from Markdown to a polished Wor
 
 ## Supported Markdown
 
-| Syntax | Word Result |
+| Syntax                              | Word Result                    |
-| :---------------------------------- | :-------------------------------- |
+| :---------------------------------- | :----------------------------- |
-| `# Heading 1` to `###### Heading 6` | Heading levels 1-6 |
+| `# Heading 1` to `###### Heading 6` | Heading levels 1-6             |
-| `**bold**` / `__bold__` | Bold text |
+| `**bold**` / `__bold__`             | Bold text                      |
-| `*italic*` / `_italic_` | Italic text |
+| `*italic*` / `_italic_`             | Italic text                    |
-| `***bold italic***` | Bold + Italic |
+| `***bold italic***`                 | Bold + Italic                  |
-| `` `inline code` `` | Monospace with gray background |
+| `` `inline code` ``                 | Monospace with gray background |
-| <code>``` code block ```</code> | Syntax-highlighted code block |
+| <code>``` code block ```</code>     | Syntax-highlighted code block  |
-| `> blockquote` | Left-bordered gray italic text |
+| `> blockquote`                      | Left-bordered gray italic text |
-| `[link](url)` | Blue underlined link |
+| `[link](url)`                       | Blue underlined link           |
-| `~~strikethrough~~` | Strikethrough |
+| `~~strikethrough~~`                 | Strikethrough                  |
-| `- item` / `* item` | Bullet list |
+| `- item` / `* item`                 | Bullet list                    |
-| `1. item` | Numbered list |
+| `1. item`                           | Numbered list                  |
-| Markdown tables | Grid table |
+| Markdown tables                     | Grid table                     |
-| `---` / `***` | Horizontal rule |
+| `---` / `***`                       | Horizontal rule                |
 
 ---
 

docs/plugins/actions/export-to-word.zh.md

@@ -17,7 +17,17 @@ Export to Word 插件会把聊天消息从 Markdown 转成精致的 Word 文档
 - :material-format-bold: **丰富 Markdown 支持**：标题、粗斜体、列表、表格
 - :material-code-tags: **语法高亮**：Pygments 驱动的代码块上色
 - :material-format-quote-close: **引用样式**：左侧边框的灰色斜体引用
-- :material-file-document-outline: **智能文件名**：优先对话标题 → Markdown 标题 → 用户/日期
+- :material-file-document-outline: **智能文件名**：可配置标题来源（对话标题、AI 生成或 Markdown 标题）
+
+---
+
+## 配置
+
+您可以通过插件设置中的 **Valves** 按钮配置以下选项：
+
+| Valve          | 说明                                                                                                             | 默认值       |
+| :------------- | :--------------------------------------------------------------------------------------------------------------- | :----------- |
+| `TITLE_SOURCE` | 文档标题/文件名的来源。选项：`chat_title` (对话标题), `ai_generated` (AI 生成), `markdown_title` (Markdown 标题) | `chat_title` |
 
 ---
 
@@ -39,21 +49,21 @@ Export to Word 插件会把聊天消息从 Markdown 转成精致的 Word 文档
 
 ## 支持的 Markdown
 
-| 语法 | Word 效果 |
+| 语法                        | Word 效果           |
-| :---------------------------------- | :-------------------------------- |
+| :-------------------------- | :------------------ |
-| `# 标题1` 到 `###### 标题6` | 标题级别 1-6 |
+| `# 标题1` 到 `###### 标题6` | 标题级别 1-6        |
-| `**粗体**` / `__粗体__` | 粗体文本 |
+| `**粗体**` / `__粗体__`     | 粗体文本            |
-| `*斜体*` / `_斜体_` | 斜体文本 |
+| `*斜体*` / `_斜体_`         | 斜体文本            |
-| `***粗斜体***` | 粗体 + 斜体 |
+| `***粗斜体***`              | 粗体 + 斜体         |
-| `` `行内代码` `` | 等宽字体 + 灰色背景 |
+| `` `行内代码` ``            | 等宽字体 + 灰色背景 |
-| <code>``` 代码块 ```</code> | 语法高亮代码块 |
+| <code>``` 代码块 ```</code> | 语法高亮代码块      |
-| `> 引用文本` | 左侧边框的灰色斜体 |
+| `> 引用文本`                | 左侧边框的灰色斜体  |
-| `[链接](url)` | 蓝色下划线链接 |
+| `[链接](url)`               | 蓝色下划线链接      |
-| `~~删除线~~` | 删除线 |
+| `~~删除线~~`                | 删除线              |
-| `- 项目` / `* 项目` | 无序列表 |
+| `- 项目` / `* 项目`         | 无序列表            |
-| `1. 项目` | 有序列表 |
+| `1. 项目`                   | 有序列表            |
-| Markdown 表格 | 带边框表格 |
+| Markdown 表格               | 带边框表格          |
-| `---` / `***` | 水平分割线 |
+| `---` / `***`               | 水平分割线          |
 
 ---
 

plugins/actions/export_to_docx/README.md

@@ -9,7 +9,16 @@ Export current conversation from Markdown to Word (.docx) with **syntax highligh
 - **Syntax Highlighting**: Code blocks are highlighted with Pygments (supports 500+ languages).
 - **Blockquote Support**: Markdown blockquotes are rendered with left border and gray styling.
 - **Multi-language Support**: Properly handles both Chinese and English text without garbled characters.
-- **Smarter Filenames**: Prefers chat title (from body or chat_id lookup) → first Markdown h1/h2 → user + date.
+- **Smarter Filenames**: Configurable title source (Chat Title, AI Generated, or Markdown Title).
+
+## Configuration
+
+You can configure the following settings via the **Valves** button in the plugin settings:
+
+- **TITLE_SOURCE**: Choose how the document title/filename is generated.
+    - `chat_title`: Use the conversation title (default).
+    - `ai_generated`: Use AI to generate a short title based on the content.
+    - `markdown_title`: Extract the first h1/h2 heading from the Markdown content.
 
 ## Supported Markdown Syntax
 

plugins/actions/export_to_docx/README_CN.md

@@ -9,7 +9,16 @@
 - **代码语法高亮**：使用 Pygments 库为代码块添加语法高亮（支持 500+ 种语言）。
 - **引用块支持**：Markdown 引用块会渲染为带左侧边框的灰色斜体样式。
 - **多语言支持**：正确处理中文和英文文本，无乱码问题。
-- **更智能的文件名**：优先使用对话标题（来自请求体或基于 chat_id 查询），其次 Markdown 一级/二级标题，最后用户+日期。
+- **更智能的文件名**：可配置标题来源（对话标题、AI 生成或 Markdown 标题）。
+
+## 配置 (Configuration)
+
+您可以通过插件设置中的 **Valves** 按钮配置以下选项：
+
+- **TITLE_SOURCE**：选择文档标题/文件名的生成方式。
+    - `chat_title`：使用对话标题（默认）。
+    - `ai_generated`：使用 AI 根据内容生成简短标题。
+    - `markdown_title`：从 Markdown 内容中提取第一个一级或二级标题。
 
 ## 支持的 Markdown 语法
 

plugins/actions/export_to_docx/export_to_word.py

@@ -25,6 +25,9 @@ from docx.enum.style import WD_STYLE_TYPE
 from docx.oxml.ns import qn
 from docx.oxml import OxmlElement
 from open_webui.models.chats import Chats
+from open_webui.models.users import Users
+from open_webui.utils.chat import generate_chat_completion
+from pydantic import BaseModel, Field
 
 # Pygments for syntax highlighting
 try:
@@ -45,8 +48,14 @@ logger = logging.getLogger(__name__)
 
 
 class Action:
+    class Valves(BaseModel):
+        TITLE_SOURCE: str = Field(
+            default="chat_title",
+            description="Title Source: 'chat_title' (Chat Title), 'ai_generated' (AI Generated), 'markdown_title' (Markdown Title)",
+        )
+
     def __init__(self):
-        pass
+        self.valves = self.Valves()
 
     async def _send_notification(self, emitter: Callable, type: str, content: str):
         await emitter(
@@ -60,6 +69,7 @@ class Action:
         __event_emitter__=None,
         __event_call__: Optional[Callable[[Any], Awaitable[None]]] = None,
         __metadata__: Optional[dict] = None,
+        __request__: Optional[Any] = None,
     ):
         logger.info(f"action:{__name__}")
 
@@ -101,26 +111,54 @@ class Action:
                     )
                     return
 
-                # Generate filename (prefer chat title; fetch via chat_id if missing; then markdown title; then fallback)
+                # Generate filename
+                title = ""
                 chat_id = self.extract_chat_id(body, __metadata__)
-                chat_title = self.extract_chat_title(body)
+
-                if not chat_title and chat_id:
+                # Fetch chat_title directly via chat_id as it's usually missing in body
+                chat_title = ""
+                if chat_id:
                     chat_title = await self.fetch_chat_title(chat_id, user_id)
-                title = self.extract_title(message_content)
+
+                if (
+                    self.valves.TITLE_SOURCE == "chat_title"
+                    or not self.valves.TITLE_SOURCE
+                ):
+                    title = chat_title
+                elif self.valves.TITLE_SOURCE == "markdown_title":
+                    title = self.extract_title(message_content)
+                elif self.valves.TITLE_SOURCE == "ai_generated":
+                    title = await self.generate_title_using_ai(
+                        body, message_content, user_id, __request__
+                    )
+
+                # Fallback logic
+                if not title:
+                    if self.valves.TITLE_SOURCE != "chat_title" and chat_title:
+                        title = chat_title
+                    elif self.valves.TITLE_SOURCE != "markdown_title":
+                        extracted = self.extract_title(message_content)
+                        if extracted:
+                            title = extracted
+
                 current_datetime = datetime.datetime.now()
                 formatted_date = current_datetime.strftime("%Y%m%d")
 
-                if chat_title:
+                if title:
-                    filename = f"{self.clean_filename(chat_title)}.docx"
-                elif title:
                     filename = f"{self.clean_filename(title)}.docx"
                 else:
                     filename = f"{user_name}_{formatted_date}.docx"
 
+                top_heading = ""
+                if chat_title:
+                    top_heading = chat_title
+                elif title:
+                    top_heading = title
+
                 # Create Word document; if no h1 exists, inject chat title as h1
                 has_h1 = bool(re.search(r"^#\s+.+$", message_content, re.MULTILINE))
                 doc = self.markdown_to_docx(
-                    message_content, top_heading=chat_title, has_h1=has_h1
+                    message_content, top_heading=top_heading, has_h1=has_h1
                 )
 
                 # Save to memory
@@ -194,6 +232,36 @@ class Action:
                     f"Error exporting Word document: {str(e)}",
                 )
 
+    async def generate_title_using_ai(
+        self, body: dict, content: str, user_id: str, request: Any
+    ) -> str:
+        if not request:
+            return ""
+
+        try:
+            user_obj = Users.get_user_by_id(user_id)
+            model = body.get("model")
+
+            payload = {
+                "model": model,
+                "messages": [
+                    {
+                        "role": "system",
+                        "content": "You are a helpful assistant. Generate a short, concise title (max 10 words) for the following text. Do not use quotes. Only output the title.",
+                    },
+                    {"role": "user", "content": content[:2000]},  # Limit content length
+                ],
+                "stream": False,
+            }
+
+            response = await generate_chat_completion(request, payload, user_obj)
+            if response and "choices" in response:
+                return response["choices"][0]["message"]["content"].strip()
+        except Exception as e:
+            logger.error(f"Error generating title: {e}")
+
+        return ""
+
     def extract_title(self, content: str) -> str:
         """Extract title from Markdown h1/h2 only"""
         lines = content.split("\n")

plugins/actions/export_to_docx/导出为Word.py

@@ -25,6 +25,9 @@ from docx.enum.style import WD_STYLE_TYPE
 from docx.oxml.ns import qn
 from docx.oxml import OxmlElement
 from open_webui.models.chats import Chats
+from open_webui.models.users import Users
+from open_webui.utils.chat import generate_chat_completion
+from pydantic import BaseModel, Field
 
 # Pygments for syntax highlighting
 try:
@@ -45,8 +48,14 @@ logger = logging.getLogger(__name__)
 
 
 class Action:
+    class Valves(BaseModel):
+        TITLE_SOURCE: str = Field(
+            default="chat_title",
+            description="标题来源: 'chat_title' (对话标题), 'ai_generated' (AI 生成), 'markdown_title' (Markdown 标题)",
+        )
+
     def __init__(self):
-        pass
+        self.valves = self.Valves()
 
     async def _send_notification(self, emitter: Callable, type: str, content: str):
         await emitter(
@@ -60,6 +69,7 @@ class Action:
         __event_emitter__=None,
         __event_call__: Optional[Callable[[Any], Awaitable[None]]] = None,
         __metadata__: Optional[dict] = None,
+        __request__: Optional[Any] = None,
     ):
         logger.info(f"action:{__name__}")
 
@@ -98,26 +108,54 @@ class Action:
                     )
                     return
 
-                # 生成文件名（优先对话标题；若缺失则通过 chat_id 查询；再到 Markdown 标题；最后用户+日期）
+                # 生成文件名
+                title = ""
                 chat_id = self.extract_chat_id(body, __metadata__)
-                chat_title = self.extract_chat_title(body)
+
-                if not chat_title and chat_id:
+                # 直接通过 chat_id 获取标题，因为 body 中通常不包含标题
+                chat_title = ""
+                if chat_id:
                     chat_title = await self.fetch_chat_title(chat_id, user_id)
-                title = self.extract_title(message_content)
+
+                # 根据配置决定文件名使用的标题
+                if (
+                    self.valves.TITLE_SOURCE == "chat_title"
+                    or not self.valves.TITLE_SOURCE
+                ):
+                    title = chat_title
+                elif self.valves.TITLE_SOURCE == "markdown_title":
+                    title = self.extract_title(message_content)
+                elif self.valves.TITLE_SOURCE == "ai_generated":
+                    title = await self.generate_title_using_ai(
+                        body, message_content, user_id, __request__
+                    )
+
                 current_datetime = datetime.datetime.now()
                 formatted_date = current_datetime.strftime("%Y%m%d")
 
-                if chat_title:
+                if title:
-                    filename = f"{self.clean_filename(chat_title)}.docx"
-                elif title:
                     filename = f"{self.clean_filename(title)}.docx"
                 else:
                     filename = f"{user_name}_{formatted_date}.docx"
 
                 # 创建 Word 文档；若正文无一级标题，使用对话标题作为一级标题
+                # 如果选择了 chat_title 且获取到了，则作为 top_heading
+                # 如果选择了其他方式，title 就是文件名，也可以作为 top_heading
+
+                # 保持原有逻辑：top_heading 主要是为了在文档开头补充标题
+                # 这里我们尽量使用 chat_title 作为 top_heading，如果它存在的话，因为它通常是对话的主题
+                # 即使文件名是 AI 生成的，文档内的标题用 chat_title 也是合理的
+                # 但如果用户选择了 markdown_title，可能不希望插入 chat_title
+
+                top_heading = ""
+                if chat_title:
+                    top_heading = chat_title
+                elif title:
+                    top_heading = title
+
                 has_h1 = bool(re.search(r"^#\s+.+$", message_content, re.MULTILINE))
                 doc = self.markdown_to_docx(
-                    message_content, top_heading=chat_title, has_h1=has_h1
+                    message_content, top_heading=top_heading, has_h1=has_h1
                 )
 
                 # 保存到内存
@@ -189,6 +227,36 @@ class Action:
                     __event_emitter__, "error", f"导出 Word 文档时出错: {str(e)}"
                 )
 
+    async def generate_title_using_ai(
+        self, body: dict, content: str, user_id: str, request: Any
+    ) -> str:
+        if not request:
+            return ""
+
+        try:
+            user_obj = Users.get_user_by_id(user_id)
+            model = body.get("model")
+
+            payload = {
+                "model": model,
+                "messages": [
+                    {
+                        "role": "system",
+                        "content": "You are a helpful assistant. Generate a short, concise title (max 10 words) for the following text. Do not use quotes. Only output the title.",
+                    },
+                    {"role": "user", "content": content[:2000]},  # Limit content length
+                ],
+                "stream": False,
+            }
+
+            response = await generate_chat_completion(request, payload, user_obj)
+            if response and "choices" in response:
+                return response["choices"][0]["message"]["content"].strip()
+        except Exception as e:
+            logger.error(f"Error generating title: {e}")
+
+        return ""
+
     def extract_title(self, content: str) -> str:
         """从 Markdown 内容提取一级/二级标题"""
         lines = content.split("\n")