feat: add note for remove_think & remove dify remove cot code

* fix: update invoke_embedding to return only embeddings from client.embed

* fix: Fixed the incorrect extraction method of sender ID when converting aiocqhttp reply messages

.gitignore

@@ -42,4 +42,5 @@ botpy.log*
 test.py
 /web_ui
 .venv/
-uv.lock
+uv.lock
+/test

README.md

@@ -6,14 +6,16 @@
 
 <div align="center">
 
-简体中文 / [English](README_EN.md) / [日本語](README_JP.md) / (PR for your language)
+<a href="https://hellogithub.com/repository/langbot-app/LangBot" target="_blank"><img src="https://abroad.hellogithub.com/v1/widgets/recommend.svg?rid=5ce8ae2aa4f74316bf393b57b952433c&claim_uid=gtmc6YWjMZkT21R" alt="Featured｜HelloGitHub" style="width: 250px; height: 54px;" width="250" height="54" /></a>
+
+[English](README_EN.md) / 简体中文 / [繁體中文](README_TW.md) / [日本語](README_JP.md) / (PR for your language)
 
 [![Discord](https://img.shields.io/discord/1335141740050649118?logo=discord&labelColor=%20%235462eb&logoColor=%20%23f5f5f5&color=%20%235462eb)](https://discord.gg/wdNEHETs87)
 [![QQ Group](https://img.shields.io/badge/%E7%A4%BE%E5%8C%BAQQ%E7%BE%A4-966235608-blue)](https://qm.qq.com/q/JLi38whHum)
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/langbot-app/LangBot)
 [![GitHub release (latest by date)](https://img.shields.io/github/v/release/langbot-app/LangBot)](https://github.com/langbot-app/LangBot/releases/latest)
 <img src="https://img.shields.io/badge/python-3.10 ~ 3.13 -blue.svg" alt="python">
-[![star](https://gitcode.com/langbot-app/LangBot/star/badge.svg)](https://gitcode.com/langbot-app/LangBot)
+[![star](https://gitcode.com/RockChinQ/LangBot/star/badge.svg)](https://gitcode.com/RockChinQ/LangBot)
 
 <a href="https://langbot.app">项目主页</a> ｜
 <a href="https://docs.langbot.app/zh/insight/guide.html">部署文档</a> ｜
@@ -25,12 +27,7 @@
 
 </p>
 
-## ✨ 特性
-
-- 💬 大模型对话、Agent：支持多种大模型，适配群聊和私聊；具有多轮对话、工具调用、多模态能力，并深度适配 [Dify](https://dify.ai)。目前支持 QQ、QQ频道、企业微信、个人微信、飞书、Discord、Telegram 等平台。
-- 🛠️ 高稳定性、功能完备：原生支持访问控制、限速、敏感词过滤等机制；配置简单，支持多种部署方式。支持多流水线配置，不同机器人用于不同应用场景。
-- 🧩 插件扩展、活跃社区：支持事件驱动、组件扩展等插件机制；适配 Anthropic [MCP 协议](https://modelcontextprotocol.io/)；目前已有数百个插件。
-- 😻 Web 管理面板：支持通过浏览器管理 LangBot 实例，不再需要手动编写配置文件。
+LangBot 是一个开源的大语言模型原生即时通信机器人开发平台，旨在提供开箱即用的 IM 机器人开发体验，具有 Agent、RAG、MCP 等多种 LLM 应用功能，适配全球主流即时通信平台，并提供丰富的 API 接口，支持自定义开发。
 
 ## 📦 开始使用
 
@@ -64,23 +61,25 @@ docker compose up -d
 
 直接使用发行版运行，查看文档[手动部署](https://docs.langbot.app/zh/deploy/langbot/manual.html)。
 
-## 📸 效果展示
+## 😎 保持更新
 
-<img alt="bots" src="https://docs.langbot.app/webui/bot-page.png" width="450px"/>
+点击仓库右上角 Star 和 Watch 按钮，获取最新动态。
 
-<img alt="bots" src="https://docs.langbot.app/webui/create-model.png" width="450px"/>
+![star gif](https://docs.langbot.app/star.gif)
 
-<img alt="bots" src="https://docs.langbot.app/webui/edit-pipeline.png" width="450px"/>
+## ✨ 特性
 
-<img alt="bots" src="https://docs.langbot.app/webui/plugin-market.png" width="450px"/>
+- 💬 大模型对话、Agent：支持多种大模型，适配群聊和私聊；具有多轮对话、工具调用、多模态、流式输出能力，自带 RAG（知识库）实现，并深度适配 [Dify](https://dify.ai)。
+- 🤖 多平台支持：目前支持 QQ、QQ频道、企业微信、个人微信、飞书、Discord、Telegram 等平台。
+- 🛠️ 高稳定性、功能完备：原生支持访问控制、限速、敏感词过滤等机制；配置简单，支持多种部署方式。支持多流水线配置，不同机器人用于不同应用场景。
+- 🧩 插件扩展、活跃社区：支持事件驱动、组件扩展等插件机制；适配 Anthropic [MCP 协议](https://modelcontextprotocol.io/)；目前已有数百个插件。
+- 😻 Web 管理面板：支持通过浏览器管理 LangBot 实例，不再需要手动编写配置文件。
 
-<img alt="回复效果（带有联网插件）" src="https://docs.langbot.app/QChatGPT-0516.png" width="500px"/>
+详细规格特性请访问[文档](https://docs.langbot.app/zh/insight/features.html)。
 
-- WebUI Demo: https://demo.langbot.dev/
-    - 登录信息：邮箱：`demo@langbot.app` 密码：`langbot123456`
-    - 注意：仅展示webui效果，公开环境，请不要在其中填入您的任何敏感信息。
-
-## 🔌 组件兼容性
+或访问 demo 环境：https://demo.langbot.dev/  
+  - 登录信息：邮箱：`demo@langbot.app` 密码：`langbot123456`
+  - 注意：仅展示 WebUI 效果，公开环境，请不要在其中填入您的任何敏感信息。
 
 ### 消息平台
 
@@ -97,10 +96,6 @@ docker compose up -d
 | Discord | ✅ |  |
 | Telegram | ✅ |  |
 | Slack | ✅ |  |
-| LINE | 🚧 |  |
-| WhatsApp | 🚧 |  |
-
-🚧: 正在开发中
 
 ### 大模型能力
 
@@ -114,6 +109,7 @@ docker compose up -d
 | [智谱AI](https://open.bigmodel.cn/) | ✅ |  |
 | [优云智算](https://www.compshare.cn/?ytag=GPU_YY-gh_langbot) | ✅ | 大模型和 GPU 资源平台 |
 | [PPIO](https://ppinfra.com/user/register?invited_by=QJKFYD&utm_source=github_langbot) | ✅ | 大模型和 GPU 资源平台 |
+| [胜算云](https://www.shengsuanyun.com/?from=CH_KYIPP758) | ✅ | 大模型和 GPU 资源平台 |
 | [302.AI](https://share.302.ai/SuTG99) | ✅ | 大模型聚合平台 |
 | [Google Gemini](https://aistudio.google.com/prompts/new_chat) | ✅ | |
 | [Dify](https://dify.ai) | ✅ | LLMOps 平台 |
@@ -147,9 +143,3 @@ docker compose up -d
 <a href="https://github.com/langbot-app/LangBot/graphs/contributors">
   <img src="https://contrib.rocks/image?repo=langbot-app/LangBot" />
 </a>
-
-## 😎 保持更新
-
-点击仓库右上角 Star 和 Watch 按钮，获取最新动态。
-
-![star gif](https://docs.langbot.app/star.gif)

README_EN.md

@@ -5,7 +5,7 @@
 
 <div align="center">
 
-[简体中文](README.md) / English / [日本語](README_JP.md) / (PR for your language)
+English / [简体中文](README.md) / [繁體中文](README_TW.md) / [日本語](README_JP.md) / (PR for your language)
 
 [![Discord](https://img.shields.io/discord/1335141740050649118?logo=discord&labelColor=%20%235462eb&logoColor=%20%23f5f5f5&color=%20%235462eb)](https://discord.gg/wdNEHETs87)
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/langbot-app/LangBot)
@@ -21,12 +21,7 @@
 
 </p>
 
-## ✨ Features
-
-- 💬 Chat with LLM / Agent: Supports multiple LLMs, adapt to group chats and private chats; Supports multi-round conversations, tool calls, and multi-modal capabilities. Deeply integrates with [Dify](https://dify.ai). Currently supports QQ, QQ Channel, WeCom, personal WeChat, Lark, DingTalk, Discord, Telegram, etc.
-- 🛠️ High Stability, Feature-rich: Native access control, rate limiting, sensitive word filtering, etc. mechanisms; Easy to use, supports multiple deployment methods. Supports multiple pipeline configurations, different bots can be used for different scenarios.
-- 🧩 Plugin Extension, Active Community: Support event-driven, component extension, etc. plugin mechanisms; Integrate Anthropic [MCP protocol](https://modelcontextprotocol.io/); Currently has hundreds of plugins.
-- 😻 [New] Web UI: Support management LangBot instance through the browser. No need to manually write configuration files.
+LangBot is an open-source LLM native instant messaging robot development platform, aiming to provide out-of-the-box IM robot development experience, with Agent, RAG, MCP and other LLM application functions, adapting to global instant messaging platforms, and providing rich API interfaces, supporting custom development.
 
 ## 📦 Getting Started
 
@@ -60,23 +55,25 @@ Community contributed Zeabur template.
 
 Directly use the released version to run, see the [Manual Deployment](https://docs.langbot.app/en/deploy/langbot/manual.html) documentation.
 
-## 📸 Demo
+## 😎 Stay Ahead
 
-<img alt="bots" src="https://docs.langbot.app/webui/bot-page.png" width="400px"/>
+Click the Star and Watch button in the upper right corner of the repository to get the latest updates.
 
-<img alt="bots" src="https://docs.langbot.app/webui/create-model.png" width="400px"/>
+![star gif](https://docs.langbot.app/star.gif)
 
-<img alt="bots" src="https://docs.langbot.app/webui/edit-pipeline.png" width="400px"/>
+## ✨ Features
 
-<img alt="bots" src="https://docs.langbot.app/webui/plugin-market.png" width="400px"/>
+- 💬 Chat with LLM / Agent: Supports multiple LLMs, adapt to group chats and private chats; Supports multi-round conversations, tool calls, multi-modal, and streaming output capabilities. Built-in RAG (knowledge base) implementation, and deeply integrates with [Dify](https://dify.ai).
+- 🤖 Multi-platform Support: Currently supports QQ, QQ Channel, WeCom, personal WeChat, Lark, DingTalk, Discord, Telegram, etc.
+- 🛠️ High Stability, Feature-rich: Native access control, rate limiting, sensitive word filtering, etc. mechanisms; Easy to use, supports multiple deployment methods. Supports multiple pipeline configurations, different bots can be used for different scenarios.
+- 🧩 Plugin Extension, Active Community: Support event-driven, component extension, etc. plugin mechanisms; Integrate Anthropic [MCP protocol](https://modelcontextprotocol.io/); Currently has hundreds of plugins.
+- 😻 Web UI: Support management LangBot instance through the browser. No need to manually write configuration files.
 
-<img alt="Reply Effect (with Internet Plugin)" src="https://docs.langbot.app/QChatGPT-0516.png" width="500px"/>
+For more detailed specifications, please refer to the [documentation](https://docs.langbot.app/en/insight/features.html).
 
-- WebUI Demo: https://demo.langbot.dev/
-    - Login information: Email: `demo@langbot.app` Password: `langbot123456`
-    - Note: Only the WebUI effect is shown, please do not fill in any sensitive information in the public environment.
-
-## 🔌 Component Compatibility
+Or visit the demo environment: https://demo.langbot.dev/
+  - Login information: Email: `demo@langbot.app` Password: `langbot123456`
+  - Note: For WebUI demo only, please do not fill in any sensitive information in the public environment.
 
 ### Message Platform
 
@@ -92,10 +89,6 @@ Directly use the released version to run, see the [Manual Deployment](https://do
 | Discord | ✅ |  |
 | Telegram | ✅ |  |
 | Slack | ✅ |  |
-| LINE | 🚧 |  |
-| WhatsApp | 🚧 |  |
-
-🚧: In development
 
 ### LLMs
 
@@ -110,6 +103,7 @@ Directly use the released version to run, see the [Manual Deployment](https://do
 | [CompShare](https://www.compshare.cn/?ytag=GPU_YY-gh_langbot) | ✅ | LLM and GPU resource platform |
 | [Dify](https://dify.ai) | ✅ | LLMOps platform |
 | [PPIO](https://ppinfra.com/user/register?invited_by=QJKFYD&utm_source=github_langbot) | ✅ | LLM and GPU resource platform |
+| [ShengSuanYun](https://www.shengsuanyun.com/?from=CH_KYIPP758) | ✅ | LLM and GPU resource platform |
 | [302.AI](https://share.302.ai/SuTG99) | ✅ | LLM gateway(MaaS) |
 | [Google Gemini](https://aistudio.google.com/prompts/new_chat) | ✅ | |
 | [Ollama](https://ollama.com/) | ✅ | Local LLM running platform |
@@ -128,9 +122,3 @@ Thank you for the following [code contributors](https://github.com/langbot-app/L
 <a href="https://github.com/langbot-app/LangBot/graphs/contributors">
   <img src="https://contrib.rocks/image?repo=langbot-app/LangBot" />
 </a>
-
-## 😎 Stay Ahead
-
-Click the Star and Watch button in the upper right corner of the repository to get the latest updates.
-
-![star gif](https://docs.langbot.app/star.gif)

README_JP.md

@@ -5,7 +5,7 @@
 
 <div align="center">
 
-[简体中文](README.md) / [English](README_EN.md) / 日本語 / (PR for your language)
+[English](README_EN.md) / [简体中文](README.md) / [繁體中文](README_TW.md) / 日本語 / (PR for your language)
 
 [![Discord](https://img.shields.io/discord/1335141740050649118?logo=discord&labelColor=%20%235462eb&logoColor=%20%23f5f5f5&color=%20%235462eb)](https://discord.gg/wdNEHETs87)
 [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/langbot-app/LangBot)
@@ -21,12 +21,7 @@
 
 </p>
 
-## ✨ 機能
-
-- 💬 LLM / エージェントとのチャット: 複数のLLMをサポートし、グループチャットとプライベートチャットに対応。マルチラウンドの会話、ツールの呼び出し、マルチモーダル機能をサポート。 [Dify](https://dify.ai) と深く統合。現在、QQ、QQ チャンネル、WeChat、個人 WeChat、Lark、DingTalk、Discord、Telegram など、複数のプラットフォームをサポートしています。
-- 🛠️ 高い安定性、豊富な機能: ネイティブのアクセス制御、レート制限、敏感な単語のフィルタリングなどのメカニズムをサポート。使いやすく、複数のデプロイ方法をサポート。複数のパイプライン設定をサポートし、異なるボットを異なる用途に使用できます。
-- 🧩 プラグイン拡張、活発なコミュニティ: イベント駆動、コンポーネント拡張などのプラグインメカニズムをサポート。適配 Anthropic [MCP プロトコル](https://modelcontextprotocol.io/)；豊富なエコシステム、現在数百のプラグインが存在。
-- 😻 Web UI: ブラウザを通じてLangBotインスタンスを管理することをサポート。
+LangBot は、エージェント、RAG、MCP などの LLM アプリケーション機能を備えた、オープンソースの LLM ネイティブのインスタントメッセージングロボット開発プラットフォームです。世界中のインスタントメッセージングプラットフォームに適応し、豊富な API インターフェースを提供し、カスタム開発をサポートします。
 
 ## 📦 始め方
 
@@ -42,7 +37,7 @@ http://localhost:5300 にアクセスして使用を開始します。
 
 詳細なドキュメントは[Dockerデプロイ](https://docs.langbot.app/en/deploy/langbot/docker.html)を参照してください。
 
-#### BTPanelでのワンクリックデプロイ
+#### Panelでのワンクリックデプロイ
 
 LangBotはBTPanelにリストされています。BTPanelをインストールしている場合は、[ドキュメント](https://docs.langbot.app/en/deploy/langbot/one-click/bt.html)を使用して使用できます。
 
@@ -60,23 +55,25 @@ LangBotはBTPanelにリストされています。BTPanelをインストール
 
 リリースバージョンを直接使用して実行します。[手動デプロイ](https://docs.langbot.app/en/deploy/langbot/manual.html)のドキュメントを参照してください。
 
-## 📸 デモ
+## 😎 最新情報を入手
 
-<img alt="bots" src="https://docs.langbot.app/webui/bot-page.png" width="400px"/>
+リポジトリの右上にある Star と Watch ボタンをクリックして、最新の更新を取得してください。
 
-<img alt="bots" src="https://docs.langbot.app/webui/create-model.png" width="400px"/>
+![star gif](https://docs.langbot.app/star.gif)
 
-<img alt="bots" src="https://docs.langbot.app/webui/edit-pipeline.png" width="400px"/>
+## ✨ 機能
 
-<img alt="bots" src="https://docs.langbot.app/webui/plugin-market.png" width="400px"/>
+- 💬 LLM / エージェントとのチャット: 複数のLLMをサポートし、グループチャットとプライベートチャットに対応。マルチラウンドの会話、ツールの呼び出し、マルチモーダル、ストリーミング出力機能をサポート、RAG（知識ベース）を組み込み、[Dify](https://dify.ai) と深く統合。
+- 🤖 多プラットフォーム対応: 現在、QQ、QQ チャンネル、WeChat、個人 WeChat、Lark、DingTalk、Discord、Telegram など、複数のプラットフォームをサポートしています。
+- 🛠️ 高い安定性、豊富な機能: ネイティブのアクセス制御、レート制限、敏感な単語のフィルタリングなどのメカニズムをサポート。使いやすく、複数のデプロイ方法をサポート。複数のパイプライン設定をサポートし、異なるボットを異なる用途に使用できます。
+- 🧩 プラグイン拡張、活発なコミュニティ: イベント駆動、コンポーネント拡張などのプラグインメカニズムをサポート。適配 Anthropic [MCP プロトコル](https://modelcontextprotocol.io/)；豊富なエコシステム、現在数百のプラグインが存在。
+- 😻 Web UI: ブラウザを通じてLangBotインスタンスを管理することをサポート。
 
-<img alt="返信効果（インターネットプラグイン付き）" src="https://docs.langbot.app/QChatGPT-0516.png" width="500px"/>
+詳細な仕様については、[ドキュメント](https://docs.langbot.app/en/insight/features.html)を参照してください。
 
-- WebUIデモ: https://demo.langbot.dev/
-    - ログイン情報: メール: `demo@langbot.app` パスワード: `langbot123456`
-    - 注意: WebUIの効果のみを示しています。公開環境では、機密情報を入力しないでください。
-
-## 🔌 コンポーネントの互換性
+または、デモ環境にアクセスしてください: https://demo.langbot.dev/
+  - ログイン情報: メール: `demo@langbot.app` パスワード: `langbot123456`
+  - 注意: WebUI のデモンストレーションのみの場合、公開環境では機密情報を入力しないでください。
 
 ### メッセージプラットフォーム
 
@@ -92,10 +89,6 @@ LangBotはBTPanelにリストされています。BTPanelをインストール
 | Discord | ✅ |  |
 | Telegram | ✅ |  |
 | Slack | ✅ |  |
-| LINE | 🚧 |  |
-| WhatsApp | 🚧 |  |
-
-🚧: 開発中
 
 ### LLMs
 
@@ -109,6 +102,7 @@ LangBotはBTPanelにリストされています。BTPanelをインストール
 | [Zhipu AI](https://open.bigmodel.cn/) | ✅ |  |
 | [CompShare](https://www.compshare.cn/?ytag=GPU_YY-gh_langbot) | ✅ | 大模型とGPUリソースプラットフォーム |
 | [PPIO](https://ppinfra.com/user/register?invited_by=QJKFYD&utm_source=github_langbot) | ✅ | 大模型とGPUリソースプラットフォーム |
+| [ShengSuanYun](https://www.shengsuanyun.com/?from=CH_KYIPP758) | ✅ | LLMとGPUリソースプラットフォーム |
 | [302.AI](https://share.302.ai/SuTG99) | ✅ | LLMゲートウェイ(MaaS) |
 | [Google Gemini](https://aistudio.google.com/prompts/new_chat) | ✅ | |
 | [Dify](https://dify.ai) | ✅ | LLMOpsプラットフォーム |
@@ -128,9 +122,3 @@ LangBot への貢献に対して、以下の [コード貢献者](https://github
 <a href="https://github.com/langbot-app/LangBot/graphs/contributors">
   <img src="https://contrib.rocks/image?repo=langbot-app/LangBot" />
 </a>
-
-## 😎 最新情報を入手
-
-リポジトリの右上にある Star と Watch ボタンをクリックして、最新の更新を取得してください。
-
-![star gif](https://docs.langbot.app/star.gif)

README_TW.md

@@ -0,0 +1,140 @@
+<p align="center">
+<a href="https://langbot.app">
+<img src="https://docs.langbot.app/social_zh.png" alt="LangBot"/>
+</a>
+
+<div align="center"><a href="https://hellogithub.com/repository/langbot-app/LangBot" target="_blank"><img src="https://abroad.hellogithub.com/v1/widgets/recommend.svg?rid=5ce8ae2aa4f74316bf393b57b952433c&claim_uid=gtmc6YWjMZkT21R" alt="Featured｜HelloGitHub" style="width: 250px; height: 54px;" width="250" height="54" /></a>
+
+[English](README_EN.md) / [简体中文](README.md) / 繁體中文 / [日本語](README_JP.md) / (PR for your language)
+
+[![Discord](https://img.shields.io/discord/1335141740050649118?logo=discord&labelColor=%20%235462eb&logoColor=%20%23f5f5f5&color=%20%235462eb)](https://discord.gg/wdNEHETs87)
+[![QQ Group](https://img.shields.io/badge/%E7%A4%BE%E5%8C%BAQQ%E7%BE%A4-966235608-blue)](https://qm.qq.com/q/JLi38whHum)
+[![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/langbot-app/LangBot)
+[![GitHub release (latest by date)](https://img.shields.io/github/v/release/langbot-app/LangBot)](https://github.com/langbot-app/LangBot/releases/latest)
+<img src="https://img.shields.io/badge/python-3.10 ~ 3.13 -blue.svg" alt="python">
+[![star](https://gitcode.com/RockChinQ/LangBot/star/badge.svg)](https://gitcode.com/RockChinQ/LangBot)
+
+<a href="https://langbot.app">主頁</a> ｜
+<a href="https://docs.langbot.app/zh/insight/guide.html">部署文件</a> ｜
+<a href="https://docs.langbot.app/zh/plugin/plugin-intro.html">外掛介紹</a> ｜
+<a href="https://github.com/langbot-app/LangBot/issues/new?assignees=&labels=%E7%8B%AC%E7%AB%8B%E6%8F%92%E4%BB%B6&projects=&template=submit-plugin.yml&title=%5BPlugin%5D%3A+%E8%AF%B7%E6%B1%82%E7%99%BB%E8%AE%B0%E6%96%B0%E6%8F%92%E4%BB%B6">提交外掛</a>
+
+</div>
+
+</p>
+
+LangBot 是一個開源的大語言模型原生即時通訊機器人開發平台，旨在提供開箱即用的 IM 機器人開發體驗，具有 Agent、RAG、MCP 等多種 LLM 應用功能，適配全球主流即時通訊平台，並提供豐富的 API 介面，支援自定義開發。
+
+## 📦 開始使用
+
+#### Docker Compose 部署
+
+```bash
+git clone https://github.com/langbot-app/LangBot
+cd LangBot
+docker compose up -d
+```
+
+訪問 http://localhost:5300 即可開始使用。
+
+詳細文件[Docker 部署](https://docs.langbot.app/zh/deploy/langbot/docker.html)。
+
+#### 寶塔面板部署
+
+已上架寶塔面板，若您已安裝寶塔面板，可以根據[文件](https://docs.langbot.app/zh/deploy/langbot/one-click/bt.html)使用。
+
+#### Zeabur 雲端部署
+
+社群貢獻的 Zeabur 模板。
+
+[![Deploy on Zeabur](https://zeabur.com/button.svg)](https://zeabur.com/zh-CN/templates/ZKTBDH)
+
+#### Railway 雲端部署
+
+[![Deploy on Railway](https://railway.com/button.svg)](https://railway.app/template/yRrAyL?referralCode=vogKPF)
+
+#### 手動部署
+
+直接使用發行版運行，查看文件[手動部署](https://docs.langbot.app/zh/deploy/langbot/manual.html)。
+
+## 😎 保持更新
+
+點擊倉庫右上角 Star 和 Watch 按鈕，獲取最新動態。
+
+![star gif](https://docs.langbot.app/star.gif)
+
+## ✨ 特性
+
+- 💬 大模型對話、Agent：支援多種大模型，適配群聊和私聊；具有多輪對話、工具調用、多模態、流式輸出能力，自帶 RAG（知識庫）實現，並深度適配 [Dify](https://dify.ai)。
+- 🤖 多平台支援：目前支援 QQ、QQ頻道、企業微信、個人微信、飛書、Discord、Telegram 等平台。
+- 🛠️ 高穩定性、功能完備：原生支援訪問控制、限速、敏感詞過濾等機制；配置簡單，支援多種部署方式。支援多流水線配置，不同機器人用於不同應用場景。
+- 🧩 外掛擴展、活躍社群：支援事件驅動、組件擴展等外掛機制；適配 Anthropic [MCP 協議](https://modelcontextprotocol.io/)；目前已有數百個外掛。
+- 😻 Web 管理面板：支援通過瀏覽器管理 LangBot 實例，不再需要手動編寫配置文件。
+
+詳細規格特性請訪問[文件](https://docs.langbot.app/zh/insight/features.html)。
+
+或訪問 demo 環境：https://demo.langbot.dev/  
+  - 登入資訊：郵箱：`demo@langbot.app` 密碼：`langbot123456`
+  - 注意：僅展示 WebUI 效果，公開環境，請不要在其中填入您的任何敏感資訊。
+
+### 訊息平台
+
+| 平台 | 狀態 | 備註 |
+| --- | --- | --- |
+| QQ 個人號 | ✅ | QQ 個人號私聊、群聊 |
+| QQ 官方機器人 | ✅ | QQ 官方機器人，支援頻道、私聊、群聊 |
+| 微信 | ✅ |  |
+| 企微對外客服 | ✅ |  |
+| 微信公眾號 | ✅ |  |
+| Lark | ✅ |  |
+| DingTalk | ✅ |  |
+| Discord | ✅ |  |
+| Telegram | ✅ |  |
+| Slack | ✅ |  |
+
+### 大模型能力
+
+| 模型 | 狀態 | 備註 |
+| --- | --- | --- |
+| [OpenAI](https://platform.openai.com/) | ✅ | 可接入任何 OpenAI 介面格式模型 |
+| [DeepSeek](https://www.deepseek.com/) | ✅ |  |
+| [Moonshot](https://www.moonshot.cn/) | ✅ |  |
+| [Anthropic](https://www.anthropic.com/) | ✅ |  |
+| [xAI](https://x.ai/) | ✅ |  |
+| [智譜AI](https://open.bigmodel.cn/) | ✅ |  |
+| [勝算雲](https://www.shengsuanyun.com/?from=CH_KYIPP758) | ✅ | 大模型和 GPU 資源平台 |
+| [優雲智算](https://www.compshare.cn/?ytag=GPU_YY-gh_langbot) | ✅ | 大模型和 GPU 資源平台 |
+| [PPIO](https://ppinfra.com/user/register?invited_by=QJKFYD&utm_source=github_langbot) | ✅ | 大模型和 GPU 資源平台 |
+| [302.AI](https://share.302.ai/SuTG99) | ✅ | 大模型聚合平台 |
+| [Google Gemini](https://aistudio.google.com/prompts/new_chat) | ✅ | |
+| [Dify](https://dify.ai) | ✅ | LLMOps 平台 |
+| [Ollama](https://ollama.com/) | ✅ | 本地大模型運行平台 |
+| [LMStudio](https://lmstudio.ai/) | ✅ | 本地大模型運行平台 |
+| [GiteeAI](https://ai.gitee.com/) | ✅ | 大模型介面聚合平台 |
+| [SiliconFlow](https://siliconflow.cn/) | ✅ | 大模型聚合平台 |
+| [阿里雲百煉](https://bailian.console.aliyun.com/) | ✅ | 大模型聚合平台, LLMOps 平台 |
+| [火山方舟](https://console.volcengine.com/ark/region:ark+cn-beijing/model?vendor=Bytedance&view=LIST_VIEW) | ✅ | 大模型聚合平台, LLMOps 平台 |
+| [ModelScope](https://modelscope.cn/docs/model-service/API-Inference/intro) | ✅ | 大模型聚合平台 |
+| [MCP](https://modelcontextprotocol.io/) | ✅ | 支援通過 MCP 協議獲取工具 |
+
+### TTS
+
+| 平台/模型 | 備註 |
+| --- | --- |
+| [FishAudio](https://fish.audio/zh-CN/discovery/) | [外掛](https://github.com/the-lazy-me/NewChatVoice) |
+| [海豚 AI](https://www.ttson.cn/?source=thelazy) | [外掛](https://github.com/the-lazy-me/NewChatVoice) |
+| [AzureTTS](https://portal.azure.com/) | [外掛](https://github.com/Ingnaryk/LangBot_AzureTTS) |
+
+### 文生圖
+
+| 平台/模型 | 備註 |
+| --- | --- |
+| 阿里雲百煉 | [外掛](https://github.com/Thetail001/LangBot_BailianTextToImagePlugin)
+
+## 😘 社群貢獻
+
+感謝以下[程式碼貢獻者](https://github.com/langbot-app/LangBot/graphs/contributors)和社群裡其他成員對 LangBot 的貢獻：
+
+<a href="https://github.com/langbot-app/LangBot/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=langbot-app/LangBot" />
+</a> 

libs/dingtalk_api/api.py

@@ -253,6 +253,43 @@ class DingTalkClient:
             await self.logger.error(f'failed to send proactive massage to group: {traceback.format_exc()}')
             raise Exception(f'failed to send proactive massage to group: {traceback.format_exc()}')
 
+    async def create_and_card(
+        self, temp_card_id: str, incoming_message: dingtalk_stream.ChatbotMessage, quote_origin: bool = False
+    ):
+        content_key = 'content'
+        card_data = {content_key: ''}
+
+        card_instance = dingtalk_stream.AICardReplier(self.client, incoming_message)
+        # print(card_instance)
+        # 先投放卡片: https://open.dingtalk.com/document/orgapp/create-and-deliver-cards
+        card_instance_id = await card_instance.async_create_and_deliver_card(
+            temp_card_id,
+            card_data,
+        )
+        return card_instance, card_instance_id
+
+    async def send_card_message(self, card_instance, card_instance_id: str, content: str, is_final: bool):
+        content_key = 'content'
+        try:
+            await card_instance.async_streaming(
+                card_instance_id,
+                content_key=content_key,
+                content_value=content,
+                append=False,
+                finished=is_final,
+                failed=False,
+            )
+        except Exception as e:
+            self.logger.exception(e)
+            await card_instance.async_streaming(
+                card_instance_id,
+                content_key=content_key,
+                content_value='',
+                append=False,
+                finished=is_final,
+                failed=True,
+            )
+
     async def start(self):
         """启动 WebSocket 连接，监听消息"""
         await self.client.start()

libs/qq_official_api/api.py

@@ -104,7 +104,7 @@ class QQOfficialClient:
             return {'code': 0, 'message': 'success'}
 
         except Exception as e:
-            await self.logger.error(f"Error in handle_callback_request: {traceback.format_exc()}")
+            await self.logger.error(f'Error in handle_callback_request: {traceback.format_exc()}')
             return {'error': str(e)}, 400
 
     async def run_task(self, host: str, port: int, *args, **kwargs):
@@ -168,7 +168,6 @@ class QQOfficialClient:
         if not await self.check_access_token():
             await self.get_access_token()
 
-
         url = self.base_url + '/v2/users/' + user_openid + '/messages'
         async with httpx.AsyncClient() as client:
             headers = {
@@ -193,7 +192,6 @@ class QQOfficialClient:
         if not await self.check_access_token():
             await self.get_access_token()
 
-
         url = self.base_url + '/v2/groups/' + group_openid + '/messages'
         async with httpx.AsyncClient() as client:
             headers = {
@@ -209,7 +207,7 @@ class QQOfficialClient:
             if response.status_code == 200:
                 return
             else:
-                await self.logger.error(f"发送群聊消息失败:{response.json()}")
+                await self.logger.error(f'发送群聊消息失败:{response.json()}')
                 raise Exception(response.read().decode())
 
     async def send_channle_group_text_msg(self, channel_id: str, content: str, msg_id: str):
@@ -217,7 +215,6 @@ class QQOfficialClient:
         if not await self.check_access_token():
             await self.get_access_token()
 
-
         url = self.base_url + '/channels/' + channel_id + '/messages'
         async with httpx.AsyncClient() as client:
             headers = {
@@ -240,7 +237,6 @@ class QQOfficialClient:
         """发送频道私聊消息"""
         if not await self.check_access_token():
             await self.get_access_token()
-        
 
         url = self.base_url + '/dms/' + guild_id + '/messages'
         async with httpx.AsyncClient() as client:

libs/slack_api/api.py

@@ -34,7 +34,6 @@ class SlackClient:
 
             if self.bot_user_id and bot_user_id == self.bot_user_id:
                 return jsonify({'status': 'ok'})
-            
 
             # 处理私信
             if data and data.get('event', {}).get('channel_type') in ['im']:
@@ -52,7 +51,7 @@ class SlackClient:
             return jsonify({'status': 'ok'})
 
         except Exception as e:
-            await self.logger.error(f"Error in handle_callback_request: {traceback.format_exc()}")
+            await self.logger.error(f'Error in handle_callback_request: {traceback.format_exc()}')
             raise (e)
 
     async def _handle_message(self, event: SlackEvent):
@@ -82,7 +81,7 @@ class SlackClient:
                 self.bot_user_id = response['message']['bot_id']
             return
         except Exception as e:
-            await self.logger.error(f"Error in send_message: {e}")
+            await self.logger.error(f'Error in send_message: {e}')
             raise e
 
     async def send_message_to_one(self, text: str, user_id: str):
@@ -93,7 +92,7 @@ class SlackClient:
 
             return
         except Exception as e:
-            await self.logger.error(f"Error in send_message: {traceback.format_exc()}")
+            await self.logger.error(f'Error in send_message: {traceback.format_exc()}')
             raise e
 
     async def run_task(self, host: str, port: int, *args, **kwargs):

libs/wechatpad_api/__init__.py

@@ -1 +1 @@
-from .client import WeChatPadClient
+from .client import WeChatPadClient as WeChatPadClient

libs/wechatpad_api/api/chatroom.py

@@ -1,4 +1,4 @@
-from libs.wechatpad_api.util.http_util import async_request, post_json
+from libs.wechatpad_api.util.http_util import post_json
 
 
 class ChatRoomApi:
@@ -7,8 +7,6 @@ class ChatRoomApi:
         self.token = token
 
     def get_chatroom_member_detail(self, chatroom_name):
-        params = {
-            "ChatRoomName": chatroom_name
-        }
+        params = {'ChatRoomName': chatroom_name}
         url = self.base_url + '/group/GetChatroomMemberDetail'
         return post_json(url, token=self.token, data=params)

libs/wechatpad_api/api/downloadpai.py

@@ -1,32 +1,23 @@
-from libs.wechatpad_api.util.http_util import async_request, post_json
+from libs.wechatpad_api.util.http_util import post_json
 import httpx
 import base64
 
+
 class DownloadApi:
     def __init__(self, base_url, token):
         self.base_url = base_url
         self.token = token
 
     def send_download(self, aeskey, file_type, file_url):
-        json_data = {
-                  "AesKey": aeskey,
-                  "FileType": file_type,
-                  "FileURL": file_url
-                }
-        url = self.base_url + "/message/SendCdnDownload"
+        json_data = {'AesKey': aeskey, 'FileType': file_type, 'FileURL': file_url}
+        url = self.base_url + '/message/SendCdnDownload'
         return post_json(url, token=self.token, data=json_data)
 
-    def get_msg_voice(self,buf_id, length, new_msgid):
-        json_data = {
-            "Bufid": buf_id,
-            "Length": length,
-            "NewMsgId": new_msgid,
-            "ToUserName": ""
-        }
-        url = self.base_url + "/message/GetMsgVoice"
+    def get_msg_voice(self, buf_id, length, new_msgid):
+        json_data = {'Bufid': buf_id, 'Length': length, 'NewMsgId': new_msgid, 'ToUserName': ''}
+        url = self.base_url + '/message/GetMsgVoice'
         return post_json(url, token=self.token, data=json_data)
 
-
     async def download_url_to_base64(self, download_url):
         async with httpx.AsyncClient() as client:
             response = await client.get(download_url)
@@ -36,4 +27,4 @@ class DownloadApi:
                 base64_str = base64.b64encode(file_bytes).decode('utf-8')  # 返回字符串格式
                 return base64_str
             else:
-                raise Exception('获取文件失败')
+                raise Exception('获取文件失败')

libs/wechatpad_api/api/friend.py

@@ -1,11 +1,6 @@
-from libs.wechatpad_api.util.http_util import post_json,async_request
-from typing import List, Dict, Any, Optional
-
-
 class FriendApi:
     """联系人API类，处理所有与联系人相关的操作"""
 
     def __init__(self, base_url: str, token: str):
         self.base_url = base_url
         self.token = token
-

libs/wechatpad_api/api/login.py

@@ -1,37 +1,34 @@
-from libs.wechatpad_api.util.http_util import async_request,post_json,get_json
+from libs.wechatpad_api.util.http_util import post_json, get_json
 
 
 class LoginApi:
     def __init__(self, base_url: str, token: str = None, admin_key: str = None):
-        '''
+        """
 
         Args:
             base_url: 原始路径
             token: token
             admin_key: 管理员key
-        '''
+        """
         self.base_url = base_url
         self.token = token
         # self.admin_key = admin_key
 
-    def get_token(self, admin_key, day: int=365):
+    def get_token(self, admin_key, day: int = 365):
         # 获取普通token
-        url = f"{self.base_url}/admin/GenAuthKey1"
-        json_data = {
-                "Count": 1,
-                "Days": day
-            }
+        url = f'{self.base_url}/admin/GenAuthKey1'
+        json_data = {'Count': 1, 'Days': day}
         return post_json(base_url=url, token=admin_key, data=json_data)
 
-    def get_login_qr(self, Proxy: str = ""):
-        '''
+    def get_login_qr(self, Proxy: str = ''):
+        """
 
         Args:
             Proxy:异地使用时代理
 
         Returns:json数据
 
-        '''
+        """
         """
         
         {
@@ -49,54 +46,37 @@ class LoginApi:
 }
         
         """
-        #获取登录二维码
-        url = f"{self.base_url}/login/GetLoginQrCodeNew"
+        # 获取登录二维码
+        url = f'{self.base_url}/login/GetLoginQrCodeNew'
         check = False
-        if Proxy != "":
+        if Proxy != '':
             check = True
-        json_data = {
-              "Check": check,
-              "Proxy": Proxy
-            }
+        json_data = {'Check': check, 'Proxy': Proxy}
         return post_json(base_url=url, token=self.token, data=json_data)
 
-
     def get_login_status(self):
         # 获取登录状态
         url = f'{self.base_url}/login/GetLoginStatus'
         return get_json(base_url=url, token=self.token)
 
-
-
     def logout(self):
         # 退出登录
         url = f'{self.base_url}/login/LogOut'
         return post_json(base_url=url, token=self.token)
 
-
-
-
-    def wake_up_login(self, Proxy: str = ""):
+    def wake_up_login(self, Proxy: str = ''):
         # 唤醒登录
         url = f'{self.base_url}/login/WakeUpLogin'
         check = False
-        if Proxy != "":
+        if Proxy != '':
             check = True
-        json_data = {
-            "Check": check,
-            "Proxy": ""
-        }
+        json_data = {'Check': check, 'Proxy': ''}
 
         return post_json(base_url=url, token=self.token, data=json_data)
 
-
-
-    def login(self,admin_key):
+    def login(self, admin_key):
         login_status = self.get_login_status()
-        if login_status["Code"] == 300 and login_status["Text"] == "你已退出微信":
-            print("token已经失效，重新获取")
+        if login_status['Code'] == 300 and login_status['Text'] == '你已退出微信':
+            print('token已经失效，重新获取')
             token_data = self.get_token(admin_key)
-            self.token = token_data["Data"][0]
-
-
-
+            self.token = token_data['Data'][0]

libs/wechatpad_api/api/message.py

@@ -1,5 +1,4 @@
-
-from libs.wechatpad_api.util.http_util import async_request, post_json
+from libs.wechatpad_api.util.http_util import post_json
 
 
 class MessageApi:
@@ -7,8 +6,8 @@ class MessageApi:
         self.base_url = base_url
         self.token = token
 
-    def post_text(self, to_wxid, content, ats: list= []):
-        '''
+    def post_text(self, to_wxid, content, ats: list = []):
+        """
 
         Args:
             app_id: 微信id
@@ -18,106 +17,64 @@ class MessageApi:
 
         Returns:
 
-        '''
-        url = self.base_url + "/message/SendTextMessage"
+        """
+        url = self.base_url + '/message/SendTextMessage'
         """发送文字消息"""
         json_data = {
-                      "MsgItem": [
-                        {
-                          "AtWxIDList": ats,
-                          "ImageContent": "",
-                          "MsgType": 0,
-                          "TextContent": content,
-                          "ToUserName": to_wxid
-                        }
-                      ]
-                    }
-        return  post_json(base_url=url, token=self.token, data=json_data)
+            'MsgItem': [
+                {'AtWxIDList': ats, 'ImageContent': '', 'MsgType': 0, 'TextContent': content, 'ToUserName': to_wxid}
+            ]
+        }
+        return post_json(base_url=url, token=self.token, data=json_data)
 
-
-
-
-    def post_image(self, to_wxid, img_url, ats: list= []):
+    def post_image(self, to_wxid, img_url, ats: list = []):
         """发送图片消息"""
         # 这里好像可以尝试发送多个暂时未测试
         json_data = {
-            "MsgItem": [
-                {
-                    "AtWxIDList": ats,
-                    "ImageContent": img_url,
-                    "MsgType": 0,
-                    "TextContent": '',
-                    "ToUserName": to_wxid
-                }
+            'MsgItem': [
+                {'AtWxIDList': ats, 'ImageContent': img_url, 'MsgType': 0, 'TextContent': '', 'ToUserName': to_wxid}
             ]
         }
-        url = self.base_url + "/message/SendImageMessage"
+        url = self.base_url + '/message/SendImageMessage'
         return post_json(base_url=url, token=self.token, data=json_data)
 
     def post_voice(self, to_wxid, voice_data, voice_forma, voice_duration):
         """发送语音消息"""
         json_data = {
-            "ToUserName": to_wxid,
-            "VoiceData": voice_data,
-            "VoiceFormat": voice_forma,
-            "VoiceSecond": voice_duration
+            'ToUserName': to_wxid,
+            'VoiceData': voice_data,
+            'VoiceFormat': voice_forma,
+            'VoiceSecond': voice_duration,
         }
-        url = self.base_url + "/message/SendVoice"
+        url = self.base_url + '/message/SendVoice'
         return post_json(base_url=url, token=self.token, data=json_data)
 
-
-
-
-
     def post_name_card(self, alias, to_wxid, nick_name, name_card_wxid, flag):
         """发送名片消息"""
         param = {
-                "CardAlias": alias,
-                "CardFlag": flag,
-                "CardNickName": nick_name,
-                "CardWxId": name_card_wxid,
-                "ToUserName": to_wxid
+            'CardAlias': alias,
+            'CardFlag': flag,
+            'CardNickName': nick_name,
+            'CardWxId': name_card_wxid,
+            'ToUserName': to_wxid,
         }
-        url = f"{self.base_url}/message/ShareCardMessage"
+        url = f'{self.base_url}/message/ShareCardMessage'
         return post_json(base_url=url, token=self.token, data=param)
 
-    def post_emoji(self, to_wxid, emoji_md5, emoji_size:int=0):
+    def post_emoji(self, to_wxid, emoji_md5, emoji_size: int = 0):
         """发送emoji消息"""
-        json_data = {
-            "EmojiList": [
-                {
-                    "EmojiMd5": emoji_md5,
-                    "EmojiSize": emoji_size,
-                    "ToUserName": to_wxid
-                }
-            ]
-        }
-        url = f"{self.base_url}/message/SendEmojiMessage"
+        json_data = {'EmojiList': [{'EmojiMd5': emoji_md5, 'EmojiSize': emoji_size, 'ToUserName': to_wxid}]}
+        url = f'{self.base_url}/message/SendEmojiMessage'
         return post_json(base_url=url, token=self.token, data=json_data)
 
-    def post_app_msg(self, to_wxid,xml_data, contenttype:int=0):
+    def post_app_msg(self, to_wxid, xml_data, contenttype: int = 0):
         """发送appmsg消息"""
-        json_data = {
-            "AppList": [
-                {
-                    "ContentType": contenttype,
-                    "ContentXML": xml_data,
-                    "ToUserName": to_wxid
-                }
-            ]
-        }
-        url = f"{self.base_url}/message/SendAppMessage"
+        json_data = {'AppList': [{'ContentType': contenttype, 'ContentXML': xml_data, 'ToUserName': to_wxid}]}
+        url = f'{self.base_url}/message/SendAppMessage'
         return post_json(base_url=url, token=self.token, data=json_data)
 
-
-
     def revoke_msg(self, to_wxid, msg_id, new_msg_id, create_time):
         """撤回消息"""
-        param = {
-              "ClientMsgId": msg_id,
-              "CreateTime": create_time,
-              "NewMsgId": new_msg_id,
-              "ToUserName": to_wxid
-            }
-        url = f"{self.base_url}/message/RevokeMsg"
-        return post_json(base_url=url, token=self.token, data=param)
+        param = {'ClientMsgId': msg_id, 'CreateTime': create_time, 'NewMsgId': new_msg_id, 'ToUserName': to_wxid}
+        url = f'{self.base_url}/message/RevokeMsg'
+        return post_json(base_url=url, token=self.token, data=param)

libs/wechatpad_api/api/user.py

@@ -12,12 +12,9 @@ class UserApi:
 
         return get_json(base_url=url, token=self.token)
 
-    def get_qr_code(self, recover:bool=True, style:int=8):
+    def get_qr_code(self, recover: bool = True, style: int = 8):
         """获取自己的二维码"""
-        param = {
-              "Recover": recover,
-              "Style": style
-            }
+        param = {'Recover': recover, 'Style': style}
         url = f'{self.base_url}/user/GetMyQRCode'
         return post_json(base_url=url, token=self.token, data=param)
 
@@ -26,12 +23,8 @@ class UserApi:
         url = f'{self.base_url}/equipment/GetSafetyInfo'
         return post_json(base_url=url, token=self.token)
 
-
-
-    async def update_head_img(self,  head_img_base64):
+    async def update_head_img(self, head_img_base64):
         """修改头像"""
-        param = {
-              "Base64": head_img_base64
-            }
+        param = {'Base64': head_img_base64}
         url = f'{self.base_url}/user/UploadHeadImage'
-        return await async_request(base_url=url, token_key=self.token, json=param)
+        return await async_request(base_url=url, token_key=self.token, json=param)

libs/wechatpad_api/client.py

@@ -1,4 +1,3 @@
-
 from libs.wechatpad_api.api.login import LoginApi
 from libs.wechatpad_api.api.friend import FriendApi
 from libs.wechatpad_api.api.message import MessageApi
@@ -7,9 +6,6 @@ from libs.wechatpad_api.api.downloadpai import DownloadApi
 from libs.wechatpad_api.api.chatroom import ChatRoomApi
 
 
-
-
-
 class WeChatPadClient:
     def __init__(self, base_url, token, logger=None):
         self._login_api = LoginApi(base_url, token)
@@ -20,16 +16,16 @@ class WeChatPadClient:
         self._chatroom_api = ChatRoomApi(base_url, token)
         self.logger = logger
 
-    def get_token(self,admin_key, day: int):
-        '''获取token'''
+    def get_token(self, admin_key, day: int):
+        """获取token"""
         return self._login_api.get_token(admin_key, day)
 
-    def get_login_qr(self, Proxy:str=""):
+    def get_login_qr(self, Proxy: str = ''):
         """登录二维码"""
         return self._login_api.get_login_qr(Proxy=Proxy)
 
-    def awaken_login(self, Proxy:str=""):
-        '''唤醒登录'''
+    def awaken_login(self, Proxy: str = ''):
+        """唤醒登录"""
         return self._login_api.wake_up_login(Proxy=Proxy)
 
     def log_out(self):
@@ -40,59 +36,57 @@ class WeChatPadClient:
         """获取登录状态"""
         return self._login_api.get_login_status()
 
-    def send_text_message(self, to_wxid, message, ats: list=[]):
+    def send_text_message(self, to_wxid, message, ats: list = []):
         """发送文本消息"""
-        return  self._message_api.post_text(to_wxid, message, ats)
+        return self._message_api.post_text(to_wxid, message, ats)
 
-    def send_image_message(self, to_wxid, img_url, ats: list=[]):
+    def send_image_message(self, to_wxid, img_url, ats: list = []):
         """发送图片消息"""
-        return  self._message_api.post_image(to_wxid, img_url, ats)
+        return self._message_api.post_image(to_wxid, img_url, ats)
 
     def send_voice_message(self, to_wxid, voice_data, voice_forma, voice_duration):
         """发送音频消息"""
-        return  self._message_api.post_voice(to_wxid, voice_data, voice_forma, voice_duration)
+        return self._message_api.post_voice(to_wxid, voice_data, voice_forma, voice_duration)
 
     def send_app_message(self, to_wxid, app_message, type):
         """发送app消息"""
-        return  self._message_api.post_app_msg(to_wxid, app_message, type)
+        return self._message_api.post_app_msg(to_wxid, app_message, type)
 
     def send_emoji_message(self, to_wxid, emoji_md5, emoji_size):
         """发送emoji消息"""
-        return  self._message_api.post_emoji(to_wxid,emoji_md5,emoji_size)
+        return self._message_api.post_emoji(to_wxid, emoji_md5, emoji_size)
 
     def revoke_msg(self, to_wxid, msg_id, new_msg_id, create_time):
         """撤回消息"""
-        return  self._message_api.revoke_msg(to_wxid, msg_id, new_msg_id, create_time)
+        return self._message_api.revoke_msg(to_wxid, msg_id, new_msg_id, create_time)
 
     def get_profile(self):
         """获取用户信息"""
         return self._user_api.get_profile()
 
-    def get_qr_code(self, recover:bool=True, style:int=8):
+    def get_qr_code(self, recover: bool = True, style: int = 8):
         """获取用户二维码"""
-        return  self._user_api.get_qr_code(recover=recover, style=style)
+        return self._user_api.get_qr_code(recover=recover, style=style)
 
     def get_safety_info(self):
         """获取设备信息"""
-        return  self._user_api.get_safety_info()
+        return self._user_api.get_safety_info()
 
-    def update_head_img(self,  head_img_base64):
+    def update_head_img(self, head_img_base64):
         """上传用户头像"""
-        return  self._user_api.update_head_img(head_img_base64)
+        return self._user_api.update_head_img(head_img_base64)
 
     def cdn_download(self, aeskey, file_type, file_url):
         """cdn下载"""
-        return self._download_api.send_download( aeskey, file_type, file_url)
+        return self._download_api.send_download(aeskey, file_type, file_url)
 
-    def get_msg_voice(self,buf_id, length, msgid):
+    def get_msg_voice(self, buf_id, length, msgid):
         """下载语音"""
         return self._download_api.get_msg_voice(buf_id, length, msgid)
 
-    async def download_base64(self,url):
+    async def download_base64(self, url):
         return await self._download_api.download_url_to_base64(download_url=url)
 
     def get_chatroom_member_detail(self, chatroom_name):
         """查看群成员详情"""
         return self._chatroom_api.get_chatroom_member_detail(chatroom_name)
-
-

libs/wechatpad_api/util/http_util.py

@@ -1,10 +1,9 @@
 import requests
+import aiohttp
+
 
 def post_json(base_url, token, data=None):
-    headers = {
-        'Content-Type': 'application/json'
-    }
-
+    headers = {'Content-Type': 'application/json'}
 
     url = base_url + f'?key={token}'
 
@@ -18,14 +17,12 @@ def post_json(base_url, token, data=None):
         else:
             raise RuntimeError(response.text)
     except Exception as e:
-        print(f"http请求失败, url={url}, exception={e}")
+        print(f'http请求失败, url={url}, exception={e}')
         raise RuntimeError(str(e))
 
-def get_json(base_url, token):
-    headers = {
-        'Content-Type': 'application/json'
-    }
 
+def get_json(base_url, token):
+    headers = {'Content-Type': 'application/json'}
 
     url = base_url + f'?key={token}'
 
@@ -39,21 +36,18 @@ def get_json(base_url, token):
         else:
             raise RuntimeError(response.text)
     except Exception as e:
-        print(f"http请求失败, url={url}, exception={e}")
+        print(f'http请求失败, url={url}, exception={e}')
         raise RuntimeError(str(e))
 
-import aiohttp
-import asyncio
-
 
 async def async_request(
-        base_url: str,
-        token_key: str,
-        method: str = 'POST',
-        params: dict = None,
-        # headers: dict = None,
-        data: dict = None,
-        json: dict = None
+    base_url: str,
+    token_key: str,
+    method: str = 'POST',
+    params: dict = None,
+    # headers: dict = None,
+    data: dict = None,
+    json: dict = None,
 ):
     """
     通用异步请求函数
@@ -67,18 +61,11 @@ async def async_request(
     :param json: JSON数据
     :return: 响应文本
     """
-    headers = {
-        'Content-Type': 'application/json'
-    }
-    url = f"{base_url}?key={token_key}"
+    headers = {'Content-Type': 'application/json'}
+    url = f'{base_url}?key={token_key}'
     async with aiohttp.ClientSession() as session:
         async with session.request(
-                method=method,
-                url=url,
-                params=params,
-                headers=headers,
-                data=data,
-                json=json
+            method=method, url=url, params=params, headers=headers, data=data, json=json
         ) as response:
             response.raise_for_status()  # 如果状态码不是200，抛出异常
             result = await response.json()
@@ -89,4 +76,3 @@ async def async_request(
             #     return await result
             # else:
             #     raise RuntimeError("请求失败",response.text)
-

libs/wechatpad_api/util/terminal_printer.py

@@ -1,31 +1,34 @@
 import qrcode
 
+
 def print_green(text):
-    print(f"\033[32m{text}\033[0m")
+    print(f'\033[32m{text}\033[0m')
+
 
 def print_yellow(text):
-    print(f"\033[33m{text}\033[0m")
+    print(f'\033[33m{text}\033[0m')
+
 
 def print_red(text):
-    print(f"\033[31m{text}\033[0m")
+    print(f'\033[31m{text}\033[0m')
+
 
 def make_and_print_qr(url):
     """生成并打印二维码
-    
+
     Args:
         url: 需要生成二维码的URL字符串
-        
+
     Returns:
         None
-        
+
     功能:
         1. 在终端打印二维码的ASCII图形
         2. 同时提供在线二维码生成链接作为备选
     """
-    print_green("请扫描下方二维码登录")
+    print_green('请扫描下方二维码登录')
     qr = qrcode.QRCode()
     qr.add_data(url)
     qr.make()
     qr.print_ascii(invert=True)
-    print_green(f"也可以访问下方链接获取二维码:\nhttps://api.qrserver.com/v1/create-qr-code/?data={url}")
-
+    print_green(f'也可以访问下方链接获取二维码:\nhttps://api.qrserver.com/v1/create-qr-code/?data={url}')

libs/wecom_api/api.py

@@ -57,7 +57,7 @@ class WecomClient:
             if 'access_token' in data:
                 return data['access_token']
             else:
-                await self.logger.error(f"获取accesstoken失败:{response.json()}")
+                await self.logger.error(f'获取accesstoken失败:{response.json()}')
                 raise Exception(f'未获取access token: {data}')
 
     async def get_users(self):
@@ -129,7 +129,7 @@ class WecomClient:
                 response = await client.post(url, json=params)
                 data = response.json()
             except Exception as e:
-                await self.logger.error(f"发送图片失败:{data}")
+                await self.logger.error(f'发送图片失败:{data}')
                 raise Exception('Failed to send image: ' + str(e))
 
             # 企业微信错误码40014和42001，代表accesstoken问题
@@ -164,7 +164,7 @@ class WecomClient:
                 self.access_token = await self.get_access_token(self.secret)
                 return await self.send_private_msg(user_id, agent_id, content)
             if data['errcode'] != 0:
-                await self.logger.error(f"发送消息失败:{data}")
+                await self.logger.error(f'发送消息失败:{data}')
                 raise Exception('Failed to send message: ' + str(data))
 
     async def handle_callback_request(self):
@@ -181,7 +181,7 @@ class WecomClient:
                 echostr = request.args.get('echostr')
                 ret, reply_echo_str = wxcpt.VerifyURL(msg_signature, timestamp, nonce, echostr)
                 if ret != 0:
-                    await self.logger.error("验证失败")
+                    await self.logger.error('验证失败')
                     raise Exception(f'验证失败，错误码: {ret}')
                 return reply_echo_str
 
@@ -189,9 +189,8 @@ class WecomClient:
                 encrypt_msg = await request.data
                 ret, xml_msg = wxcpt.DecryptMsg(encrypt_msg, msg_signature, timestamp, nonce)
                 if ret != 0:
-                    await self.logger.error("消息解密失败")
+                    await self.logger.error('消息解密失败')
                     raise Exception(f'消息解密失败，错误码: {ret}')
-                
 
                 # 解析消息并处理
                 message_data = await self.get_message(xml_msg)
@@ -202,7 +201,7 @@ class WecomClient:
 
                 return 'success'
         except Exception as e:
-            await self.logger.error(f"Error in handle_callback_request: {traceback.format_exc()}")
+            await self.logger.error(f'Error in handle_callback_request: {traceback.format_exc()}')
             return f'Error processing request: {str(e)}', 400
 
     async def run_task(self, host: str, port: int, *args, **kwargs):
@@ -301,7 +300,7 @@ class WecomClient:
             except binascii.Error as e:
                 raise ValueError(f'Invalid base64 string: {str(e)}')
         else:
-            await self.logger.error("Image对象出错")
+            await self.logger.error('Image对象出错')
             raise ValueError('image对象出错')
 
         # 设置 multipart/form-data 格式的文件
@@ -325,7 +324,7 @@ class WecomClient:
                 self.access_token = await self.get_access_token(self.secret)
                 media_id = await self.upload_to_work(image)
             if data.get('errcode', 0) != 0:
-                await self.logger.error(f"上传图片失败:{data}")
+                await self.logger.error(f'上传图片失败:{data}')
                 raise Exception('failed to upload file')
 
             media_id = data.get('media_id')

libs/wecom_customer_service_api/api.py

@@ -187,7 +187,7 @@ class WecomCSClient:
                 self.access_token = await self.get_access_token(self.secret)
                 return await self.send_text_msg(open_kfid, external_userid, msgid, content)
             if data['errcode'] != 0:
-                await self.logger.error(f"发送消息失败：{data}")
+                await self.logger.error(f'发送消息失败：{data}')
                 raise Exception('Failed to send message')
             return data
 
@@ -227,7 +227,7 @@ class WecomCSClient:
                 return 'success'
         except Exception as e:
             if self.logger:
-                await self.logger.error(f"Error in handle_callback_request: {traceback.format_exc()}")
+                await self.logger.error(f'Error in handle_callback_request: {traceback.format_exc()}')
             else:
                 traceback.print_exc()
             return f'Error processing request: {str(e)}', 400

pkg/api/http/controller/group.py

@@ -14,8 +14,8 @@ preregistered_groups: list[type[RouterGroup]] = []
 """Pre-registered list of RouterGroup"""
 
 
-def group_class(name: str, path: str) -> None:
-    """Register a RouterGroup"""
+def group_class(name: str, path: str) -> typing.Callable[[typing.Type[RouterGroup]], typing.Type[RouterGroup]]:
+    """注册一个 RouterGroup"""
 
     def decorator(cls: typing.Type[RouterGroup]) -> typing.Type[RouterGroup]:
         cls.name = name
@@ -86,10 +86,11 @@ class RouterGroup(abc.ABC):
 
                 try:
                     return await f(*args, **kwargs)
-                except Exception:  # auto 500
+
+                except Exception as e:  # 自动 500
                     traceback.print_exc()
                     # return self.http_status(500, -2, str(e))
-                    return self.http_status(500, -2, 'internal server error')
+                    return self.http_status(500, -2, str(e))
 
             new_f = handler_error
             new_f.__name__ = (self.name + rule).replace('/', '__')
@@ -120,6 +121,6 @@ class RouterGroup(abc.ABC):
             }
         )
 
-    def http_status(self, status: int, code: int, msg: str) -> quart.Response:
-        """Return a response with a specified status code"""
-        return self.fail(code, msg), status
+    def http_status(self, status: int, code: int, msg: str) -> typing.Tuple[quart.Response, int]:
+        """返回一个指定状态码的响应"""
+        return (self.fail(code, msg), status)

pkg/api/http/controller/groups/files.py

@@ -2,6 +2,10 @@ from __future__ import annotations
 
 import quart
 import mimetypes
+import uuid
+import asyncio
+
+import quart.datastructures
 
 from .. import group
 
@@ -20,3 +24,23 @@ class FilesRouterGroup(group.RouterGroup):
                 mime_type = 'image/jpeg'
 
             return quart.Response(image_bytes, mimetype=mime_type)
+
+        @self.route('/documents', methods=['POST'], auth_type=group.AuthType.USER_TOKEN)
+        async def _() -> quart.Response:
+            request = quart.request
+            # get file bytes from 'file'
+            file = (await request.files)['file']
+            assert isinstance(file, quart.datastructures.FileStorage)
+
+            file_bytes = await asyncio.to_thread(file.stream.read)
+            extension = file.filename.split('.')[-1]
+            file_name = file.filename.split('.')[0]
+
+            file_key = file_name + '_' + str(uuid.uuid4())[:8] + '.' + extension
+            # save file to storage
+            await self.ap.storage_mgr.storage_provider.save(file_key, file_bytes)
+            return self.success(
+                data={
+                    'file_id': file_key,
+                }
+            )

pkg/api/http/controller/groups/knowledge/base.py

@@ -0,0 +1,90 @@
+import quart
+from ... import group
+
+
+@group.group_class('knowledge_base', '/api/v1/knowledge/bases')
+class KnowledgeBaseRouterGroup(group.RouterGroup):
+    async def initialize(self) -> None:
+        @self.route('', methods=['POST', 'GET'])
+        async def handle_knowledge_bases() -> quart.Response:
+            if quart.request.method == 'GET':
+                knowledge_bases = await self.ap.knowledge_service.get_knowledge_bases()
+                return self.success(data={'bases': knowledge_bases})
+
+            elif quart.request.method == 'POST':
+                json_data = await quart.request.json
+                knowledge_base_uuid = await self.ap.knowledge_service.create_knowledge_base(json_data)
+                return self.success(data={'uuid': knowledge_base_uuid})
+
+            return self.http_status(405, -1, 'Method not allowed')
+
+        @self.route(
+            '/<knowledge_base_uuid>',
+            methods=['GET', 'DELETE', 'PUT'],
+        )
+        async def handle_specific_knowledge_base(knowledge_base_uuid: str) -> quart.Response:
+            if quart.request.method == 'GET':
+                knowledge_base = await self.ap.knowledge_service.get_knowledge_base(knowledge_base_uuid)
+
+                if knowledge_base is None:
+                    return self.http_status(404, -1, 'knowledge base not found')
+
+                return self.success(
+                    data={
+                        'base': knowledge_base,
+                    }
+                )
+
+            elif quart.request.method == 'PUT':
+                json_data = await quart.request.json
+                await self.ap.knowledge_service.update_knowledge_base(knowledge_base_uuid, json_data)
+                return self.success({})
+
+            elif quart.request.method == 'DELETE':
+                await self.ap.knowledge_service.delete_knowledge_base(knowledge_base_uuid)
+                return self.success({})
+
+        @self.route(
+            '/<knowledge_base_uuid>/files',
+            methods=['GET', 'POST'],
+        )
+        async def get_knowledge_base_files(knowledge_base_uuid: str) -> str:
+            if quart.request.method == 'GET':
+                files = await self.ap.knowledge_service.get_files_by_knowledge_base(knowledge_base_uuid)
+                return self.success(
+                    data={
+                        'files': files,
+                    }
+                )
+
+            elif quart.request.method == 'POST':
+                json_data = await quart.request.json
+                file_id = json_data.get('file_id')
+                if not file_id:
+                    return self.http_status(400, -1, 'File ID is required')
+
+                # 调用服务层方法将文件与知识库关联
+                task_id = await self.ap.knowledge_service.store_file(knowledge_base_uuid, file_id)
+                return self.success(
+                    {
+                        'task_id': task_id,
+                    }
+                )
+
+        @self.route(
+            '/<knowledge_base_uuid>/files/<file_id>',
+            methods=['DELETE'],
+        )
+        async def delete_specific_file_in_kb(file_id: str, knowledge_base_uuid: str) -> str:
+            await self.ap.knowledge_service.delete_file(knowledge_base_uuid, file_id)
+            return self.success({})
+
+        @self.route(
+            '/<knowledge_base_uuid>/retrieve',
+            methods=['POST'],
+        )
+        async def retrieve_knowledge_base(knowledge_base_uuid: str) -> str:
+            json_data = await quart.request.json
+            query = json_data.get('query')
+            results = await self.ap.knowledge_service.retrieve_knowledge_base(knowledge_base_uuid, query)
+            return self.success(data={'results': results})

pkg/api/http/controller/groups/pipelines/pipelines.py

@@ -11,7 +11,11 @@ class PipelinesRouterGroup(group.RouterGroup):
         @self.route('', methods=['GET', 'POST'])
         async def _() -> str:
             if quart.request.method == 'GET':
-                return self.success(data={'pipelines': await self.ap.pipeline_service.get_pipelines()})
+                sort_by = quart.request.args.get('sort_by', 'created_at')
+                sort_order = quart.request.args.get('sort_order', 'DESC')
+                return self.success(
+                    data={'pipelines': await self.ap.pipeline_service.get_pipelines(sort_by, sort_order)}
+                )
             elif quart.request.method == 'POST':
                 json_data = await quart.request.json
 

pkg/api/http/controller/groups/pipelines/webchat.py

@@ -1,3 +1,5 @@
+import json
+
 import quart
 
 from ... import group
@@ -9,10 +11,18 @@ class WebChatDebugRouterGroup(group.RouterGroup):
         @self.route('/send', methods=['POST'])
         async def send_message(pipeline_uuid: str) -> str:
             """Send a message to the pipeline for debugging"""
+
+            async def stream_generator(generator):
+                yield 'data: {"type": "start"}\n\n'
+                async for message in generator:
+                    yield f'data: {json.dumps({"message": message})}\n\n'
+                yield 'data: {"type": "end"}\n\n'
+
             try:
                 data = await quart.request.get_json()
                 session_type = data.get('session_type', 'person')
                 message_chain_obj = data.get('message', [])
+                is_stream = data.get('is_stream', False)
 
                 if not message_chain_obj:
                     return self.http_status(400, -1, 'message is required')
@@ -25,13 +35,33 @@ class WebChatDebugRouterGroup(group.RouterGroup):
                 if not webchat_adapter:
                     return self.http_status(404, -1, 'WebChat adapter not found')
 
-                result = await webchat_adapter.send_webchat_message(pipeline_uuid, session_type, message_chain_obj)
-
-                return self.success(
-                    data={
-                        'message': result,
+                if is_stream:
+                    generator = webchat_adapter.send_webchat_message(
+                        pipeline_uuid, session_type, message_chain_obj, is_stream
+                    )
+                    # 设置正确的响应头
+                    headers = {
+                        'Content-Type': 'text/event-stream',
+                        'Transfer-Encoding': 'chunked',
+                        'Cache-Control': 'no-cache',
+                        'Connection': 'keep-alive'
                     }
-                )
+                    return quart.Response(stream_generator(generator), mimetype='text/event-stream',headers=headers)
+
+                else:  # non-stream
+                    result = None
+                    async for message in webchat_adapter.send_webchat_message(
+                        pipeline_uuid, session_type, message_chain_obj
+                    ):
+                        result = message
+                    if result is not None:
+                        return self.success(
+                            data={
+                                'message': result,
+                            }
+                        )
+                    else:
+                        return self.http_status(400, -1, 'message is required')
 
             except Exception as e:
                 return self.http_status(500, -1, f'Internal server error: {str(e)}')

pkg/api/http/controller/groups/provider/models.py

@@ -9,18 +9,18 @@ class LLMModelsRouterGroup(group.RouterGroup):
         @self.route('', methods=['GET', 'POST'])
         async def _() -> str:
             if quart.request.method == 'GET':
-                return self.success(data={'models': await self.ap.model_service.get_llm_models()})
+                return self.success(data={'models': await self.ap.llm_model_service.get_llm_models()})
             elif quart.request.method == 'POST':
                 json_data = await quart.request.json
 
-                model_uuid = await self.ap.model_service.create_llm_model(json_data)
+                model_uuid = await self.ap.llm_model_service.create_llm_model(json_data)
 
                 return self.success(data={'uuid': model_uuid})
 
         @self.route('/<model_uuid>', methods=['GET', 'PUT', 'DELETE'])
         async def _(model_uuid: str) -> str:
             if quart.request.method == 'GET':
-                model = await self.ap.model_service.get_llm_model(model_uuid)
+                model = await self.ap.llm_model_service.get_llm_model(model_uuid)
 
                 if model is None:
                     return self.http_status(404, -1, 'model not found')
@@ -29,11 +29,11 @@ class LLMModelsRouterGroup(group.RouterGroup):
             elif quart.request.method == 'PUT':
                 json_data = await quart.request.json
 
-                await self.ap.model_service.update_llm_model(model_uuid, json_data)
+                await self.ap.llm_model_service.update_llm_model(model_uuid, json_data)
 
                 return self.success()
             elif quart.request.method == 'DELETE':
-                await self.ap.model_service.delete_llm_model(model_uuid)
+                await self.ap.llm_model_service.delete_llm_model(model_uuid)
 
                 return self.success()
 
@@ -41,6 +41,49 @@ class LLMModelsRouterGroup(group.RouterGroup):
         async def _(model_uuid: str) -> str:
             json_data = await quart.request.json
 
-            await self.ap.model_service.test_llm_model(model_uuid, json_data)
+            await self.ap.llm_model_service.test_llm_model(model_uuid, json_data)
+
+            return self.success()
+
+
+@group.group_class('models/embedding', '/api/v1/provider/models/embedding')
+class EmbeddingModelsRouterGroup(group.RouterGroup):
+    async def initialize(self) -> None:
+        @self.route('', methods=['GET', 'POST'])
+        async def _() -> str:
+            if quart.request.method == 'GET':
+                return self.success(data={'models': await self.ap.embedding_models_service.get_embedding_models()})
+            elif quart.request.method == 'POST':
+                json_data = await quart.request.json
+
+                model_uuid = await self.ap.embedding_models_service.create_embedding_model(json_data)
+
+                return self.success(data={'uuid': model_uuid})
+
+        @self.route('/<model_uuid>', methods=['GET', 'PUT', 'DELETE'])
+        async def _(model_uuid: str) -> str:
+            if quart.request.method == 'GET':
+                model = await self.ap.embedding_models_service.get_embedding_model(model_uuid)
+
+                if model is None:
+                    return self.http_status(404, -1, 'model not found')
+
+                return self.success(data={'model': model})
+            elif quart.request.method == 'PUT':
+                json_data = await quart.request.json
+
+                await self.ap.embedding_models_service.update_embedding_model(model_uuid, json_data)
+
+                return self.success()
+            elif quart.request.method == 'DELETE':
+                await self.ap.embedding_models_service.delete_embedding_model(model_uuid)
+
+                return self.success()
+
+        @self.route('/<model_uuid>/test', methods=['POST'])
+        async def _(model_uuid: str) -> str:
+            json_data = await quart.request.json
+
+            await self.ap.embedding_models_service.test_embedding_model(model_uuid, json_data)
 
             return self.success()

pkg/api/http/controller/groups/provider/requesters.py

@@ -8,7 +8,8 @@ class RequestersRouterGroup(group.RouterGroup):
     async def initialize(self) -> None:
         @self.route('', methods=['GET'])
         async def _() -> quart.Response:
-            return self.success(data={'requesters': self.ap.model_mgr.get_available_requesters_info()})
+            model_type = quart.request.args.get('type', '')
+            return self.success(data={'requesters': self.ap.model_mgr.get_available_requesters_info(model_type)})
 
         @self.route('/<requester_name>', methods=['GET'])
         async def _(requester_name: str) -> quart.Response:

pkg/api/http/controller/groups/user.py

@@ -67,3 +67,19 @@ class UserRouterGroup(group.RouterGroup):
             await self.ap.user_service.reset_password(user_email, new_password)
 
             return self.success(data={'user': user_email})
+
+        @self.route('/change-password', methods=['POST'], auth_type=group.AuthType.USER_TOKEN)
+        async def _(user_email: str) -> str:
+            json_data = await quart.request.json
+            
+            current_password = json_data['current_password']
+            new_password = json_data['new_password']
+            
+            try:
+                await self.ap.user_service.change_password(user_email, current_password, new_password)
+            except argon2.exceptions.VerifyMismatchError:
+                return self.http_status(400, -1, 'Current password is incorrect')
+            except ValueError as e:
+                return self.http_status(400, -1, str(e))
+            
+            return self.success(data={'user': user_email})

pkg/api/http/controller/main.py

@@ -14,11 +14,13 @@ from . import group
 from .groups import provider as groups_provider
 from .groups import platform as groups_platform
 from .groups import pipelines as groups_pipelines
+from .groups import knowledge as groups_knowledge
 
 importutil.import_modules_in_pkg(groups)
 importutil.import_modules_in_pkg(groups_provider)
 importutil.import_modules_in_pkg(groups_platform)
 importutil.import_modules_in_pkg(groups_pipelines)
+importutil.import_modules_in_pkg(groups_knowledge)
 
 
 class HTTPController:

pkg/api/http/service/knowledge.py

@@ -0,0 +1,120 @@
+from __future__ import annotations
+
+import uuid
+import sqlalchemy
+
+from ....core import app
+from ....entity.persistence import rag as persistence_rag
+
+
+class KnowledgeService:
+    """知识库服务"""
+
+    ap: app.Application
+
+    def __init__(self, ap: app.Application) -> None:
+        self.ap = ap
+
+    async def get_knowledge_bases(self) -> list[dict]:
+        """获取所有知识库"""
+        result = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_rag.KnowledgeBase))
+        knowledge_bases = result.all()
+        return [
+            self.ap.persistence_mgr.serialize_model(persistence_rag.KnowledgeBase, knowledge_base)
+            for knowledge_base in knowledge_bases
+        ]
+
+    async def get_knowledge_base(self, kb_uuid: str) -> dict | None:
+        """获取知识库"""
+        result = await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.select(persistence_rag.KnowledgeBase).where(persistence_rag.KnowledgeBase.uuid == kb_uuid)
+        )
+        knowledge_base = result.first()
+        if knowledge_base is None:
+            return None
+        return self.ap.persistence_mgr.serialize_model(persistence_rag.KnowledgeBase, knowledge_base)
+
+    async def create_knowledge_base(self, kb_data: dict) -> str:
+        """创建知识库"""
+        kb_data['uuid'] = str(uuid.uuid4())
+        await self.ap.persistence_mgr.execute_async(sqlalchemy.insert(persistence_rag.KnowledgeBase).values(kb_data))
+
+        kb = await self.get_knowledge_base(kb_data['uuid'])
+
+        await self.ap.rag_mgr.load_knowledge_base(kb)
+
+        return kb_data['uuid']
+
+    async def update_knowledge_base(self, kb_uuid: str, kb_data: dict) -> None:
+        """更新知识库"""
+        if 'uuid' in kb_data:
+            del kb_data['uuid']
+
+        if 'embedding_model_uuid' in kb_data:
+            del kb_data['embedding_model_uuid']
+
+        await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.update(persistence_rag.KnowledgeBase)
+            .values(kb_data)
+            .where(persistence_rag.KnowledgeBase.uuid == kb_uuid)
+        )
+        await self.ap.rag_mgr.remove_knowledge_base_from_runtime(kb_uuid)
+
+        kb = await self.get_knowledge_base(kb_uuid)
+
+        await self.ap.rag_mgr.load_knowledge_base(kb)
+
+    async def store_file(self, kb_uuid: str, file_id: str) -> int:
+        """存储文件"""
+        # await self.ap.persistence_mgr.execute_async(sqlalchemy.insert(persistence_rag.File).values(kb_id=kb_uuid, file_id=file_id))
+        # await self.ap.rag_mgr.store_file(file_id)
+        runtime_kb = await self.ap.rag_mgr.get_knowledge_base_by_uuid(kb_uuid)
+        if runtime_kb is None:
+            raise Exception('Knowledge base not found')
+        return await runtime_kb.store_file(file_id)
+
+    async def retrieve_knowledge_base(self, kb_uuid: str, query: str) -> list[dict]:
+        """检索知识库"""
+        runtime_kb = await self.ap.rag_mgr.get_knowledge_base_by_uuid(kb_uuid)
+        if runtime_kb is None:
+            raise Exception('Knowledge base not found')
+        return [
+            result.model_dump() for result in await runtime_kb.retrieve(query, runtime_kb.knowledge_base_entity.top_k)
+        ]
+
+    async def get_files_by_knowledge_base(self, kb_uuid: str) -> list[dict]:
+        """获取知识库文件"""
+        result = await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.select(persistence_rag.File).where(persistence_rag.File.kb_id == kb_uuid)
+        )
+        files = result.all()
+        return [self.ap.persistence_mgr.serialize_model(persistence_rag.File, file) for file in files]
+
+    async def delete_file(self, kb_uuid: str, file_id: str) -> None:
+        """删除文件"""
+        runtime_kb = await self.ap.rag_mgr.get_knowledge_base_by_uuid(kb_uuid)
+        if runtime_kb is None:
+            raise Exception('Knowledge base not found')
+        await runtime_kb.delete_file(file_id)
+
+    async def delete_knowledge_base(self, kb_uuid: str) -> None:
+        """删除知识库"""
+        await self.ap.rag_mgr.delete_knowledge_base(kb_uuid)
+
+        await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.delete(persistence_rag.KnowledgeBase).where(persistence_rag.KnowledgeBase.uuid == kb_uuid)
+        )
+
+        # delete files
+        files = await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.select(persistence_rag.File).where(persistence_rag.File.kb_id == kb_uuid)
+        )
+        for file in files:
+            # delete chunks
+            await self.ap.persistence_mgr.execute_async(
+                sqlalchemy.delete(persistence_rag.Chunk).where(persistence_rag.Chunk.file_id == file.uuid)
+            )
+            # delete file
+            await self.ap.persistence_mgr.execute_async(
+                sqlalchemy.delete(persistence_rag.File).where(persistence_rag.File.uuid == file.uuid)
+            )

pkg/api/http/service/model.py

@@ -10,7 +10,7 @@ from ....provider.modelmgr import requester as model_requester
 from ....provider import entities as llm_entities
 
 
-class ModelsService:
+class LLMModelsService:
     ap: app.Application
 
     def __init__(self, ap: app.Application) -> None:
@@ -101,5 +101,91 @@ class ModelsService:
             model=runtime_llm_model,
             messages=[llm_entities.Message(role='user', content='Hello, world!')],
             funcs=[],
+            extra_args=model_data.get('extra_args', {}),
+        )
+
+
+class EmbeddingModelsService:
+    ap: app.Application
+
+    def __init__(self, ap: app.Application) -> None:
+        self.ap = ap
+
+    async def get_embedding_models(self) -> list[dict]:
+        result = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_model.EmbeddingModel))
+
+        models = result.all()
+        return [self.ap.persistence_mgr.serialize_model(persistence_model.EmbeddingModel, model) for model in models]
+
+    async def create_embedding_model(self, model_data: dict) -> str:
+        model_data['uuid'] = str(uuid.uuid4())
+
+        await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.insert(persistence_model.EmbeddingModel).values(**model_data)
+        )
+
+        embedding_model = await self.get_embedding_model(model_data['uuid'])
+
+        await self.ap.model_mgr.load_embedding_model(embedding_model)
+
+        return model_data['uuid']
+
+    async def get_embedding_model(self, model_uuid: str) -> dict | None:
+        result = await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.select(persistence_model.EmbeddingModel).where(
+                persistence_model.EmbeddingModel.uuid == model_uuid
+            )
+        )
+
+        model = result.first()
+
+        if model is None:
+            return None
+
+        return self.ap.persistence_mgr.serialize_model(persistence_model.EmbeddingModel, model)
+
+    async def update_embedding_model(self, model_uuid: str, model_data: dict) -> None:
+        if 'uuid' in model_data:
+            del model_data['uuid']
+
+        await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.update(persistence_model.EmbeddingModel)
+            .where(persistence_model.EmbeddingModel.uuid == model_uuid)
+            .values(**model_data)
+        )
+
+        await self.ap.model_mgr.remove_embedding_model(model_uuid)
+
+        embedding_model = await self.get_embedding_model(model_uuid)
+
+        await self.ap.model_mgr.load_embedding_model(embedding_model)
+
+    async def delete_embedding_model(self, model_uuid: str) -> None:
+        await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.delete(persistence_model.EmbeddingModel).where(
+                persistence_model.EmbeddingModel.uuid == model_uuid
+            )
+        )
+
+        await self.ap.model_mgr.remove_embedding_model(model_uuid)
+
+    async def test_embedding_model(self, model_uuid: str, model_data: dict) -> None:
+        runtime_embedding_model: model_requester.RuntimeEmbeddingModel | None = None
+
+        if model_uuid != '_':
+            for model in self.ap.model_mgr.embedding_models:
+                if model.model_entity.uuid == model_uuid:
+                    runtime_embedding_model = model
+                    break
+
+            if runtime_embedding_model is None:
+                raise Exception('model not found')
+
+        else:
+            runtime_embedding_model = await self.ap.model_mgr.init_runtime_embedding_model(model_data)
+
+        await runtime_embedding_model.requester.invoke_embedding(
+            model=runtime_embedding_model,
+            input_text=['Hello, world!'],
             extra_args={},
         )

pkg/api/http/service/pipeline.py

@@ -38,9 +38,21 @@ class PipelineService:
             self.ap.pipeline_config_meta_output.data,
         ]
 
-    async def get_pipelines(self) -> list[dict]:
-        result = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_pipeline.LegacyPipeline))
+    async def get_pipelines(self, sort_by: str = 'created_at', sort_order: str = 'DESC') -> list[dict]:
+        query = sqlalchemy.select(persistence_pipeline.LegacyPipeline)
 
+        if sort_by == 'created_at':
+            if sort_order == 'DESC':
+                query = query.order_by(persistence_pipeline.LegacyPipeline.created_at.desc())
+            else:
+                query = query.order_by(persistence_pipeline.LegacyPipeline.created_at.asc())
+        elif sort_by == 'updated_at':
+            if sort_order == 'DESC':
+                query = query.order_by(persistence_pipeline.LegacyPipeline.updated_at.desc())
+            else:
+                query = query.order_by(persistence_pipeline.LegacyPipeline.updated_at.asc())
+
+        result = await self.ap.persistence_mgr.execute_async(query)
         pipelines = result.all()
         return [
             self.ap.persistence_mgr.serialize_model(persistence_pipeline.LegacyPipeline, pipeline)

pkg/api/http/service/user.py

@@ -82,3 +82,18 @@ class UserService:
         await self.ap.persistence_mgr.execute_async(
             sqlalchemy.update(user.User).where(user.User.user == user_email).values(password=hashed_password)
         )
+
+    async def change_password(self, user_email: str, current_password: str, new_password: str) -> None:
+        ph = argon2.PasswordHasher()
+        
+        user_obj = await self.get_user_by_email(user_email)
+        if user_obj is None:
+            raise ValueError('User not found')
+        
+        ph.verify(user_obj.password, current_password)
+        
+        hashed_password = ph.hash(new_password)
+        
+        await self.ap.persistence_mgr.execute_async(
+            sqlalchemy.update(user.User).where(user.User.user == user_email).values(password=hashed_password)
+        )

pkg/core/app.py

@@ -22,11 +22,14 @@ from ..api.http.service import user as user_service
 from ..api.http.service import model as model_service
 from ..api.http.service import pipeline as pipeline_service
 from ..api.http.service import bot as bot_service
+from ..api.http.service import knowledge as knowledge_service
 from ..discover import engine as discover_engine
 from ..storage import mgr as storagemgr
 from ..utils import logcache
 from . import taskmgr
 from . import entities as core_entities
+from ..rag.knowledge import kbmgr as rag_mgr
+from ..vector import mgr as vectordb_mgr
 
 
 class Application:
@@ -47,6 +50,8 @@ class Application:
 
     model_mgr: llm_model_mgr.ModelManager = None
 
+    rag_mgr: rag_mgr.RAGManager = None
+
     # TODO move to pipeline
     tool_mgr: llm_tool_mgr.ToolManager = None
 
@@ -93,6 +98,8 @@ class Application:
 
     persistence_mgr: persistencemgr.PersistenceManager = None
 
+    vector_db_mgr: vectordb_mgr.VectorDBManager = None
+
     http_ctrl: http_controller.HTTPController = None
 
     log_cache: logcache.LogCache = None
@@ -103,12 +110,16 @@ class Application:
 
     user_service: user_service.UserService = None
 
-    model_service: model_service.ModelsService = None
+    llm_model_service: model_service.LLMModelsService = None
+
+    embedding_models_service: model_service.EmbeddingModelsService = None
 
     pipeline_service: pipeline_service.PipelineService = None
 
     bot_service: bot_service.BotService = None
 
+    knowledge_service: knowledge_service.KnowledgeService = None
+
     def __init__(self):
         pass
 
@@ -143,6 +154,7 @@ class Application:
                 name='http-api-controller',
                 scopes=[core_entities.LifecycleControlScope.APPLICATION],
             )
+
             self.task_mgr.create_task(
                 never_ending(),
                 name='never-ending-task',

pkg/core/entities.py

@@ -87,7 +87,9 @@ class Query(pydantic.BaseModel):
     """使用的函数，由前置处理器阶段设置"""
 
     resp_messages: (
-        typing.Optional[list[llm_entities.Message]] | typing.Optional[list[platform_message.MessageChain]]
+        typing.Optional[list[llm_entities.Message]]
+        | typing.Optional[list[platform_message.MessageChain]]
+        | typing.Optional[list[llm_entities.MessageChunk]]
     ) = []
     """由Process阶段生成的回复消息对象列表"""
 

pkg/core/stages/build_app.py

@@ -9,6 +9,7 @@ from ...command import cmdmgr
 from ...provider.session import sessionmgr as llm_session_mgr
 from ...provider.modelmgr import modelmgr as llm_model_mgr
 from ...provider.tools import toolmgr as llm_tool_mgr
+from ...rag.knowledge import kbmgr as rag_mgr
 from ...platform import botmgr as im_mgr
 from ...persistence import mgr as persistencemgr
 from ...api.http.controller import main as http_controller
@@ -16,9 +17,11 @@ from ...api.http.service import user as user_service
 from ...api.http.service import model as model_service
 from ...api.http.service import pipeline as pipeline_service
 from ...api.http.service import bot as bot_service
+from ...api.http.service import knowledge as knowledge_service
 from ...discover import engine as discover_engine
 from ...storage import mgr as storagemgr
 from ...utils import logcache
+from ...vector import mgr as vectordb_mgr
 from .. import taskmgr
 
 
@@ -88,6 +91,15 @@ class BuildAppStage(stage.BootingStage):
         await pipeline_mgr.initialize()
         ap.pipeline_mgr = pipeline_mgr
 
+        rag_mgr_inst = rag_mgr.RAGManager(ap)
+        await rag_mgr_inst.initialize()
+        ap.rag_mgr = rag_mgr_inst
+
+        # 初始化向量数据库管理器
+        vectordb_mgr_inst = vectordb_mgr.VectorDBManager(ap)
+        await vectordb_mgr_inst.initialize()
+        ap.vector_db_mgr = vectordb_mgr_inst
+
         http_ctrl = http_controller.HTTPController(ap)
         await http_ctrl.initialize()
         ap.http_ctrl = http_ctrl
@@ -95,8 +107,11 @@ class BuildAppStage(stage.BootingStage):
         user_service_inst = user_service.UserService(ap)
         ap.user_service = user_service_inst
 
-        model_service_inst = model_service.ModelsService(ap)
-        ap.model_service = model_service_inst
+        llm_model_service_inst = model_service.LLMModelsService(ap)
+        ap.llm_model_service = llm_model_service_inst
+
+        embedding_models_service_inst = model_service.EmbeddingModelsService(ap)
+        ap.embedding_models_service = embedding_models_service_inst
 
         pipeline_service_inst = pipeline_service.PipelineService(ap)
         ap.pipeline_service = pipeline_service_inst
@@ -104,5 +119,8 @@ class BuildAppStage(stage.BootingStage):
         bot_service_inst = bot_service.BotService(ap)
         ap.bot_service = bot_service_inst
 
+        knowledge_service_inst = knowledge_service.KnowledgeService(ap)
+        ap.knowledge_service = knowledge_service_inst
+
         ctrl = controller.Controller(ap)
         ap.ctrl = ctrl

pkg/entity/persistence/model.py

@@ -23,3 +23,24 @@ class LLMModel(Base):
         server_default=sqlalchemy.func.now(),
         onupdate=sqlalchemy.func.now(),
     )
+
+
+class EmbeddingModel(Base):
+    """Embedding 模型"""
+
+    __tablename__ = 'embedding_models'
+
+    uuid = sqlalchemy.Column(sqlalchemy.String(255), primary_key=True, unique=True)
+    name = sqlalchemy.Column(sqlalchemy.String(255), nullable=False)
+    description = sqlalchemy.Column(sqlalchemy.String(255), nullable=False)
+    requester = sqlalchemy.Column(sqlalchemy.String(255), nullable=False)
+    requester_config = sqlalchemy.Column(sqlalchemy.JSON, nullable=False, default={})
+    api_keys = sqlalchemy.Column(sqlalchemy.JSON, nullable=False)
+    extra_args = sqlalchemy.Column(sqlalchemy.JSON, nullable=False, default={})
+    created_at = sqlalchemy.Column(sqlalchemy.DateTime, nullable=False, server_default=sqlalchemy.func.now())
+    updated_at = sqlalchemy.Column(
+        sqlalchemy.DateTime,
+        nullable=False,
+        server_default=sqlalchemy.func.now(),
+        onupdate=sqlalchemy.func.now(),
+    )

pkg/entity/persistence/pipeline.py

@@ -20,7 +20,6 @@ class LegacyPipeline(Base):
     )
     for_version = sqlalchemy.Column(sqlalchemy.String(255), nullable=False)
     is_default = sqlalchemy.Column(sqlalchemy.Boolean, nullable=False, default=False)
-
     stages = sqlalchemy.Column(sqlalchemy.JSON, nullable=False)
     config = sqlalchemy.Column(sqlalchemy.JSON, nullable=False)
 
@@ -43,3 +42,4 @@ class PipelineRunRecord(Base):
     started_at = sqlalchemy.Column(sqlalchemy.DateTime, nullable=False)
     finished_at = sqlalchemy.Column(sqlalchemy.DateTime, nullable=False)
     result = sqlalchemy.Column(sqlalchemy.JSON, nullable=False)
+    knowledge_base_uuid = sqlalchemy.Column(sqlalchemy.String(255), nullable=True)

pkg/entity/persistence/rag.py

@@ -0,0 +1,50 @@
+import sqlalchemy
+from .base import Base
+
+# Base = declarative_base()
+# DATABASE_URL = os.getenv('DATABASE_URL', 'sqlite:///./rag_knowledge.db')
+# print("Using database URL:", DATABASE_URL)
+
+
+# engine = create_engine(DATABASE_URL, connect_args={'check_same_thread': False})
+
+# SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
+
+# def create_db_and_tables():
+#     """Creates all database tables defined in the Base."""
+#     Base.metadata.create_all(bind=engine)
+#     print('Database tables created or already exist.')
+
+
+class KnowledgeBase(Base):
+    __tablename__ = 'knowledge_bases'
+    uuid = sqlalchemy.Column(sqlalchemy.String(255), primary_key=True, unique=True)
+    name = sqlalchemy.Column(sqlalchemy.String, index=True)
+    description = sqlalchemy.Column(sqlalchemy.Text)
+    created_at = sqlalchemy.Column(sqlalchemy.DateTime, default=sqlalchemy.func.now())
+    embedding_model_uuid = sqlalchemy.Column(sqlalchemy.String, default='')
+    top_k = sqlalchemy.Column(sqlalchemy.Integer, default=5)
+
+
+class File(Base):
+    __tablename__ = 'knowledge_base_files'
+    uuid = sqlalchemy.Column(sqlalchemy.String(255), primary_key=True, unique=True)
+    kb_id = sqlalchemy.Column(sqlalchemy.String(255), nullable=True)
+    file_name = sqlalchemy.Column(sqlalchemy.String)
+    extension = sqlalchemy.Column(sqlalchemy.String)
+    created_at = sqlalchemy.Column(sqlalchemy.DateTime, default=sqlalchemy.func.now())
+    status = sqlalchemy.Column(sqlalchemy.String, default='pending')  # pending, processing, completed, failed
+
+
+class Chunk(Base):
+    __tablename__ = 'knowledge_base_chunks'
+    uuid = sqlalchemy.Column(sqlalchemy.String(255), primary_key=True, unique=True)
+    file_id = sqlalchemy.Column(sqlalchemy.String(255), nullable=True)
+    text = sqlalchemy.Column(sqlalchemy.Text)
+
+
+# class Vector(Base):
+#     __tablename__ = 'knowledge_base_vectors'
+#     uuid = sqlalchemy.Column(sqlalchemy.String(255), primary_key=True, unique=True)
+#     chunk_id = sqlalchemy.Column(sqlalchemy.String, nullable=True)
+#     embedding = sqlalchemy.Column(sqlalchemy.LargeBinary)

pkg/entity/persistence/vector.py

@@ -0,0 +1,13 @@
+from sqlalchemy import Column, Integer, ForeignKey, LargeBinary
+from sqlalchemy.orm import declarative_base, relationship
+
+Base = declarative_base()
+
+
+class Vector(Base):
+    __tablename__ = 'vectors'
+    id = Column(Integer, primary_key=True, index=True)
+    chunk_id = Column(Integer, ForeignKey('chunks.id'), unique=True)
+    embedding = Column(LargeBinary)  # Store embeddings as binary
+
+    chunk = relationship('Chunk', back_populates='vector')

pkg/entity/rag/retriever.py

@@ -0,0 +1,13 @@
+from __future__ import annotations
+
+import pydantic
+
+from typing import Any
+
+
+class RetrieveResultEntry(pydantic.BaseModel):
+    id: str
+
+    metadata: dict[str, Any]
+
+    distance: float

pkg/persistence/mgr.py

@@ -79,7 +79,7 @@ class PersistenceManager:
                 'stages': pipeline_service.default_stage_order,
                 'is_default': True,
                 'name': 'ChatPipeline',
-                'description': 'Default pipeline provided, your new bots will be automatically bound to this pipeline | 默认提供的流水线，您配置的机器人将自动绑定到此流水线',
+                'description': 'Default pipeline, new bots will be bound to this pipeline | 默认提供的流水线，您配置的机器人将自动绑定到此流水线',
                 'config': pipeline_config,
             }
 

pkg/persistence/migrations/dbm004_rag_kb_uuid.py

@@ -0,0 +1,38 @@
+from .. import migration
+
+import sqlalchemy
+
+from ...entity.persistence import pipeline as persistence_pipeline
+
+
+@migration.migration_class(4)
+class DBMigrateRAGKBUUID(migration.DBMigration):
+    """RAG知识库UUID"""
+
+    async def upgrade(self):
+        """升级"""
+        # read all pipelines
+        pipelines = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_pipeline.LegacyPipeline))
+
+        for pipeline in pipelines:
+            serialized_pipeline = self.ap.persistence_mgr.serialize_model(persistence_pipeline.LegacyPipeline, pipeline)
+
+            config = serialized_pipeline['config']
+
+            if 'knowledge-base' not in config['ai']['local-agent']:
+                config['ai']['local-agent']['knowledge-base'] = ''
+
+            await self.ap.persistence_mgr.execute_async(
+                sqlalchemy.update(persistence_pipeline.LegacyPipeline)
+                .where(persistence_pipeline.LegacyPipeline.uuid == serialized_pipeline['uuid'])
+                .values(
+                    {
+                        'config': config,
+                        'for_version': self.ap.ver_mgr.get_current_version(),
+                    }
+                )
+            )
+
+    async def downgrade(self):
+        """降级"""
+        pass

pkg/persistence/migrations/dbm005_pipeline_remove_cot_config.py

@@ -0,0 +1,38 @@
+from .. import migration
+
+import sqlalchemy
+
+from ...entity.persistence import pipeline as persistence_pipeline
+
+
+@migration.migration_class(5)
+class DBMigratePipelineRemoveCotConfig(migration.DBMigration):
+    """Pipeline remove cot config"""
+
+    async def upgrade(self):
+        """Upgrade"""
+        # read all pipelines
+        pipelines = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_pipeline.LegacyPipeline))
+
+        for pipeline in pipelines:
+            serialized_pipeline = self.ap.persistence_mgr.serialize_model(persistence_pipeline.LegacyPipeline, pipeline)
+
+            config = serialized_pipeline['config']
+
+            if 'remove-think' not in config['output']['misc']:
+                config['output']['misc']['remove-think'] = False
+
+            await self.ap.persistence_mgr.execute_async(
+                sqlalchemy.update(persistence_pipeline.LegacyPipeline)
+                .where(persistence_pipeline.LegacyPipeline.uuid == serialized_pipeline['uuid'])
+                .values(
+                    {
+                        'config': config,
+                        'for_version': self.ap.ver_mgr.get_current_version(),
+                    }
+                )
+            )
+
+    async def downgrade(self):
+        """Downgrade"""
+        pass

pkg/pipeline/cntfilter/cntfilter.py

@@ -67,7 +67,7 @@ class ContentFilterStage(stage.PipelineStage):
         if query.pipeline_config['safety']['content-filter']['scope'] == 'output-msg':
             return entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)
         if not message.strip():
-            return entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)    
+            return entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)
         else:
             for filter in self.filter_chain:
                 if filter_entities.EnableStage.PRE in filter.enable_stages:

pkg/pipeline/pipelinemgr.py

@@ -93,12 +93,20 @@ class RuntimePipeline:
                 query.message_event, platform_events.GroupMessage
             ):
                 result.user_notice.insert(0, platform_message.At(query.message_event.sender.id))
-
-            await query.adapter.reply_message(
-                message_source=query.message_event,
-                message=result.user_notice,
-                quote_origin=query.pipeline_config['output']['misc']['quote-origin'],
-            )
+            if await query.adapter.is_stream_output_supported():
+                await query.adapter.reply_message_chunk(
+                    message_source=query.message_event,
+                    bot_message=query.resp_messages[-1],
+                    message=result.user_notice,
+                    quote_origin=query.pipeline_config['output']['misc']['quote-origin'],
+                    is_final=[msg.is_final for msg in query.resp_messages][0]
+                )
+            else:
+                await query.adapter.reply_message(
+                    message_source=query.message_event,
+                    message=result.user_notice,
+                    quote_origin=query.pipeline_config['output']['misc']['quote-origin'],
+                )
         if result.debug_notice:
             self.ap.logger.debug(result.debug_notice)
         if result.console_notice:
@@ -144,23 +152,27 @@ class RuntimePipeline:
                 result = await result
 
             if isinstance(result, pipeline_entities.StageProcessResult):  # 直接返回结果
-                self.ap.logger.debug(f'Stage {stage_container.inst_name} processed query {query} res {result}')
+                self.ap.logger.debug(
+                    f'Stage {stage_container.inst_name} processed query {query.query_id} res {result.result_type}'
+                )
                 await self._check_output(query, result)
 
                 if result.result_type == pipeline_entities.ResultType.INTERRUPT:
-                    self.ap.logger.debug(f'Stage {stage_container.inst_name} interrupted query {query}')
+                    self.ap.logger.debug(f'Stage {stage_container.inst_name} interrupted query {query.query_id}')
                     break
                 elif result.result_type == pipeline_entities.ResultType.CONTINUE:
                     query = result.new_query
             elif isinstance(result, typing.AsyncGenerator):  # 生成器
-                self.ap.logger.debug(f'Stage {stage_container.inst_name} processed query {query} gen')
+                self.ap.logger.debug(f'Stage {stage_container.inst_name} processed query {query.query_id} gen')
 
                 async for sub_result in result:
-                    self.ap.logger.debug(f'Stage {stage_container.inst_name} processed query {query} res {sub_result}')
+                    self.ap.logger.debug(
+                        f'Stage {stage_container.inst_name} processed query {query.query_id} res {sub_result.result_type}'
+                    )
                     await self._check_output(query, sub_result)
 
                     if sub_result.result_type == pipeline_entities.ResultType.INTERRUPT:
-                        self.ap.logger.debug(f'Stage {stage_container.inst_name} interrupted query {query}')
+                        self.ap.logger.debug(f'Stage {stage_container.inst_name} interrupted query {query.query_id}')
                         break
                     elif sub_result.result_type == pipeline_entities.ResultType.CONTINUE:
                         query = sub_result.new_query
@@ -192,7 +204,7 @@ class RuntimePipeline:
             if event_ctx.is_prevented_default():
                 return
 
-            self.ap.logger.debug(f'Processing query {query}')
+            self.ap.logger.debug(f'Processing query {query.query_id}')
 
             await self._execute_from_stage(0, query)
         except Exception as e:
@@ -200,7 +212,7 @@ class RuntimePipeline:
             self.ap.logger.error(f'处理请求时出错 query_id={query.query_id} stage={inst_name} : {e}')
             self.ap.logger.error(f'Traceback: {traceback.format_exc()}')
         finally:
-            self.ap.logger.debug(f'Query {query} processed')
+            self.ap.logger.debug(f'Query {query.query_id} processed')
 
 
 class PipelineManager:

pkg/pipeline/preproc/preproc.py

@@ -80,14 +80,15 @@ class PreProcessor(stage.PipelineStage):
                         if me.type == 'image_url':
                             msg.content.remove(me)
 
-        content_list = []
+        content_list: list[llm_entities.ContentElement] = []
 
         plain_text = ''
         qoute_msg = query.pipeline_config['trigger'].get('misc', '').get('combine-quote-message')
 
+        # tidy the content_list
+        # combine all text content into one, and put it in the first position
         for me in query.message_chain:
             if isinstance(me, platform_message.Plain):
-                content_list.append(llm_entities.ContentElement.from_text(me.text))
                 plain_text += me.text
             elif isinstance(me, platform_message.Image):
                 if selected_runner != 'local-agent' or query.use_llm_model.model_entity.abilities.__contains__(
@@ -106,6 +107,8 @@ class PreProcessor(stage.PipelineStage):
                             if msg.base64 is not None:
                                 content_list.append(llm_entities.ContentElement.from_image_base64(msg.base64))
 
+        content_list.insert(0, llm_entities.ContentElement.from_text(plain_text))
+
         query.variables['user_message_text'] = plain_text
 
         query.user_message = llm_entities.Message(role='user', content=content_list)

pkg/pipeline/process/handlers/chat.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import uuid
 import typing
 import traceback
 
@@ -22,11 +23,11 @@ class ChatMessageHandler(handler.MessageHandler):
         self,
         query: core_entities.Query,
     ) -> typing.AsyncGenerator[entities.StageProcessResult, None]:
-        """Process"""
-        # Call API
-        #   generator
+        """处理"""
+        # 调API
+        #   生成器
 
-        # Trigger plugin event
+        # 触发插件事件
         event_class = (
             events.PersonNormalMessageReceived
             if query.launcher_type == core_entities.LauncherTypes.PERSON
@@ -46,7 +47,6 @@ class ChatMessageHandler(handler.MessageHandler):
         if event_ctx.is_prevented_default():
             if event_ctx.event.reply is not None:
                 mc = platform_message.MessageChain(event_ctx.event.reply)
-
                 query.resp_messages.append(mc)
 
                 yield entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)
@@ -54,10 +54,14 @@ class ChatMessageHandler(handler.MessageHandler):
                 yield entities.StageProcessResult(result_type=entities.ResultType.INTERRUPT, new_query=query)
         else:
             if event_ctx.event.alter is not None:
-                # if isinstance(event_ctx.event, str):  # Currently not considering multi-modal alter
+                # if isinstance(event_ctx.event, str):  # 现在暂时不考虑多模态alter
                 query.user_message.content = event_ctx.event.alter
 
             text_length = 0
+            try:
+                is_stream = await query.adapter.is_stream_output_supported()
+            except AttributeError:
+                is_stream = False
 
             try:
                 for r in runner_module.preregistered_runners:
@@ -65,22 +69,42 @@ class ChatMessageHandler(handler.MessageHandler):
                         runner = r(self.ap, query.pipeline_config)
                         break
                 else:
-                    raise ValueError(f'Request runner not found: {query.pipeline_config["ai"]["runner"]["runner"]}')
+                    raise ValueError(f'未找到请求运行器: {query.pipeline_config["ai"]["runner"]["runner"]}')
+                if is_stream:
+                    resp_message_id = uuid.uuid4()
+                    await query.adapter.create_message_card(str(resp_message_id), query.message_event)
+                    async for result in runner.run(query):
+                        result.resp_message_id = str(resp_message_id)
+                        if query.resp_messages:
+                            query.resp_messages.pop()
+                        if query.resp_message_chain:
+                            query.resp_message_chain.pop()
 
-                async for result in runner.run(query):
-                    query.resp_messages.append(result)
+                        query.resp_messages.append(result)
+                        self.ap.logger.info(f'对话({query.query_id})流式响应: {self.cut_str(result.readable_str())}')
 
-                    self.ap.logger.info(f'Response({query.query_id}): {self.cut_str(result.readable_str())}')
+                        if result.content is not None:
+                            text_length += len(result.content)
 
-                    if result.content is not None:
-                        text_length += len(result.content)
+                        yield entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)
 
-                    yield entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)
+                else:
+                    async for result in runner.run(query):
+                        query.resp_messages.append(result)
+
+                        self.ap.logger.info(f'对话({query.query_id})响应: {self.cut_str(result.readable_str())}')
+
+                        if result.content is not None:
+                            text_length += len(result.content)
+
+                        yield entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)
 
                 query.session.using_conversation.messages.append(query.user_message)
+
                 query.session.using_conversation.messages.extend(query.resp_messages)
             except Exception as e:
-                self.ap.logger.error(f'Request failed({query.query_id}): {type(e).__name__} {str(e)}')
+                self.ap.logger.error(f'对话({query.query_id})请求失败: {type(e).__name__} {str(e)}')
+                traceback.print_exc()
 
                 hide_exception_info = query.pipeline_config['output']['misc']['hide-exception']
 
@@ -93,4 +117,4 @@ class ChatMessageHandler(handler.MessageHandler):
                 )
             finally:
                 # TODO statistics
-                pass
+                pass

pkg/pipeline/respback/respback.py

@@ -7,6 +7,10 @@ import asyncio
 from ...platform.types import events as platform_events
 from ...platform.types import message as platform_message
 
+from ...provider import entities as llm_entities
+
+
+
 from .. import stage, entities
 from ...core import entities as core_entities
 
@@ -36,10 +40,22 @@ class SendResponseBackStage(stage.PipelineStage):
 
         quote_origin = query.pipeline_config['output']['misc']['quote-origin']
 
-        await query.adapter.reply_message(
-            message_source=query.message_event,
-            message=query.resp_message_chain[-1],
-            quote_origin=quote_origin,
-        )
+        has_chunks = any(isinstance(msg, llm_entities.MessageChunk) for msg in query.resp_messages)
+        # TODO 命令与流式的兼容性问题
+        if await query.adapter.is_stream_output_supported() and has_chunks:
+            is_final = [msg.is_final for msg in query.resp_messages][0]
+            await query.adapter.reply_message_chunk(
+                message_source=query.message_event,
+                bot_message=query.resp_messages[-1],
+                message=query.resp_message_chain[-1],
+                quote_origin=quote_origin,
+                is_final=is_final,
+            )
+        else:
+            await query.adapter.reply_message(
+                message_source=query.message_event,
+                message=query.resp_message_chain[-1],
+                quote_origin=quote_origin,
+            )
 
         return entities.StageProcessResult(result_type=entities.ResultType.CONTINUE, new_query=query)

pkg/platform/adapter.py

@@ -61,14 +61,40 @@ class MessagePlatformAdapter(metaclass=abc.ABCMeta):
         """
         raise NotImplementedError
 
+    async def reply_message_chunk(
+        self,
+        message_source: platform_events.MessageEvent,
+        bot_message: dict,
+        message: platform_message.MessageChain,
+        quote_origin: bool = False,
+        is_final: bool = False,
+    ):
+        """回复消息（流式输出）
+        Args:
+            message_source (platform.types.MessageEvent): 消息源事件
+            message_id (int): 消息ID
+            message (platform.types.MessageChain): 消息链
+            quote_origin (bool, optional): 是否引用原消息. Defaults to False.
+            is_final (bool, optional): 流式是否结束. Defaults to False.
+        """
+        raise NotImplementedError
+
+    async def create_message_card(self, message_id: typing.Type[str, int], event: platform_events.MessageEvent) -> bool:
+        """创建卡片消息
+        Args:
+            message_id (str): 消息ID
+            event (platform_events.MessageEvent): 消息源事件
+        """
+        return False
+
     async def is_muted(self, group_id: int) -> bool:
         """获取账号是否在指定群被禁言"""
         raise NotImplementedError
 
     def register_listener(
         self,
-        event_type: typing.Type[platform_message.Event],
-        callback: typing.Callable[[platform_message.Event, MessagePlatformAdapter], None],
+        event_type: typing.Type[platform_events.Event],
+        callback: typing.Callable[[platform_events.Event, MessagePlatformAdapter], None],
     ):
         """注册事件监听器
 
@@ -80,8 +106,8 @@ class MessagePlatformAdapter(metaclass=abc.ABCMeta):
 
     def unregister_listener(
         self,
-        event_type: typing.Type[platform_message.Event],
-        callback: typing.Callable[[platform_message.Event, MessagePlatformAdapter], None],
+        event_type: typing.Type[platform_events.Event],
+        callback: typing.Callable[[platform_events.Event, MessagePlatformAdapter], None],
     ):
         """注销事件监听器
 
@@ -95,6 +121,10 @@ class MessagePlatformAdapter(metaclass=abc.ABCMeta):
         """异步运行"""
         raise NotImplementedError
 
+    async def is_stream_output_supported(self) -> bool:
+        """是否支持流式输出"""
+        return False
+
     async def kill(self) -> bool:
         """关闭适配器
 
@@ -136,7 +166,7 @@ class EventConverter:
     """事件转换器基类"""
 
     @staticmethod
-    def yiri2target(event: typing.Type[platform_message.Event]):
+    def yiri2target(event: typing.Type[platform_events.Event]):
         """将源平台事件转换为目标平台事件
 
         Args:
@@ -148,7 +178,7 @@ class EventConverter:
         raise NotImplementedError
 
     @staticmethod
-    def target2yiri(event: typing.Any) -> platform_message.Event:
+    def target2yiri(event: typing.Any) -> platform_events.Event:
         """将目标平台事件的调用参数转换为源平台的事件参数对象
 
         Args:

pkg/platform/botmgr.py

@@ -120,8 +120,10 @@ class RuntimeBot:
                 if isinstance(e, asyncio.CancelledError):
                     self.task_context.set_current_action('Exited.')
                     return
+
+                traceback_str = traceback.format_exc()
                 self.task_context.set_current_action('Exited with error.')
-                await self.logger.error(f'平台适配器运行出错:\n{e}\n{traceback.format_exc()}')
+                await self.logger.error(f'平台适配器运行出错:\n{e}\n{traceback_str}')
 
         self.task_wrapper = self.ap.task_mgr.create_task(
             exception_wrapper(),

pkg/platform/logger.py

@@ -119,7 +119,7 @@ class EventLogger:
     async def _truncate_logs(self):
         if len(self.logs) > MAX_LOG_COUNT:
             for i in range(DELETE_COUNT_PER_TIME):
-                for image_key in self.logs[i].images:
+                for image_key in self.logs[i].images:  # type: ignore
                     await self.ap.storage_mgr.storage_provider.delete(image_key)
             self.logs = self.logs[DELETE_COUNT_PER_TIME:]
 

pkg/platform/sources/aiocqhttp.py

@@ -266,7 +266,7 @@ class AiocqhttpMessageConverter(adapter.MessageConverter):
                     await process_message_data(msg_data, reply_list)
 
                 reply_msg = platform_message.Quote(
-                    message_id=msg.data['id'], sender_id=msg_datas['user_id'], origin=reply_list
+                    message_id=msg.data['id'], sender_id=msg_datas['sender']['user_id'], origin=reply_list
                 )
                 yiri_msg_list.append(reply_msg)
 

pkg/platform/sources/dingtalk.py

@@ -1,3 +1,4 @@
+from re import S
 import traceback
 import typing
 from libs.dingtalk_api.dingtalkevent import DingTalkEvent
@@ -99,11 +100,15 @@ class DingTalkAdapter(adapter.MessagePlatformAdapter):
     message_converter: DingTalkMessageConverter = DingTalkMessageConverter()
     event_converter: DingTalkEventConverter = DingTalkEventConverter()
     config: dict
+    card_instance_id_dict: dict  # 回复卡片消息字典，key为消息id，value为回复卡片实例id，用于在流式消息时判断是否发送到指定卡片
+    seq: int  # 消息顺序，直接以seq作为标识
 
     def __init__(self, config: dict, ap: app.Application, logger: EventLogger):
         self.config = config
         self.ap = ap
         self.logger = logger
+        self.card_instance_id_dict = {}
+        # self.seq = 1
         required_keys = [
             'client_id',
             'client_secret',
@@ -139,6 +144,34 @@ class DingTalkAdapter(adapter.MessagePlatformAdapter):
         content, at = await DingTalkMessageConverter.yiri2target(message)
         await self.bot.send_message(content, incoming_message, at)
 
+    async def reply_message_chunk(
+        self,
+        message_source: platform_events.MessageEvent,
+        bot_message,
+        message: platform_message.MessageChain,
+        quote_origin: bool = False,
+        is_final: bool = False,
+    ):
+        # event = await DingTalkEventConverter.yiri2target(
+        #     message_source,
+        # )
+        # incoming_message = event.incoming_message
+
+        # msg_id = incoming_message.message_id
+        message_id = bot_message.resp_message_id
+        msg_seq = bot_message.msg_sequence
+
+        if (msg_seq - 1) % 8 == 0 or is_final:
+
+            content, at = await DingTalkMessageConverter.yiri2target(message)
+
+            card_instance, card_instance_id = self.card_instance_id_dict[message_id]
+            # print(card_instance_id)
+            await self.bot.send_card_message(card_instance, card_instance_id, content, is_final)
+            if is_final and bot_message.tool_calls is None:
+                # self.seq = 1  # 消息回复结束之后重置seq
+                self.card_instance_id_dict.pop(message_id)  # 消息回复结束之后删除卡片实例id
+
     async def send_message(self, target_type: str, target_id: str, message: platform_message.MessageChain):
         content = await DingTalkMessageConverter.yiri2target(message)
         if target_type == 'person':
@@ -146,6 +179,20 @@ class DingTalkAdapter(adapter.MessagePlatformAdapter):
         if target_type == 'group':
             await self.bot.send_proactive_message_to_group(target_id, content)
 
+    async def is_stream_output_supported(self) -> bool:
+        is_stream = False
+        if self.config.get('enable-stream-reply', None):
+            is_stream = True
+        return is_stream
+
+    async def create_message_card(self, message_id, event):
+        card_template_id = self.config['card_template_id']
+        incoming_message = event.source_platform_object.incoming_message
+        # message_id = incoming_message.message_id
+        card_instance, card_instance_id = await self.bot.create_and_card(card_template_id, incoming_message)
+        self.card_instance_id_dict[message_id] = (card_instance, card_instance_id)
+        return True
+
     def register_listener(
         self,
         event_type: typing.Type[platform_events.Event],

pkg/platform/sources/dingtalk.yaml

@@ -46,6 +46,23 @@ spec:
       type: boolean
       required: false
       default: true
+    - name: enable-stream-reply
+      label:
+        en_US: Enable Stream Reply Mode
+        zh_Hans: 启用钉钉卡片流式回复模式
+      description:
+        en_US: If enabled, the bot will use the stream of lark reply mode
+        zh_Hans: 如果启用，将使用钉钉卡片流式方式来回复内容
+      type: boolean
+      required: true
+      default: false
+    - name: card_template_id
+      label:
+        en_US: card template id
+        zh_Hans: 卡片模板ID
+      type: string
+      required: true
+      default: "填写你的卡片template_id"
 execution:
   python:
     path: ./dingtalk.py

pkg/platform/sources/lark.py

@@ -17,6 +17,7 @@ import aiohttp
 import lark_oapi.ws.exception
 import quart
 from lark_oapi.api.im.v1 import *
+from lark_oapi.api.cardkit.v1 import *
 
 from .. import adapter
 from ...core import app
@@ -320,6 +321,10 @@ class LarkEventConverter(adapter.EventConverter):
             )
 
 
+CARD_ID_CACHE_SIZE = 500
+CARD_ID_CACHE_MAX_LIFETIME = 20 * 60  # 20分钟
+
+
 class LarkAdapter(adapter.MessagePlatformAdapter):
     bot: lark_oapi.ws.Client
     api_client: lark_oapi.Client
@@ -339,12 +344,20 @@ class LarkAdapter(adapter.MessagePlatformAdapter):
     quart_app: quart.Quart
     ap: app.Application
 
+
+    card_id_dict: dict[str, str]  # 消息id到卡片id的映射，便于创建卡片后的发送消息到指定卡片
+
+    seq: int  # 用于在发送卡片消息中识别消息顺序，直接以seq作为标识
+
     def __init__(self, config: dict, ap: app.Application, logger: EventLogger):
         self.config = config
         self.ap = ap
         self.logger = logger
         self.quart_app = quart.Quart(__name__)
         self.listeners = {}
+        self.card_id_dict = {}
+        self.seq = 1
+
 
         @self.quart_app.route('/lark/callback', methods=['POST'])
         async def lark_callback():
@@ -378,15 +391,15 @@ class LarkAdapter(adapter.MessagePlatformAdapter):
                 if 'im.message.receive_v1' == type:
                     try:
                         event = await self.event_converter.target2yiri(p2v1, self.api_client)
-                    except Exception as e:
-                        await self.logger.error(f"Error in lark callback: {traceback.format_exc()}")
+                    except Exception:
+                        await self.logger.error(f'Error in lark callback: {traceback.format_exc()}')
 
                     if event.__class__ in self.listeners:
                         await self.listeners[event.__class__](event, self)
 
                 return {'code': 200, 'message': 'ok'}
-            except Exception as e:
-                await self.logger.error(f"Error in lark callback: {traceback.format_exc()}")
+            except Exception:
+                await self.logger.error(f'Error in lark callback: {traceback.format_exc()}')
                 return {'code': 500, 'message': 'error'}
 
         async def on_message(event: lark_oapi.im.v1.P2ImMessageReceiveV1):
@@ -409,6 +422,216 @@ class LarkAdapter(adapter.MessagePlatformAdapter):
     async def send_message(self, target_type: str, target_id: str, message: platform_message.MessageChain):
         pass
 
+    async def is_stream_output_supported(self) -> bool:
+        is_stream = False
+        if self.config.get('enable-stream-reply', None):
+            is_stream = True
+        return is_stream
+
+    async def create_card_id(self, message_id):
+        try:
+            self.ap.logger.debug('飞书支持stream输出,创建卡片......')
+
+            card_data = {"schema": "2.0", "config": {"update_multi": True, "streaming_mode": True,
+                                                     "streaming_config": {"print_step": {"default": 1},
+                                                                          "print_frequency_ms": {"default": 70},
+                                                                          "print_strategy": "fast"}},
+                         "body": {"direction": "vertical", "padding": "12px 12px 12px 12px", "elements": [{"tag": "div",
+                                                                                                           "text": {
+                                                                                                               "tag": "plain_text",
+                                                                                                               "content": "LangBot",
+                                                                                                               "text_size": "normal",
+                                                                                                               "text_align": "left",
+                                                                                                               "text_color": "default"},
+                                                                                                           "icon": {
+                                                                                                               "tag": "custom_icon",
+                                                                                                               "img_key": "img_v3_02p3_05c65d5d-9bad-440a-a2fb-c89571bfd5bg"}},
+                                                                                                          {
+                                                                                                              "tag": "markdown",
+                                                                                                              "content": "",
+                                                                                                              "text_align": "left",
+                                                                                                              "text_size": "normal",
+                                                                                                              "margin": "0px 0px 0px 0px",
+                                                                                                              "element_id": "streaming_txt"},
+                                                                                                          {
+                                                                                                              "tag": "markdown",
+                                                                                                              "content": "",
+                                                                                                              "text_align": "left",
+                                                                                                              "text_size": "normal",
+                                                                                                              "margin": "0px 0px 0px 0px"},
+                                                                                                          {
+                                                                                                              "tag": "column_set",
+                                                                                                              "horizontal_spacing": "8px",
+                                                                                                              "horizontal_align": "left",
+                                                                                                              "columns": [
+                                                                                                                  {
+                                                                                                                      "tag": "column",
+                                                                                                                      "width": "weighted",
+                                                                                                                      "elements": [
+                                                                                                                          {
+                                                                                                                              "tag": "markdown",
+                                                                                                                              "content": "",
+                                                                                                                              "text_align": "left",
+                                                                                                                              "text_size": "normal",
+                                                                                                                              "margin": "0px 0px 0px 0px"},
+                                                                                                                          {
+                                                                                                                              "tag": "markdown",
+                                                                                                                              "content": "",
+                                                                                                                              "text_align": "left",
+                                                                                                                              "text_size": "normal",
+                                                                                                                              "margin": "0px 0px 0px 0px"},
+                                                                                                                          {
+                                                                                                                              "tag": "markdown",
+                                                                                                                              "content": "",
+                                                                                                                              "text_align": "left",
+                                                                                                                              "text_size": "normal",
+                                                                                                                              "margin": "0px 0px 0px 0px"}],
+                                                                                                                      "padding": "0px 0px 0px 0px",
+                                                                                                                      "direction": "vertical",
+                                                                                                                      "horizontal_spacing": "8px",
+                                                                                                                      "vertical_spacing": "2px",
+                                                                                                                      "horizontal_align": "left",
+                                                                                                                      "vertical_align": "top",
+                                                                                                                      "margin": "0px 0px 0px 0px",
+                                                                                                                      "weight": 1}],
+                                                                                                              "margin": "0px 0px 0px 0px"},
+                                                                                                          {"tag": "hr",
+                                                                                                           "margin": "0px 0px 0px 0px"},
+                                                                                                          {
+                                                                                                              "tag": "column_set",
+                                                                                                              "horizontal_spacing": "12px",
+                                                                                                              "horizontal_align": "right",
+                                                                                                              "columns": [
+                                                                                                                  {
+                                                                                                                      "tag": "column",
+                                                                                                                      "width": "weighted",
+                                                                                                                      "elements": [
+                                                                                                                          {
+                                                                                                                              "tag": "markdown",
+                                                                                                                              "content": "<font color=\"grey-600\">以上内容由 AI 生成，仅供参考。更多详细、准确信息可点击引用链接查看</font>",
+                                                                                                                              "text_align": "left",
+                                                                                                                              "text_size": "notation",
+                                                                                                                              "margin": "4px 0px 0px 0px",
+                                                                                                                              "icon": {
+                                                                                                                                  "tag": "standard_icon",
+                                                                                                                                  "token": "robot_outlined",
+                                                                                                                                  "color": "grey"}}],
+                                                                                                                      "padding": "0px 0px 0px 0px",
+                                                                                                                      "direction": "vertical",
+                                                                                                                      "horizontal_spacing": "8px",
+                                                                                                                      "vertical_spacing": "8px",
+                                                                                                                      "horizontal_align": "left",
+                                                                                                                      "vertical_align": "top",
+                                                                                                                      "margin": "0px 0px 0px 0px",
+                                                                                                                      "weight": 1},
+                                                                                                                  {
+                                                                                                                      "tag": "column",
+                                                                                                                      "width": "20px",
+                                                                                                                      "elements": [
+                                                                                                                          {
+                                                                                                                              "tag": "button",
+                                                                                                                              "text": {
+                                                                                                                                  "tag": "plain_text",
+                                                                                                                                  "content": ""},
+                                                                                                                              "type": "text",
+                                                                                                                              "width": "fill",
+                                                                                                                              "size": "medium",
+                                                                                                                              "icon": {
+                                                                                                                                  "tag": "standard_icon",
+                                                                                                                                  "token": "thumbsup_outlined"},
+                                                                                                                              "hover_tips": {
+                                                                                                                                  "tag": "plain_text",
+                                                                                                                                  "content": "有帮助"},
+                                                                                                                              "margin": "0px 0px 0px 0px"}],
+                                                                                                                      "padding": "0px 0px 0px 0px",
+                                                                                                                      "direction": "vertical",
+                                                                                                                      "horizontal_spacing": "8px",
+                                                                                                                      "vertical_spacing": "8px",
+                                                                                                                      "horizontal_align": "left",
+                                                                                                                      "vertical_align": "top",
+                                                                                                                      "margin": "0px 0px 0px 0px"},
+                                                                                                                  {
+                                                                                                                      "tag": "column",
+                                                                                                                      "width": "30px",
+                                                                                                                      "elements": [
+                                                                                                                          {
+                                                                                                                              "tag": "button",
+                                                                                                                              "text": {
+                                                                                                                                  "tag": "plain_text",
+                                                                                                                                  "content": ""},
+                                                                                                                              "type": "text",
+                                                                                                                              "width": "default",
+                                                                                                                              "size": "medium",
+                                                                                                                              "icon": {
+                                                                                                                                  "tag": "standard_icon",
+                                                                                                                                  "token": "thumbdown_outlined"},
+                                                                                                                              "hover_tips": {
+                                                                                                                                  "tag": "plain_text",
+                                                                                                                                  "content": "无帮助"},
+                                                                                                                              "margin": "0px 0px 0px 0px"}],
+                                                                                                                      "padding": "0px 0px 0px 0px",
+                                                                                                                      "vertical_spacing": "8px",
+                                                                                                                      "horizontal_align": "left",
+                                                                                                                      "vertical_align": "top",
+                                                                                                                      "margin": "0px 0px 0px 0px"}],
+                                                                                                              "margin": "0px 0px 4px 0px"}]}}
+            # delay / fast 创建卡片模板，delay 延迟打印，fast 实时打印，可以自定义更好看的消息模板
+
+            request: CreateCardRequest = (
+                CreateCardRequest.builder()
+                .request_body(CreateCardRequestBody.builder().type('card_json').data(json.dumps(card_data)).build())
+                .build()
+            )
+
+            # 发起请求
+            response: CreateCardResponse = self.api_client.cardkit.v1.card.create(request)
+
+            # 处理失败返回
+            if not response.success():
+                raise Exception(
+                    f'client.cardkit.v1.card.create failed, code: {response.code}, msg: {response.msg}, log_id: {response.get_log_id()}, resp: \n{json.dumps(json.loads(response.raw.content), indent=4, ensure_ascii=False)}'
+                )
+
+            self.ap.logger.debug(f'飞书卡片创建成功,卡片ID: {response.data.card_id}')
+            self.card_id_dict[message_id] = response.data.card_id
+
+            card_id = response.data.card_id
+            return card_id
+
+        except Exception as e:
+            self.ap.logger.error(f'飞书卡片创建失败,错误信息: {e}')
+
+    async def create_message_card(self, message_id, event) -> str:
+        """
+        创建卡片消息。
+        使用卡片消息是因为普通消息更新次数有限制，而大模型流式返回结果可能很多而超过限制，而飞书卡片没有这个限制（api免费次数有限）
+        """
+        # message_id = event.message_chain.message_id
+
+        card_id = await self.create_card_id(message_id)
+        content = {
+            'type': 'card',
+            'data': {'card_id': card_id, 'template_variable': {'content': 'Thinking...'}},
+        }   # 当收到消息时发送消息模板，可添加模板变量，详情查看飞书中接口文档
+        request: ReplyMessageRequest = (
+            ReplyMessageRequest.builder()
+            .message_id(event.message_chain.message_id)
+            .request_body(
+                ReplyMessageRequestBody.builder().content(json.dumps(content)).msg_type('interactive').build()
+            )
+            .build()
+        )
+
+        # 发起请求
+        response: ReplyMessageResponse = await self.api_client.im.v1.message.areply(request)
+
+        # 处理失败返回
+        if not response.success():
+            raise Exception(
+                f'client.im.v1.message.reply failed, code: {response.code}, msg: {response.msg}, log_id: {response.get_log_id()}, resp: \n{json.dumps(json.loads(response.raw.content), indent=4, ensure_ascii=False)}'
+            )
+        return True
+
     async def reply_message(
         self,
         message_source: platform_events.MessageEvent,
@@ -447,6 +670,64 @@ class LarkAdapter(adapter.MessagePlatformAdapter):
                 f'client.im.v1.message.reply failed, code: {response.code}, msg: {response.msg}, log_id: {response.get_log_id()}, resp: \n{json.dumps(json.loads(response.raw.content), indent=4, ensure_ascii=False)}'
             )
 
+    async def reply_message_chunk(
+        self,
+        message_source: platform_events.MessageEvent,
+        bot_message,
+        message: platform_message.MessageChain,
+        quote_origin: bool = False,
+        is_final: bool = False,
+    ):
+        """
+        回复消息变成更新卡片消息
+        """
+        # self.seq += 1
+        message_id = bot_message.resp_message_id
+        msg_seq = bot_message.msg_sequence
+        if msg_seq % 8 == 0 or is_final:
+
+            lark_message = await self.message_converter.yiri2target(message, self.api_client)
+
+
+            text_message = ''
+            for ele in lark_message[0]:
+                if ele['tag'] == 'text':
+                    text_message += ele['text']
+                elif ele['tag'] == 'md':
+                    text_message += ele['text']
+
+            # content = {
+            #     'type': 'card_json',
+            #     'data': {'card_id': self.card_id_dict[message_id], 'elements': {'content': text_message}},
+            # }
+
+            request: ContentCardElementRequest = (
+                ContentCardElementRequest.builder()
+                .card_id(self.card_id_dict[message_id])
+                .element_id('streaming_txt')
+                .request_body(
+                    ContentCardElementRequestBody.builder()
+                    # .uuid("a0d69e20-1dd1-458b-k525-dfeca4015204")
+                    .content(text_message)
+                    .sequence(msg_seq)
+                    .build()
+                )
+                .build()
+            )
+
+            if is_final and bot_message.tool_calls is None:
+                # self.seq = 1  # 消息回复结束之后重置seq
+                self.card_id_dict.pop(message_id)  # 清理已经使用过的卡片
+            # 发起请求
+            response: ContentCardElementResponse = self.api_client.cardkit.v1.card_element.content(request)
+
+            # 处理失败返回
+            if not response.success():
+                raise Exception(
+                    f'client.im.v1.message.patch failed, code: {response.code}, msg: {response.msg}, log_id: {response.get_log_id()}, resp: \n{json.dumps(json.loads(response.raw.content), indent=4, ensure_ascii=False)}'
+                )
+                return
+
     async def is_muted(self, group_id: int) -> bool:
         return False
 
@@ -492,4 +773,9 @@ class LarkAdapter(adapter.MessagePlatformAdapter):
             )
 
     async def kill(self) -> bool:
-        return False
+        # 需要断开连接，不然旧的连接会继续运行，导致飞书消息来时会随机选择一个连接
+        # 断开时lark.ws.Client的_receive_message_loop会打印error日志: receive message loop exit。然后进行重连，
+        # 所以要设置_auto_reconnect=False,让其不重连。
+        self.bot._auto_reconnect = False
+        await self.bot._disconnect()
+        return False

pkg/platform/sources/lark.yaml

@@ -65,6 +65,16 @@ spec:
       type: string
       required: true
       default: ""
+    - name: enable-stream-reply
+      label:
+        en_US: Enable Stream Reply Mode
+        zh_Hans: 启用飞书流式回复模式
+      description:
+        en_US: If enabled, the bot will use the stream of lark reply mode
+        zh_Hans: 如果启用，将使用飞书流式方式来回复内容
+      type: boolean
+      required: true
+      default: false
 execution:
   python:
     path: ./lark.py

pkg/platform/sources/nakuru.py

@@ -72,8 +72,9 @@ class NakuruProjectMessageConverter(adapter_model.MessageConverter):
                             content=content_list,
                         )
                         nakuru_forward_node_list.append(nakuru_forward_node)
-                    except Exception as e:
+                    except Exception:
                         import traceback
+
                         traceback.print_exc()
 
                 nakuru_msg_list.append(nakuru_forward_node_list)
@@ -276,7 +277,7 @@ class NakuruAdapter(adapter_model.MessagePlatformAdapter):
             # 注册监听器
             self.bot.receiver(source_cls.__name__)(listener_wrapper)
         except Exception as e:
-            self.logger.error(f"Error in nakuru register_listener: {traceback.format_exc()}")
+            self.logger.error(f'Error in nakuru register_listener: {traceback.format_exc()}')
             raise e
 
     def unregister_listener(

pkg/platform/sources/officialaccount.py

@@ -125,8 +125,8 @@ class OfficialAccountAdapter(adapter.MessagePlatformAdapter):
             self.bot_account_id = event.receiver_id
             try:
                 return await callback(await self.event_converter.target2yiri(event), self)
-            except Exception as e:
-                await self.logger.error(f"Error in officialaccount callback: {traceback.format_exc()}")
+            except Exception:
+                await self.logger.error(f'Error in officialaccount callback: {traceback.format_exc()}')
 
         if event_type == platform_events.FriendMessage:
             self.bot.on_message('text')(on_message)

pkg/platform/sources/qqbotpy.py

@@ -501,7 +501,7 @@ class OfficialAdapter(adapter_model.MessagePlatformAdapter):
             for event_handler in event_handler_mapping[event_type]:
                 setattr(self.bot, event_handler, wrapper)
         except Exception as e:
-            self.logger.error(f"Error in qqbotpy callback: {traceback.format_exc()}")
+            self.logger.error(f'Error in qqbotpy callback: {traceback.format_exc()}')
             raise e
 
     def unregister_listener(

pkg/platform/sources/qqofficial.py

@@ -154,10 +154,7 @@ class QQOfficialAdapter(adapter.MessagePlatformAdapter):
             raise ParamNotEnoughError('QQ官方机器人缺少相关配置项，请查看文档或联系管理员')
 
         self.bot = QQOfficialClient(
-            app_id=config['appid'],
-            secret=config['secret'],
-            token=config['token'],
-            logger=self.logger
+            app_id=config['appid'], secret=config['secret'], token=config['token'], logger=self.logger
         )
 
     async def reply_message(
@@ -224,8 +221,8 @@ class QQOfficialAdapter(adapter.MessagePlatformAdapter):
             self.bot_account_id = 'justbot'
             try:
                 return await callback(await self.event_converter.target2yiri(event), self)
-            except Exception as e:
-                await self.logger.error(f"Error in qqofficial callback: {traceback.format_exc()}")
+            except Exception:
+                await self.logger.error(f'Error in qqofficial callback: {traceback.format_exc()}')
 
         if event_type == platform_events.FriendMessage:
             self.bot.on_message('DIRECT_MESSAGE_CREATE')(on_message)

pkg/platform/sources/slack.py

@@ -104,7 +104,9 @@ class SlackAdapter(adapter.MessagePlatformAdapter):
         if missing_keys:
             raise ParamNotEnoughError('Slack机器人缺少相关配置项，请查看文档或联系管理员')
 
-        self.bot = SlackClient(bot_token=self.config['bot_token'], signing_secret=self.config['signing_secret'], logger=self.logger)
+        self.bot = SlackClient(
+            bot_token=self.config['bot_token'], signing_secret=self.config['signing_secret'], logger=self.logger
+        )
 
     async def reply_message(
         self,
@@ -139,8 +141,8 @@ class SlackAdapter(adapter.MessagePlatformAdapter):
             self.bot_account_id = 'SlackBot'
             try:
                 return await callback(await self.event_converter.target2yiri(event, self.bot), self)
-            except Exception as e:
-                await self.logger.error(f"Error in slack callback: {traceback.format_exc()}")
+            except Exception:
+                await self.logger.error(f'Error in slack callback: {traceback.format_exc()}')
 
         if event_type == platform_events.FriendMessage:
             self.bot.on_message('im')(on_message)

pkg/platform/sources/telegram.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+
 import telegram
 import telegram.ext
 from telegram import Update
@@ -143,6 +144,10 @@ class TelegramAdapter(adapter.MessagePlatformAdapter):
     config: dict
     ap: app.Application
 
+    msg_stream_id: dict  # 流式消息id字典，key为流式消息id，value为首次消息源id，用于在流式消息时判断编辑那条消息
+
+    seq: int  # 消息中识别消息顺序，直接以seq作为标识
+
     listeners: typing.Dict[
         typing.Type[platform_events.Event],
         typing.Callable[[platform_events.Event, adapter.MessagePlatformAdapter], None],
@@ -152,6 +157,8 @@ class TelegramAdapter(adapter.MessagePlatformAdapter):
         self.config = config
         self.ap = ap
         self.logger = logger
+        self.msg_stream_id = {}
+        # self.seq = 1
 
         async def telegram_callback(update: Update, context: ContextTypes.DEFAULT_TYPE):
             if update.message.from_user.is_bot:
@@ -160,8 +167,9 @@ class TelegramAdapter(adapter.MessagePlatformAdapter):
             try:
                 lb_event = await self.event_converter.target2yiri(update, self.bot, self.bot_account_id)
                 await self.listeners[type(lb_event)](lb_event, self)
-            except Exception as e:
-                await self.logger.error(f"Error in telegram callback: {traceback.format_exc()}")
+                await self.is_stream_output_supported()
+            except Exception:
+                await self.logger.error(f'Error in telegram callback: {traceback.format_exc()}')
 
         self.application = ApplicationBuilder().token(self.config['token']).build()
         self.bot = self.application.bot
@@ -200,6 +208,70 @@ class TelegramAdapter(adapter.MessagePlatformAdapter):
 
         await self.bot.send_message(**args)
 
+    async def reply_message_chunk(
+        self,
+        message_source: platform_events.MessageEvent,
+        bot_message,
+        message: platform_message.MessageChain,
+        quote_origin: bool = False,
+        is_final: bool = False,
+    ):
+        msg_seq = bot_message.msg_sequence
+        if (msg_seq - 1) % 8 == 0 or is_final:
+            assert isinstance(message_source.source_platform_object, Update)
+            components = await TelegramMessageConverter.yiri2target(message, self.bot)
+            args = {}
+            message_id = message_source.source_platform_object.message.id
+            if quote_origin:
+                args['reply_to_message_id'] = message_source.source_platform_object.message.id
+
+            component = components[0]
+            if message_id not in self.msg_stream_id:  # 当消息回复第一次时，发送新消息
+                # time.sleep(0.6)
+                if component['type'] == 'text':
+                    if self.config['markdown_card'] is True:
+                        content = telegramify_markdown.markdownify(
+                            content=component['text'],
+                        )
+                    else:
+                        content = component['text']
+                    args = {
+                        'chat_id': message_source.source_platform_object.effective_chat.id,
+                        'text': content,
+                    }
+                    if self.config['markdown_card'] is True:
+                        args['parse_mode'] = 'MarkdownV2'
+
+                send_msg = await self.bot.send_message(**args)
+                send_msg_id = send_msg.message_id
+                self.msg_stream_id[message_id] = send_msg_id
+            else:  # 存在消息的时候直接编辑消息1
+                if component['type'] == 'text':
+                    if self.config['markdown_card'] is True:
+                        content = telegramify_markdown.markdownify(
+                            content=component['text'],
+                        )
+                    else:
+                        content = component['text']
+                    args = {
+                        'message_id': self.msg_stream_id[message_id],
+                        'chat_id': message_source.source_platform_object.effective_chat.id,
+                        'text': content,
+                    }
+                    if self.config['markdown_card'] is True:
+                        args['parse_mode'] = 'MarkdownV2'
+
+                await self.bot.edit_message_text(**args)
+            if is_final and bot_message.tool_calls is None:
+                # self.seq = 1  # 消息回复结束之后重置seq
+                self.msg_stream_id.pop(message_id)  # 消息回复结束之后删除流式消息id
+
+    async def is_stream_output_supported(self) -> bool:
+        is_stream = False
+        if self.config.get('enable-stream-reply', None):
+            is_stream = True
+        return is_stream
+
     async def is_muted(self, group_id: int) -> bool:
         return False
 
@@ -222,8 +294,12 @@ class TelegramAdapter(adapter.MessagePlatformAdapter):
         self.bot_account_id = (await self.bot.get_me()).username
         await self.application.updater.start_polling(allowed_updates=Update.ALL_TYPES)
         await self.application.start()
+        await self.logger.info('Telegram adapter running')
 
     async def kill(self) -> bool:
         if self.application.running:
             await self.application.stop()
+            if self.application.updater:
+                await self.application.updater.stop()
+            await self.logger.info('Telegram adapter stopped')
         return True

pkg/platform/sources/telegram.yaml

@@ -25,6 +25,16 @@ spec:
       type: boolean
       required: false
       default: true
+    - name: enable-stream-reply
+      label:
+        en_US: Enable Stream Reply Mode
+        zh_Hans: 启用电报流式回复模式
+      description:
+        en_US: If enabled, the bot will use the stream of telegram reply mode
+        zh_Hans: 如果启用，将使用电报流式方式来回复内容
+      type: boolean
+      required: true
+      default: false
 execution:
   python:
     path: ./telegram.py

pkg/platform/sources/webchat.py

@@ -19,17 +19,20 @@ class WebChatMessage(BaseModel):
     content: str
     message_chain: list[dict]
     timestamp: str
+    is_final: bool = False
 
 
 class WebChatSession:
     id: str
     message_lists: dict[str, list[WebChatMessage]] = {}
     resp_waiters: dict[int, asyncio.Future[WebChatMessage]]
+    resp_queues: dict[int, asyncio.Queue[WebChatMessage]]
 
     def __init__(self, id: str):
         self.id = id
         self.message_lists = {}
         self.resp_waiters = {}
+        self.resp_queues = {}
 
     def get_message_list(self, pipeline_uuid: str) -> list[WebChatMessage]:
         if pipeline_uuid not in self.message_lists:
@@ -49,6 +52,8 @@ class WebChatAdapter(msadapter.MessagePlatformAdapter):
         typing.Callable[[platform_events.Event, msadapter.MessagePlatformAdapter], None],
     ] = {}
 
+    is_stream: bool
+
     def __init__(self, config: dict, ap: app.Application, logger: EventLogger):
         self.ap = ap
         self.logger = logger
@@ -59,6 +64,8 @@ class WebChatAdapter(msadapter.MessagePlatformAdapter):
 
         self.bot_account_id = 'webchatbot'
 
+        self.is_stream = False
+
     async def send_message(
         self,
         target_type: str,
@@ -102,12 +109,53 @@ class WebChatAdapter(msadapter.MessagePlatformAdapter):
 
         # notify waiter
         if isinstance(message_source, platform_events.FriendMessage):
-            self.webchat_person_session.resp_waiters[message_source.message_chain.message_id].set_result(message_data)
+            await self.webchat_person_session.resp_queues[message_source.message_chain.message_id].put(message_data)
         elif isinstance(message_source, platform_events.GroupMessage):
-            self.webchat_group_session.resp_waiters[message_source.message_chain.message_id].set_result(message_data)
+            await self.webchat_group_session.resp_queues[message_source.message_chain.message_id].put(message_data)
 
         return message_data.model_dump()
 
+    async def reply_message_chunk(
+        self,
+        message_source: platform_events.MessageEvent,
+        bot_message,
+        message: platform_message.MessageChain,
+        quote_origin: bool = False,
+        is_final: bool = False,
+    ) -> dict:
+        """回复消息"""
+        message_data = WebChatMessage(
+            id=-1,
+            role='assistant',
+            content=str(message),
+            message_chain=[component.__dict__ for component in message],
+            timestamp=datetime.now().isoformat(),
+        )
+
+        # notify waiter
+        session = (
+            self.webchat_group_session
+            if isinstance(message_source, platform_events.GroupMessage)
+            else self.webchat_person_session
+        )
+        if message_source.message_chain.message_id not in session.resp_waiters:
+            # session.resp_waiters[message_source.message_chain.message_id] = asyncio.Queue()
+            queue = session.resp_queues[message_source.message_chain.message_id]
+
+        # if isinstance(message_source, platform_events.FriendMessage):
+        #     queue = self.webchat_person_session.resp_queues[message_source.message_chain.message_id]
+        # elif isinstance(message_source, platform_events.GroupMessage):
+        #     queue = self.webchat_group_session.resp_queues[message_source.message_chain.message_id]
+        if is_final and bot_message.tool_calls is None:
+            message_data.is_final = True
+        # print(message_data)
+        await queue.put(message_data)
+
+        return message_data.model_dump()
+
+    async def is_stream_output_supported(self) -> bool:
+        return self.is_stream
+
     def register_listener(
         self,
         event_type: typing.Type[platform_events.Event],
@@ -140,8 +188,13 @@ class WebChatAdapter(msadapter.MessagePlatformAdapter):
         await self.logger.info('WebChat调试适配器正在停止')
 
     async def send_webchat_message(
-        self, pipeline_uuid: str, session_type: str, message_chain_obj: typing.List[dict]
+        self,
+        pipeline_uuid: str,
+        session_type: str,
+        message_chain_obj: typing.List[dict],
+        is_stream: bool = False,
     ) -> dict:
+        self.is_stream = is_stream
         """发送调试消息到流水线"""
         if session_type == 'person':
             use_session = self.webchat_person_session
@@ -152,6 +205,9 @@ class WebChatAdapter(msadapter.MessagePlatformAdapter):
 
         message_id = len(use_session.get_message_list(pipeline_uuid)) + 1
 
+        use_session.resp_queues[message_id] = asyncio.Queue()
+        logger.debug(f'Initialized queue for message_id: {message_id}')
+
         use_session.get_message_list(pipeline_uuid).append(
             WebChatMessage(
                 id=message_id,
@@ -185,21 +241,46 @@ class WebChatAdapter(msadapter.MessagePlatformAdapter):
 
         self.ap.platform_mgr.webchat_proxy_bot.bot_entity.use_pipeline_uuid = pipeline_uuid
 
+        # trigger pipeline
         if event.__class__ in self.listeners:
             await self.listeners[event.__class__](event, self)
 
-        # set waiter
-        waiter = asyncio.Future[WebChatMessage]()
-        use_session.resp_waiters[message_id] = waiter
-        waiter.add_done_callback(lambda future: use_session.resp_waiters.pop(message_id))
+        if is_stream:
+            queue = use_session.resp_queues[message_id]
+            msg_id = len(use_session.get_message_list(pipeline_uuid)) + 1
+            while True:
+                resp_message = await queue.get()
+                resp_message.id = msg_id
+                if resp_message.is_final:
+                    resp_message.id = msg_id
+                    use_session.get_message_list(pipeline_uuid).append(resp_message)
+                    yield resp_message.model_dump()
+                    break
+                yield resp_message.model_dump()
+            use_session.resp_queues.pop(message_id)
 
-        resp_message = await waiter
+        else:  # non-stream
+            # set waiter
+            # waiter = asyncio.Future[WebChatMessage]()
+            # use_session.resp_waiters[message_id] = waiter
+            # # waiter.add_done_callback(lambda future: use_session.resp_waiters.pop(message_id))
+            #
+            # resp_message = await waiter
+            #
+            # resp_message.id = len(use_session.get_message_list(pipeline_uuid)) + 1
+            #
+            # use_session.get_message_list(pipeline_uuid).append(resp_message)
+            #
+            # yield resp_message.model_dump()
+            msg_id = len(use_session.get_message_list(pipeline_uuid)) + 1
 
-        resp_message.id = len(use_session.get_message_list(pipeline_uuid)) + 1
+            queue = use_session.resp_queues[message_id]
+            resp_message = await queue.get()
+            use_session.get_message_list(pipeline_uuid).append(resp_message)
+            resp_message.id = msg_id
+            resp_message.is_final = True
 
-        use_session.get_message_list(pipeline_uuid).append(resp_message)
-
-        return resp_message.model_dump()
+            yield resp_message.model_dump()
 
     def get_webchat_messages(self, pipeline_uuid: str, session_type: str) -> list[dict]:
         """获取调试消息历史"""

pkg/platform/sources/webchat.yaml

@@ -9,7 +9,8 @@ metadata:
     en_US: "WebChat adapter for pipeline debugging"
     zh_Hans: "用于流水线调试的网页聊天适配器"
   icon: ""
-spec: {}
+spec:
+  config: []
 execution:
   python:
     path: "webchat.py"

pkg/platform/sources/wechatpad.py

@@ -1,5 +1,4 @@
 import requests
-import websockets
 import websocket
 import json
 import time
@@ -10,53 +9,42 @@ from libs.wechatpad_api.client import WeChatPadClient
 import typing
 import asyncio
 import traceback
-import time
 import re
 import base64
-import uuid
-import json
-import os
 import copy
-import datetime
 import threading
 
 import quart
-import aiohttp
 
 from .. import adapter
-from ...pipeline.longtext.strategies import forward
 from ...core import app
 from ..types import message as platform_message
 from ..types import events as platform_events
 from ..types import entities as platform_entities
-from ...utils import image
 from ..logger import EventLogger
 import xml.etree.ElementTree as ET
-from typing import Optional, List, Tuple
+from typing import Optional, Tuple
 from functools import partial
 import logging
 
-class WeChatPadMessageConverter(adapter.MessageConverter):
 
+class WeChatPadMessageConverter(adapter.MessageConverter):
     def __init__(self, config: dict, logger: logging.Logger):
         self.config = config
-        self.bot = WeChatPadClient(self.config["wechatpad_url"],self.config["token"])
+        self.bot = WeChatPadClient(self.config['wechatpad_url'], self.config['token'])
         self.logger = logger
 
     @staticmethod
-    async def yiri2target(
-            message_chain: platform_message.MessageChain
-    ) -> list[dict]:
+    async def yiri2target(message_chain: platform_message.MessageChain) -> list[dict]:
         content_list = []
-        current_file_path = os.path.abspath(__file__)
-
-
 
         for component in message_chain:
-            if isinstance(component, platform_message.At):
-                content_list.append({"type": "at", "target": component.target})
+            if isinstance(component, platform_message.AtAll):
+                content_list.append({'type': 'at', 'target': 'all'})
+            elif isinstance(component, platform_message.At):
+                content_list.append({'type': 'at', 'target': component.target})
             elif isinstance(component, platform_message.Plain):
-                content_list.append({"type": "text", "content": component.text})
+                content_list.append({'type': 'text', 'content': component.text})
             elif isinstance(component, platform_message.Image):
                 if component.url:
                     async with httpx.AsyncClient() as client:
@@ -68,15 +56,16 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
                         else:
                             raise Exception('获取文件失败')
                     # pass
-                    content_list.append({"type": "image", "image": base64_str})
+                    content_list.append({'type': 'image', 'image': base64_str})
                 elif component.base64:
-                    content_list.append({"type": "image", "image": component.base64})
+                    content_list.append({'type': 'image', 'image': component.base64})
 
             elif isinstance(component, platform_message.WeChatEmoji):
                 content_list.append(
-                    {'type': 'WeChatEmoji', 'emoji_md5': component.emoji_md5, 'emoji_size': component.emoji_size})
+                    {'type': 'WeChatEmoji', 'emoji_md5': component.emoji_md5, 'emoji_size': component.emoji_size}
+                )
             elif isinstance(component, platform_message.Voice):
-                content_list.append({"type": "voice", "data": component.url, "duration": component.length, "forma": 0})
+                content_list.append({'type': 'voice', 'data': component.url, 'duration': component.length, 'forma': 0})
             elif isinstance(component, platform_message.WeChatAppMsg):
                 content_list.append({'type': 'WeChatAppMsg', 'app_msg': component.app_msg})
             elif isinstance(component, platform_message.Forward):
@@ -86,37 +75,37 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
 
         return content_list
 
-
     async def target2yiri(
-            self,
-            message: dict,
-            bot_account_id: str,
+        self,
+        message: dict,
+        bot_account_id: str,
     ) -> platform_message.MessageChain:
         """外部消息转平台消息"""
         # 数据预处理
         message_list = []
         bot_wxid = self.config['wxid']
         ats_bot = False  # 是否被@
-        content = message["content"]["str"]
+        content = message['content']['str']
         content_no_preifx = content  # 群消息则去掉前缀
         is_group_message = self._is_group_message(message)
         if is_group_message:
             ats_bot = self._ats_bot(message, bot_account_id)
-            self.logger.info(f"ats_bot: {ats_bot}; bot_account_id: {bot_account_id}; bot_wxid: {bot_wxid}")
-            if "@所有人" in content:
+
+            self.logger.info(f'ats_bot: {ats_bot}; bot_account_id: {bot_account_id}; bot_wxid: {bot_wxid}')
+            if '@所有人' in content:
                 message_list.append(platform_message.AtAll())
-            elif ats_bot:
+            if ats_bot:
                 message_list.append(platform_message.At(target=bot_account_id))
-            
+
             # 解析@信息并生成At组件
             at_targets = self._extract_at_targets(message)
             for target_id in at_targets:
                 if target_id != bot_wxid:  # 避免重复添加机器人的At
                     message_list.append(platform_message.At(target=target_id))
-            
+
             content_no_preifx, _ = self._extract_content_and_sender(content)
 
-        msg_type = message["msg_type"]
+        msg_type = message['msg_type']
 
         # 映射消息类型到处理器方法
         handler_map = {
@@ -138,11 +127,7 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
 
         return platform_message.MessageChain(message_list)
 
-    async def _handler_text(
-            self,
-            message: Optional[dict],
-            content_no_preifx: str
-    ) -> platform_message.MessageChain:
+    async def _handler_text(self, message: Optional[dict], content_no_preifx: str) -> platform_message.MessageChain:
         """处理文本消息 (msg_type=1)"""
         if message and self._is_group_message(message):
             pattern = r'@\S{1,20}'
@@ -150,16 +135,12 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
 
         return platform_message.MessageChain([platform_message.Plain(content_no_preifx)])
 
-    async def _handler_image(
-            self,
-            message: Optional[dict],
-            content_no_preifx: str
-    ) -> platform_message.MessageChain:
+    async def _handler_image(self, message: Optional[dict], content_no_preifx: str) -> platform_message.MessageChain:
         """处理图像消息 (msg_type=3)"""
         try:
             image_xml = content_no_preifx
             if not image_xml:
-                return platform_message.MessageChain([platform_message.Unknown("[图片内容为空]")])
+                return platform_message.MessageChain([platform_message.Unknown('[图片内容为空]')])
             root = ET.fromstring(image_xml)
 
             # 提取img标签的属性
@@ -169,28 +150,22 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
                 cdnthumburl = img_tag.get('cdnthumburl')
                 # cdnmidimgurl = img_tag.get('cdnmidimgurl')
 
-
             image_data = self.bot.cdn_download(aeskey=aeskey, file_type=1, file_url=cdnthumburl)
-            if image_data["Data"]['FileData'] == '':
+            if image_data['Data']['FileData'] == '':
                 image_data = self.bot.cdn_download(aeskey=aeskey, file_type=2, file_url=cdnthumburl)
-            base64_str = image_data["Data"]['FileData']
+            base64_str = image_data['Data']['FileData']
             # self.logger.info(f"data:image/png;base64,{base64_str}")
 
-
             elements = [
-                platform_message.Image(base64=f"data:image/png;base64,{base64_str}"),
+                platform_message.Image(base64=f'data:image/png;base64,{base64_str}'),
                 # platform_message.WeChatForwardImage(xml_data=image_xml)  # 微信消息转发
             ]
             return platform_message.MessageChain(elements)
         except Exception as e:
-            self.logger.error(f"处理图片失败: {str(e)}")
-            return platform_message.MessageChain([platform_message.Unknown("[图片处理失败]")])
+            self.logger.error(f'处理图片失败: {str(e)}')
+            return platform_message.MessageChain([platform_message.Unknown('[图片处理失败]')])
 
-    async def _handler_voice(
-            self,
-            message: Optional[dict],
-            content_no_preifx: str
-    ) -> platform_message.MessageChain:
+    async def _handler_voice(self, message: Optional[dict], content_no_preifx: str) -> platform_message.MessageChain:
         """处理语音消息 (msg_type=34)"""
         message_List = []
         try:
@@ -206,39 +181,33 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
                 bufid = voicemsg.get('bufid')
                 length = voicemsg.get('voicelength')
             voice_data = self.bot.get_msg_voice(buf_id=str(bufid), length=int(length), msgid=str(new_msg_id))
-            audio_base64 = voice_data["Data"]['Base64']
+            audio_base64 = voice_data['Data']['Base64']
 
             # 验证语音数据有效性
             if not audio_base64:
-                message_List.append(platform_message.Unknown(text="[语音内容为空]"))
+                message_List.append(platform_message.Unknown(text='[语音内容为空]'))
                 return platform_message.MessageChain(message_List)
 
             # 转换为平台支持的语音格式（如 Silk 格式）
-            voice_element = platform_message.Voice(
-                base64=f"data:audio/silk;base64,{audio_base64}"
-            )
+            voice_element = platform_message.Voice(base64=f'data:audio/silk;base64,{audio_base64}')
             message_List.append(voice_element)
 
         except KeyError as e:
-            self.logger.error(f"语音数据字段缺失: {str(e)}")
-            message_List.append(platform_message.Unknown(text="[语音数据解析失败]"))
+            self.logger.error(f'语音数据字段缺失: {str(e)}')
+            message_List.append(platform_message.Unknown(text='[语音数据解析失败]'))
         except Exception as e:
-            self.logger.error(f"处理语音消息异常: {str(e)}")
-            message_List.append(platform_message.Unknown(text="[语音处理失败]"))
+            self.logger.error(f'处理语音消息异常: {str(e)}')
+            message_List.append(platform_message.Unknown(text='[语音处理失败]'))
 
         return platform_message.MessageChain(message_List)
 
-    async def _handler_compound(
-            self,
-            message: Optional[dict],
-            content_no_preifx: str
-    ) -> platform_message.MessageChain:
+    async def _handler_compound(self, message: Optional[dict], content_no_preifx: str) -> platform_message.MessageChain:
         """处理复合消息 (msg_type=49)，根据子类型分派"""
         try:
             xml_data = ET.fromstring(content_no_preifx)
             appmsg_data = xml_data.find('.//appmsg')
             if appmsg_data:
-                data_type = appmsg_data.findtext('.//type', "")
+                data_type = appmsg_data.findtext('.//type', '')
                 # 二次分派处理器
                 sub_handler_map = {
                     '57': self._handler_compound_quote,
@@ -247,9 +216,9 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
                     '74': self._handler_compound_file,
                     '33': self._handler_compound_mini_program,
                     '36': self._handler_compound_mini_program,
-                    '2000': partial(self._handler_compound_unsupported, text="[转账消息]"),
-                    '2001': partial(self._handler_compound_unsupported, text="[红包消息]"),
-                    '51': partial(self._handler_compound_unsupported, text="[视频号消息]"),
+                    '2000': partial(self._handler_compound_unsupported, text='[转账消息]'),
+                    '2001': partial(self._handler_compound_unsupported, text='[红包消息]'),
+                    '51': partial(self._handler_compound_unsupported, text='[视频号消息]'),
                 }
 
                 handler = sub_handler_map.get(data_type, self._handler_compound_unsupported)
@@ -260,56 +229,51 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
             else:
                 return platform_message.MessageChain([platform_message.Unknown(text=content_no_preifx)])
         except Exception as e:
-            self.logger.error(f"解析复合消息失败: {str(e)}")
+            self.logger.error(f'解析复合消息失败: {str(e)}')
             return platform_message.MessageChain([platform_message.Unknown(text=content_no_preifx)])
 
     async def _handler_compound_quote(
-            self,
-            message: Optional[dict],
-            xml_data: ET.Element
+        self, message: Optional[dict], xml_data: ET.Element
     ) -> platform_message.MessageChain:
         """处理引用消息 (data_type=57)"""
         message_list = []
-#         self.logger.info("_handler_compound_quote", ET.tostring(xml_data, encoding='unicode'))
+        #         self.logger.info("_handler_compound_quote", ET.tostring(xml_data, encoding='unicode'))
         appmsg_data = xml_data.find('.//appmsg')
-        quote_data = ""  # 引用原文
-        quote_id = None  # 引用消息的原发送者
-        tousername = None  # 接收方: 所属微信的wxid
-        user_data = ""  # 用户消息
+        quote_data = ''  # 引用原文
+        # quote_id = None  # 引用消息的原发送者
+        # tousername = None  # 接收方: 所属微信的wxid
+        user_data = ''  # 用户消息
         sender_id = xml_data.findtext('.//fromusername')  # 发送方：单聊用户/群member
 
         # 引用消息转发
         if appmsg_data:
-            user_data = appmsg_data.findtext('.//title') or ""
+            user_data = appmsg_data.findtext('.//title') or ''
             quote_data = appmsg_data.find('.//refermsg').findtext('.//content')
-            quote_id = appmsg_data.find('.//refermsg').findtext('.//chatusr')
-            message_list.append(
-                platform_message.WeChatAppMsg(
-                    app_msg=ET.tostring(appmsg_data, encoding='unicode'))
-            )
-        if message:
-            tousername = message['to_user_name']["str"]
-        
+            # quote_id = appmsg_data.find('.//refermsg').findtext('.//chatusr')
+            message_list.append(platform_message.WeChatAppMsg(app_msg=ET.tostring(appmsg_data, encoding='unicode')))
+        # if message:
+        #     tousername = message['to_user_name']['str']
+
         if quote_data:
             quote_data_message_list = platform_message.MessageChain()
             # 文本消息
             try:
-                if "<msg>" not in quote_data:
+                if '<msg>' not in quote_data:
                     quote_data_message_list.append(platform_message.Plain(quote_data))
                 else:
                     # 引用消息展开
                     quote_data_xml = ET.fromstring(quote_data)
-                    if quote_data_xml.find("img"):
+                    if quote_data_xml.find('img'):
                         quote_data_message_list.extend(await self._handler_image(None, quote_data))
-                    elif quote_data_xml.find("voicemsg"):
+                    elif quote_data_xml.find('voicemsg'):
                         quote_data_message_list.extend(await self._handler_voice(None, quote_data))
-                    elif quote_data_xml.find("videomsg"):
+                    elif quote_data_xml.find('videomsg'):
                         quote_data_message_list.extend(await self._handler_default(None, quote_data))  # 先不处理
                     else:
                         # appmsg
                         quote_data_message_list.extend(await self._handler_compound(None, quote_data))
             except Exception as e:
-                self.logger.error(f"处理引用消息异常 expcetion:{e}")
+                self.logger.error(f'处理引用消息异常 expcetion:{e}')
                 quote_data_message_list.append(platform_message.Plain(quote_data))
             message_list.append(
                 platform_message.Quote(
@@ -324,15 +288,11 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
 
         return platform_message.MessageChain(message_list)
 
-    async def _handler_compound_file(
-            self,
-            message: dict,
-            xml_data: ET.Element
-    ) -> platform_message.MessageChain:
+    async def _handler_compound_file(self, message: dict, xml_data: ET.Element) -> platform_message.MessageChain:
         """处理文件消息 (data_type=6)"""
         file_data = xml_data.find('.//appmsg')
 
-        if file_data.findtext('.//type', "") == "74":
+        if file_data.findtext('.//type', '') == '74':
             return None
 
         else:
@@ -355,22 +315,21 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
 
             file_data = self.bot.cdn_download(aeskey=aeskey, file_type=5, file_url=cdnthumburl)
 
-            file_base64 = file_data["Data"]['FileData']
+            file_base64 = file_data['Data']['FileData']
             # print(file_data)
-            file_size = file_data["Data"]['TotalSize']
+            file_size = file_data['Data']['TotalSize']
 
             # print(file_base64)
-            return platform_message.MessageChain([
-                platform_message.WeChatFile(file_id=file_id, file_name=file_name, file_size=file_size,
-                                            file_base64=file_base64),
-                platform_message.WeChatForwardFile(xml_data=xml_data_str)
-            ])
+            return platform_message.MessageChain(
+                [
+                    platform_message.WeChatFile(
+                        file_id=file_id, file_name=file_name, file_size=file_size, file_base64=file_base64
+                    ),
+                    platform_message.WeChatForwardFile(xml_data=xml_data_str),
+                ]
+            )
 
-    async def _handler_compound_link(
-            self,
-            message: dict,
-            xml_data: ET.Element
-    ) -> platform_message.MessageChain:
+    async def _handler_compound_link(self, message: dict, xml_data: ET.Element) -> platform_message.MessageChain:
         """处理链接消息（如公众号文章、外部网页）"""
         message_list = []
         try:
@@ -383,56 +342,38 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
                     link_title=appmsg.findtext('title', ''),
                     link_desc=appmsg.findtext('des', ''),
                     link_url=appmsg.findtext('url', ''),
-                    link_thumb_url=appmsg.findtext("thumburl", '')  # 这个字段拿不到
+                    link_thumb_url=appmsg.findtext('thumburl', ''),  # 这个字段拿不到
                 )
             )
             # 还没有发链接的接口, 暂时还需要自己构造appmsg, 先用WeChatAppMsg。
-            message_list.append(
-                platform_message.WeChatAppMsg(
-                    app_msg=ET.tostring(appmsg, encoding='unicode')
-                )
-            )
+            message_list.append(platform_message.WeChatAppMsg(app_msg=ET.tostring(appmsg, encoding='unicode')))
         except Exception as e:
-            self.logger.error(f"解析链接消息失败: {str(e)}")
+            self.logger.error(f'解析链接消息失败: {str(e)}')
         return platform_message.MessageChain(message_list)
 
     async def _handler_compound_mini_program(
-            self,
-            message: dict,
-            xml_data: ET.Element
+        self, message: dict, xml_data: ET.Element
     ) -> platform_message.MessageChain:
         """处理小程序消息（如小程序卡片、服务通知）"""
         xml_data_str = ET.tostring(xml_data, encoding='unicode')
-        return platform_message.MessageChain([
-            platform_message.WeChatForwardMiniPrograms(xml_data=xml_data_str)
-        ])
+        return platform_message.MessageChain([platform_message.WeChatForwardMiniPrograms(xml_data=xml_data_str)])
 
-    async def _handler_default(
-            self,
-            message: Optional[dict],
-            content_no_preifx: str
-    ) -> platform_message.MessageChain:
+    async def _handler_default(self, message: Optional[dict], content_no_preifx: str) -> platform_message.MessageChain:
         """处理未知消息类型"""
         if message:
-            msg_type = message["msg_type"]
+            msg_type = message['msg_type']
         else:
-            msg_type = ""
-        return platform_message.MessageChain([
-            platform_message.Unknown(text=f"[未知消息类型 msg_type:{msg_type}]")
-        ])
+            msg_type = ''
+        return platform_message.MessageChain([platform_message.Unknown(text=f'[未知消息类型 msg_type:{msg_type}]')])
 
     def _handler_compound_unsupported(
-            self,
-            message: dict,
-            xml_data: str,
-            text: Optional[str] = None
+        self, message: dict, xml_data: str, text: Optional[str] = None
     ) -> platform_message.MessageChain:
         """处理未支持复合消息类型(msg_type=49)子类型"""
         if not text:
-            text = f"[xml_data={xml_data}]"
+            text = f'[xml_data={xml_data}]'
         content_list = []
-        content_list.append(
-            platform_message.Unknown(text=f"[处理未支持复合消息类型[msg_type=49]|{text}"))
+        content_list.append(platform_message.Unknown(text=f'[处理未支持复合消息类型[msg_type=49]|{text}'))
 
         return platform_message.MessageChain(content_list)
 
@@ -441,7 +382,7 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
         ats_bot = False
         try:
             to_user_name = message['to_user_name']['str']  # 接收方: 所属微信的wxid
-            raw_content = message["content"]["str"]  # 原始消息内容
+            raw_content = message['content']['str']  # 原始消息内容
             content_no_prefix, _ = self._extract_content_and_sender(raw_content)
             # 直接艾特机器人（这个有bug，当被引用的消息里面有@bot,会套娃
             # ats_bot =  ats_bot or (f"@{bot_account_id}" in content_no_prefix)
@@ -452,7 +393,7 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
             msg_source = message.get('msg_source', '') or ''
             if len(msg_source) > 0:
                 msg_source_data = ET.fromstring(msg_source)
-                at_user_list = msg_source_data.findtext("atuserlist") or ""
+                at_user_list = msg_source_data.findtext('atuserlist') or ''
                 ats_bot = ats_bot or (to_user_name in at_user_list)
             # 引用bot
             if message.get('msg_type', 0) == 49:
@@ -463,7 +404,7 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
                     quote_id = appmsg_data.find('.//refermsg').findtext('.//chatusr')  # 引用消息的原发送者
                     ats_bot = ats_bot or (quote_id == tousername)
         except Exception as e:
-            self.logger.error(f"_ats_bot got except: {e}")
+            self.logger.error(f'_ats_bot got except: {e}')
         finally:
             return ats_bot
 
@@ -476,60 +417,58 @@ class WeChatPadMessageConverter(adapter.MessageConverter):
             msg_source = message.get('msg_source', '') or ''
             if len(msg_source) > 0:
                 msg_source_data = ET.fromstring(msg_source)
-                at_user_list = msg_source_data.findtext("atuserlist") or ""
+                at_user_list = msg_source_data.findtext('atuserlist') or ''
                 if at_user_list:
                     # atuserlist格式通常是逗号分隔的用户ID列表
                     at_targets = [user_id.strip() for user_id in at_user_list.split(',') if user_id.strip()]
         except Exception as e:
-            self.logger.error(f"_extract_at_targets got except: {e}")
+            self.logger.error(f'_extract_at_targets got except: {e}')
         return at_targets
-    
+
     # 提取一下content前面的sender_id, 和去掉前缀的内容
     def _extract_content_and_sender(self, raw_content: str) -> Tuple[str, Optional[str]]:
         try:
             # 检查消息开头，如果有 wxid_sbitaz0mt65n22:\n 则删掉
             # add: 有些用户的wxid不是上述格式。换成user_name:
-            regex = re.compile(r"^[a-zA-Z0-9_\-]{5,20}:")
-            line_split = raw_content.split("\n")
+            regex = re.compile(r'^[a-zA-Z0-9_\-]{5,20}:')
+            line_split = raw_content.split('\n')
             if len(line_split) > 0 and regex.match(line_split[0]):
-                raw_content = "\n".join(line_split[1:])
-                sender_id = line_split[0].strip(":")
+                raw_content = '\n'.join(line_split[1:])
+                sender_id = line_split[0].strip(':')
                 return raw_content, sender_id
         except Exception as e:
-            self.logger.error(f"_extract_content_and_sender got except: {e}")
+            self.logger.error(f'_extract_content_and_sender got except: {e}')
         finally:
             return raw_content, None
 
     # 是否是群消息
     def _is_group_message(self, message: dict) -> bool:
         from_user_name = message['from_user_name']['str']
-        return from_user_name.endswith("@chatroom")
+        return from_user_name.endswith('@chatroom')
 
 
 class WeChatPadEventConverter(adapter.EventConverter):
-
     def __init__(self, config: dict, logger: logging.Logger):
         self.config = config
         self.message_converter = WeChatPadMessageConverter(config, logger)
         self.logger = logger
-        
+
     @staticmethod
-    async def yiri2target(
-            event: platform_events.MessageEvent
-    ) -> dict:
+    async def yiri2target(event: platform_events.MessageEvent) -> dict:
         pass
 
     async def target2yiri(
-            self,
-            event: dict,
-            bot_account_id: str,
+        self,
+        event: dict,
+        bot_account_id: str,
     ) -> platform_events.MessageEvent:
-
         # 排除公众号以及微信团队消息
-        if event['from_user_name']['str'].startswith('gh_') \
-                or event['from_user_name']['str']=='weixin'\
-                or event['from_user_name']['str'] == "newsapp"\
-                or event['from_user_name']['str'] == self.config["wxid"]:
+        if (
+            event['from_user_name']['str'].startswith('gh_')
+            or event['from_user_name']['str'] == 'weixin'
+            or event['from_user_name']['str'] == 'newsapp'
+            or event['from_user_name']['str'] == self.config['wxid']
+        ):
             return None
         message_chain = await self.message_converter.target2yiri(copy.deepcopy(event), bot_account_id)
 
@@ -538,7 +477,7 @@ class WeChatPadEventConverter(adapter.EventConverter):
 
         if '@chatroom' in event['from_user_name']['str']:
             # 找出开头的 wxid_ 字符串，以:结尾
-            sender_wxid = event['content']['str'].split(":")[0]
+            sender_wxid = event['content']['str'].split(':')[0]
 
             return platform_events.GroupMessage(
                 sender=platform_entities.GroupMember(
@@ -550,13 +489,13 @@ class WeChatPadEventConverter(adapter.EventConverter):
                         name=event['from_user_name']['str'],
                         permission=platform_entities.Permission.Member,
                     ),
-                    special_title="",
+                    special_title='',
                     join_timestamp=0,
                     last_speak_timestamp=0,
                     mute_time_remaining=0,
                 ),
                 message_chain=message_chain,
-                time=event["create_time"],
+                time=event['create_time'],
                 source_platform_object=event,
             )
         else:
@@ -567,13 +506,13 @@ class WeChatPadEventConverter(adapter.EventConverter):
                     remark='',
                 ),
                 message_chain=message_chain,
-                time=event["create_time"],
+                time=event['create_time'],
                 source_platform_object=event,
             )
 
 
 class WeChatPadAdapter(adapter.MessagePlatformAdapter):
-    name: str = "WeChatPad"  # 定义适配器名称
+    name: str = 'WeChatPad'  # 定义适配器名称
 
     bot: WeChatPadClient
     quart_app: quart.Quart
@@ -606,27 +545,21 @@ class WeChatPadAdapter(adapter.MessagePlatformAdapter):
         # self.ap.logger.debug(f"Gewechat callback event: {data}")
         # print(data)
 
-
         try:
             event = await self.event_converter.target2yiri(data.copy(), self.bot_account_id)
-        except Exception as e:
-            await self.logger.error(f"Error in wechatpad callback: {traceback.format_exc()}")
+        except Exception:
+            await self.logger.error(f'Error in wechatpad callback: {traceback.format_exc()}')
 
         if event.__class__ in self.listeners:
             await self.listeners[event.__class__](event, self)
 
         return 'ok'
 
-
-    async def _handle_message(
-            self,
-            message: platform_message.MessageChain,
-            target_id: str
-    ):
+    async def _handle_message(self, message: platform_message.MessageChain, target_id: str):
         """统一消息处理核心逻辑"""
         content_list = await self.message_converter.yiri2target(message)
         # print(content_list)
-        at_targets = [item["target"] for item in content_list if item["type"] == "at"]
+        at_targets = [item['target'] for item in content_list if item['type'] == 'at']
         # print(at_targets)
         # 处理@逻辑
         at_targets = at_targets or []
@@ -634,71 +567,62 @@ class WeChatPadAdapter(adapter.MessagePlatformAdapter):
         if at_targets:
             member_info = self.bot.get_chatroom_member_detail(
                 target_id,
-            )["Data"]["member_data"]["chatroom_member_list"]
+            )['Data']['member_data']['chatroom_member_list']
 
         # 处理消息组件
         for msg in content_list:
             # 文本消息处理@
             if msg['type'] == 'text' and at_targets:
-                at_nick_name_list = []
-                for member in member_info:
-                    if member["user_name"] in at_targets:
-                        at_nick_name_list.append(f'@{member["nick_name"]}')
-                msg['content'] = f'{" ".join(at_nick_name_list)} {msg["content"]}'
+                if 'all' in at_targets:
+                    msg['content'] = f'@所有人 {msg["content"]}'
+                else:
+                    at_nick_name_list = []
+                    for member in member_info:
+                        if member['user_name'] in at_targets:
+                            at_nick_name_list.append(f'@{member["nick_name"]}')
+                    msg['content'] = f'{" ".join(at_nick_name_list)} {msg["content"]}'
 
             # 统一消息派发
             handler_map = {
                 'text': lambda msg: self.bot.send_text_message(
-                    to_wxid=target_id,
-                    message=msg['content'],
-                    ats=at_targets
+                    to_wxid=target_id, message=msg['content'], ats=['notify@all'] if 'all' in at_targets else at_targets
                 ),
                 'image': lambda msg: self.bot.send_image_message(
-                    to_wxid=target_id,
-                    img_url=msg["image"],
-                    ats = at_targets
+                    to_wxid=target_id, img_url=msg['image'], ats=['notify@all'] if 'all' in at_targets else at_targets
                 ),
                 'WeChatEmoji': lambda msg: self.bot.send_emoji_message(
-                    to_wxid=target_id,
-                    emoji_md5=msg['emoji_md5'],
-                    emoji_size=msg['emoji_size']
+                    to_wxid=target_id, emoji_md5=msg['emoji_md5'], emoji_size=msg['emoji_size']
                 ),
-
                 'voice': lambda msg: self.bot.send_voice_message(
                     to_wxid=target_id,
                     voice_data=msg['data'],
-                    voice_duration=msg["duration"],
-                    voice_forma=msg["forma"],
+                    voice_duration=msg['duration'],
+                    voice_forma=msg['forma'],
                 ),
                 'WeChatAppMsg': lambda msg: self.bot.send_app_message(
                     to_wxid=target_id,
                     app_message=msg['app_msg'],
                     type=0,
                 ),
-                'at': lambda msg: None
+                'at': lambda msg: None,
             }
 
             if handler := handler_map.get(msg['type']):
                 handler(msg)
                 # self.ap.logger.warning(f"未处理的消息类型: {ret}")
             else:
-                self.ap.logger.warning(f"未处理的消息类型: {msg['type']}")
+                self.ap.logger.warning(f'未处理的消息类型: {msg["type"]}')
                 continue
 
-    async def send_message(
-            self,
-            target_type: str,
-            target_id: str,
-            message: platform_message.MessageChain
-    ):
+    async def send_message(self, target_type: str, target_id: str, message: platform_message.MessageChain):
         """主动发送消息"""
         return await self._handle_message(message, target_id)
 
     async def reply_message(
-            self,
-            message_source: platform_events.MessageEvent,
-            message: platform_message.MessageChain,
-            quote_origin: bool = False
+        self,
+        message_source: platform_events.MessageEvent,
+        message: platform_message.MessageChain,
+        quote_origin: bool = False,
     ):
         """回复消息"""
         if message_source.source_platform_object:
@@ -709,58 +633,49 @@ class WeChatPadAdapter(adapter.MessagePlatformAdapter):
         pass
 
     def register_listener(
-            self,
-            event_type: typing.Type[platform_events.Event],
-            callback: typing.Callable[[platform_events.Event, adapter.MessagePlatformAdapter], None]
+        self,
+        event_type: typing.Type[platform_events.Event],
+        callback: typing.Callable[[platform_events.Event, adapter.MessagePlatformAdapter], None],
     ):
         self.listeners[event_type] = callback
 
     def unregister_listener(
-            self,
-            event_type: typing.Type[platform_events.Event],
-            callback: typing.Callable[[platform_events.Event, adapter.MessagePlatformAdapter], None]
+        self,
+        event_type: typing.Type[platform_events.Event],
+        callback: typing.Callable[[platform_events.Event, adapter.MessagePlatformAdapter], None],
     ):
         pass
 
     async def run_async(self):
-
-        if not self.config["admin_key"] and not self.config["token"]:
-            raise RuntimeError("无wechatpad管理密匙，请填入配置文件后重启")
+        if not self.config['admin_key'] and not self.config['token']:
+            raise RuntimeError('无wechatpad管理密匙，请填入配置文件后重启')
         else:
-            if self.config["token"]:
-                self.bot = WeChatPadClient(
-                    self.config['wechatpad_url'],
-                    self.config["token"]
-                )
+            if self.config['token']:
+                self.bot = WeChatPadClient(self.config['wechatpad_url'], self.config['token'])
                 data = self.bot.get_login_status()
                 self.ap.logger.info(data)
-                if data["Code"] == 300 and data["Text"] == "你已退出微信":
+                if data['Code'] == 300 and data['Text'] == '你已退出微信':
                     response = requests.post(
-                        f"{self.config['wechatpad_url']}/admin/GenAuthKey1?key={self.config['admin_key']}",
-                        json={"Count": 1, "Days": 365}
+                        f'{self.config["wechatpad_url"]}/admin/GenAuthKey1?key={self.config["admin_key"]}',
+                        json={'Count': 1, 'Days': 365},
                     )
                     if response.status_code != 200:
-                        raise Exception(f"获取token失败: {response.text}")
-                    self.config["token"] = response.json()["Data"][0]
+                        raise Exception(f'获取token失败: {response.text}')
+                    self.config['token'] = response.json()['Data'][0]
 
-            elif not self.config["token"]:
+            elif not self.config['token']:
                 response = requests.post(
-                    f"{self.config['wechatpad_url']}/admin/GenAuthKey1?key={self.config['admin_key']}",
-                    json={"Count": 1, "Days": 365}
+                    f'{self.config["wechatpad_url"]}/admin/GenAuthKey1?key={self.config["admin_key"]}',
+                    json={'Count': 1, 'Days': 365},
                 )
                 if response.status_code != 200:
-                    raise Exception(f"获取token失败: {response.text}")
-                self.config["token"] = response.json()["Data"][0]
+                    raise Exception(f'获取token失败: {response.text}')
+                self.config['token'] = response.json()['Data'][0]
 
-        self.bot = WeChatPadClient(
-            self.config['wechatpad_url'],
-            self.config["token"],
-            logger=self.logger
-        )
-        self.ap.logger.info(self.config["token"])
+        self.bot = WeChatPadClient(self.config['wechatpad_url'], self.config['token'], logger=self.logger)
+        self.ap.logger.info(self.config['token'])
         thread_1 = threading.Event()
 
-
         def wechat_login_process():
             # 不登录，这些先注释掉，避免登陆态尝试拉qrcode。
             # login_data =self.bot.get_login_qr()
@@ -768,67 +683,54 @@ class WeChatPadAdapter(adapter.MessagePlatformAdapter):
             # url = login_data['Data']["QrCodeUrl"]
             # self.ap.logger.info(login_data)
 
-
-            profile =self.bot.get_profile()
+            profile = self.bot.get_profile()
             self.ap.logger.info(profile)
 
-            self.bot_account_id = profile["Data"]["userInfo"]["nickName"]["str"]
-            self.config["wxid"] = profile["Data"]["userInfo"]["userName"]["str"]
+            self.bot_account_id = profile['Data']['userInfo']['nickName']['str']
+            self.config['wxid'] = profile['Data']['userInfo']['userName']['str']
             thread_1.set()
 
-
         # asyncio.create_task(wechat_login_process)
         threading.Thread(target=wechat_login_process).start()
 
         def connect_websocket_sync() -> None:
-
             thread_1.wait()
-            uri = f"{self.config['wechatpad_ws']}/GetSyncMsg?key={self.config['token']}"
-            self.ap.logger.info(f"Connecting to WebSocket: {uri}")
+            uri = f'{self.config["wechatpad_ws"]}/GetSyncMsg?key={self.config["token"]}'
+            self.ap.logger.info(f'Connecting to WebSocket: {uri}')
+
             def on_message(ws, message):
                 try:
                     data = json.loads(message)
-                    self.ap.logger.debug(f"Received message: {data}")
+                    self.ap.logger.debug(f'Received message: {data}')
                     # 这里需要确保ws_message是同步的，或者使用asyncio.run调用异步方法
                     asyncio.run(self.ws_message(data))
                 except json.JSONDecodeError:
-                    self.ap.logger.error(f"Non-JSON message: {message[:100]}...")
+                    self.ap.logger.error(f'Non-JSON message: {message[:100]}...')
 
             def on_error(ws, error):
-                self.ap.logger.error(f"WebSocket error: {str(error)[:200]}")
+                self.ap.logger.error(f'WebSocket error: {str(error)[:200]}')
 
             def on_close(ws, close_status_code, close_msg):
-                self.ap.logger.info("WebSocket closed, reconnecting...")
+                self.ap.logger.info('WebSocket closed, reconnecting...')
                 time.sleep(5)
                 connect_websocket_sync()  # 自动重连
 
             def on_open(ws):
-                self.ap.logger.info("WebSocket connected successfully!")
+                self.ap.logger.info('WebSocket connected successfully!')
 
             ws = websocket.WebSocketApp(
-                uri,
-                on_message=on_message,
-                on_error=on_error,
-                on_close=on_close,
-                on_open=on_open
-            )
-            ws.run_forever(
-                ping_interval=60,
-                ping_timeout=20
+                uri, on_message=on_message, on_error=on_error, on_close=on_close, on_open=on_open
             )
+            ws.run_forever(ping_interval=60, ping_timeout=20)
 
         # 直接调用同步版本（会阻塞）
         # connect_websocket_sync()
 
         # 这行代码会在WebSocket连接断开后才会执行
         # self.ap.logger.info("WebSocket client thread started")
-        thread = threading.Thread(
-            target=connect_websocket_sync,
-            name="WebSocketClientThread",
-            daemon=True
-        )
+        thread = threading.Thread(target=connect_websocket_sync, name='WebSocketClientThread', daemon=True)
         thread.start()
-        self.ap.logger.info("WebSocket client thread started")
+        self.ap.logger.info('WebSocket client thread started')
 
     async def kill(self) -> bool:
         pass

pkg/platform/sources/wecom.py

@@ -157,7 +157,7 @@ class WecomAdapter(adapter.MessagePlatformAdapter):
             token=config['token'],
             EncodingAESKey=config['EncodingAESKey'],
             contacts_secret=config['contacts_secret'],
-            logger=self.logger
+            logger=self.logger,
         )
 
     async def reply_message(
@@ -201,8 +201,8 @@ class WecomAdapter(adapter.MessagePlatformAdapter):
             self.bot_account_id = event.receiver_id
             try:
                 return await callback(await self.event_converter.target2yiri(event), self)
-            except Exception as e:
-                await self.logger.error(f"Error in wecom callback: {traceback.format_exc()}")
+            except Exception:
+                await self.logger.error(f'Error in wecom callback: {traceback.format_exc()}')
 
         if event_type == platform_events.FriendMessage:
             self.bot.on_message('text')(on_message)

pkg/platform/sources/wecomcs.py

@@ -145,7 +145,7 @@ class WecomCSAdapter(adapter.MessagePlatformAdapter):
             secret=config['secret'],
             token=config['token'],
             EncodingAESKey=config['EncodingAESKey'],
-            logger=self.logger
+            logger=self.logger,
         )
 
     async def reply_message(
@@ -178,8 +178,8 @@ class WecomCSAdapter(adapter.MessagePlatformAdapter):
             self.bot_account_id = event.receiver_id
             try:
                 return await callback(await self.event_converter.target2yiri(event), self)
-            except Exception as e:
-                await self.logger.error(f"Error in wecomcs callback: {traceback.format_exc()}")
+            except Exception:
+                await self.logger.error(f'Error in wecomcs callback: {traceback.format_exc()}')
 
         if event_type == platform_events.FriendMessage:
             self.bot.on_message('text')(on_message)

pkg/platform/types/message.py

@@ -812,12 +812,14 @@ class File(MessageComponent):
     def __str__(self):
         return f'[文件]{self.name}'
 
+
 class Face(MessageComponent):
     """系统表情
     此处将超级表情骰子/划拳，一同归类于face
     当face_type为rps(划拳)时 face_id 对应的是手势
     当face_type为dice(骰子)时 face_id 对应的是点数
     """
+
     type: str = 'Face'
     """表情类型"""
     face_type: str = 'face'
@@ -834,15 +836,15 @@ class Face(MessageComponent):
         elif self.face_type == 'rps':
             return f'[表情]{self.face_name}({self.rps_data(self.face_id)})'
 
-
-    def rps_data(self,face_id):
-        rps_dict ={
-            1 : "布",
-            2 : "剪刀",
-            3 : "石头",
+    def rps_data(self, face_id):
+        rps_dict = {
+            1: '布',
+            2: '剪刀',
+            3: '石头',
         }
         return rps_dict[face_id]
 
+
 # ================ 个人微信专用组件 ================
 
 
@@ -971,5 +973,6 @@ class WeChatFile(MessageComponent):
     """文件地址"""
     file_base64: str = ''
     """base64"""
+
     def __str__(self):
-        return f'[文件]{self.file_name}'
+        return f'[文件]{self.file_name}'

pkg/provider/entities.py

@@ -125,6 +125,95 @@ class Message(pydantic.BaseModel):
             return platform_message.MessageChain(mc)
 
 
+class MessageChunk(pydantic.BaseModel):
+    """消息"""
+
+    resp_message_id: typing.Optional[str] = None
+    """消息id"""
+
+    role: str  # user, system, assistant, tool, command, plugin
+    """消息的角色"""
+
+    name: typing.Optional[str] = None
+    """名称，仅函数调用返回时设置"""
+
+    all_content: typing.Optional[str] = None
+    """所有内容"""
+
+    content: typing.Optional[list[ContentElement]] | typing.Optional[str] = None
+    """内容"""
+
+    tool_calls: typing.Optional[list[ToolCall]] = None
+    """工具调用"""
+
+    tool_call_id: typing.Optional[str] = None
+
+    is_final: bool = False
+    """是否是结束"""
+
+    msg_sequence: int = 0
+    """消息迭代次数"""
+
+    def readable_str(self) -> str:
+        if self.content is not None:
+            return str(self.role) + ': ' + str(self.get_content_platform_message_chain())
+        elif self.tool_calls is not None:
+            return f'调用工具: {self.tool_calls[0].id}'
+        else:
+            return '未知消息'
+
+    def get_content_platform_message_chain(self, prefix_text: str = '') -> platform_message.MessageChain | None:
+        """将内容转换为平台消息 MessageChain 对象
+
+        Args:
+            prefix_text (str): 首个文字组件的前缀文本
+        """
+
+        if self.content is None:
+            return None
+        elif isinstance(self.content, str):
+            return platform_message.MessageChain([platform_message.Plain(prefix_text + self.content)])
+        elif isinstance(self.content, list):
+            mc = []
+            for ce in self.content:
+                if ce.type == 'text':
+                    mc.append(platform_message.Plain(ce.text))
+                elif ce.type == 'image_url':
+                    if ce.image_url.url.startswith('http'):
+                        mc.append(platform_message.Image(url=ce.image_url.url))
+                    else:  # base64
+                        b64_str = ce.image_url.url
+
+                        if b64_str.startswith('data:'):
+                            b64_str = b64_str.split(',')[1]
+
+                        mc.append(platform_message.Image(base64=b64_str))
+
+            # 找第一个文字组件
+            if prefix_text:
+                for i, c in enumerate(mc):
+                    if isinstance(c, platform_message.Plain):
+                        mc[i] = platform_message.Plain(prefix_text + c.text)
+                        break
+                else:
+                    mc.insert(0, platform_message.Plain(prefix_text))
+
+            return platform_message.MessageChain(mc)
+
+
+class ToolCallChunk(pydantic.BaseModel):
+    """工具调用"""
+
+    id: str
+    """工具调用ID"""
+
+    type: str
+    """工具调用类型"""
+
+    function: FunctionCall
+    """函数调用"""
+
+
 class Prompt(pydantic.BaseModel):
     """供AI使用的Prompt"""
 

pkg/provider/modelmgr/entities.py

@@ -17,7 +17,7 @@ class LLMModelInfo(pydantic.BaseModel):
 
     token_mgr: token.TokenManager
 
-    requester: requester.LLMAPIRequester
+    requester: requester.ProviderAPIRequester
 
     tool_call_supported: typing.Optional[bool] = False
 

pkg/provider/modelmgr/modelmgr.py

@@ -18,7 +18,7 @@ class ModelManager:
 
     model_list: list[entities.LLMModelInfo]  # deprecated
 
-    requesters: dict[str, requester.LLMAPIRequester]  # deprecated
+    requesters: dict[str, requester.ProviderAPIRequester]  # deprecated
 
     token_mgrs: dict[str, token.TokenManager]  # deprecated
 
@@ -28,9 +28,11 @@ class ModelManager:
 
     llm_models: list[requester.RuntimeLLMModel]
 
+    embedding_models: list[requester.RuntimeEmbeddingModel]
+
     requester_components: list[engine.Component]
 
-    requester_dict: dict[str, type[requester.LLMAPIRequester]]  # cache
+    requester_dict: dict[str, type[requester.ProviderAPIRequester]]  # cache
 
     def __init__(self, ap: app.Application):
         self.ap = ap
@@ -38,6 +40,7 @@ class ModelManager:
         self.requesters = {}
         self.token_mgrs = {}
         self.llm_models = []
+        self.embedding_models = []
         self.requester_components = []
         self.requester_dict = {}
 
@@ -45,7 +48,7 @@ class ModelManager:
         self.requester_components = self.ap.discover.get_components_by_kind('LLMAPIRequester')
 
         # forge requester class dict
-        requester_dict: dict[str, type[requester.LLMAPIRequester]] = {}
+        requester_dict: dict[str, type[requester.ProviderAPIRequester]] = {}
         for component in self.requester_components:
             requester_dict[component.metadata.name] = component.get_python_component_class()
 
@@ -58,13 +61,11 @@ class ModelManager:
         self.ap.logger.info('Loading models from db...')
 
         self.llm_models = []
+        self.embedding_models = []
 
         # llm models
         result = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_model.LLMModel))
-
         llm_models = result.all()
-
-        # load models
         for llm_model in llm_models:
             try:
                 await self.load_llm_model(llm_model)
@@ -73,11 +74,17 @@ class ModelManager:
             except Exception as e:
                 self.ap.logger.error(f'Failed to load model {llm_model.uuid}: {e}\n{traceback.format_exc()}')
 
+        # embedding models
+        result = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_model.EmbeddingModel))
+        embedding_models = result.all()
+        for embedding_model in embedding_models:
+            await self.load_embedding_model(embedding_model)
+
     async def init_runtime_llm_model(
         self,
         model_info: persistence_model.LLMModel | sqlalchemy.Row[persistence_model.LLMModel] | dict,
     ):
-        """初始化运行时模型"""
+        """初始化运行时 LLM 模型"""
         if isinstance(model_info, sqlalchemy.Row):
             model_info = persistence_model.LLMModel(**model_info._mapping)
         elif isinstance(model_info, dict):
@@ -101,14 +108,47 @@ class ModelManager:
 
         return runtime_llm_model
 
+    async def init_runtime_embedding_model(
+        self,
+        model_info: persistence_model.EmbeddingModel | sqlalchemy.Row[persistence_model.EmbeddingModel] | dict,
+    ):
+        """初始化运行时 Embedding 模型"""
+        if isinstance(model_info, sqlalchemy.Row):
+            model_info = persistence_model.EmbeddingModel(**model_info._mapping)
+        elif isinstance(model_info, dict):
+            model_info = persistence_model.EmbeddingModel(**model_info)
+
+        requester_inst = self.requester_dict[model_info.requester](ap=self.ap, config=model_info.requester_config)
+
+        await requester_inst.initialize()
+
+        runtime_embedding_model = requester.RuntimeEmbeddingModel(
+            model_entity=model_info,
+            token_mgr=token.TokenManager(
+                name=model_info.uuid,
+                tokens=model_info.api_keys,
+            ),
+            requester=requester_inst,
+        )
+
+        return runtime_embedding_model
+
     async def load_llm_model(
         self,
         model_info: persistence_model.LLMModel | sqlalchemy.Row[persistence_model.LLMModel] | dict,
     ):
-        """加载模型"""
+        """加载 LLM 模型"""
         runtime_llm_model = await self.init_runtime_llm_model(model_info)
         self.llm_models.append(runtime_llm_model)
 
+    async def load_embedding_model(
+        self,
+        model_info: persistence_model.EmbeddingModel | sqlalchemy.Row[persistence_model.EmbeddingModel] | dict,
+    ):
+        """加载 Embedding 模型"""
+        runtime_embedding_model = await self.init_runtime_embedding_model(model_info)
+        self.embedding_models.append(runtime_embedding_model)
+
     async def get_model_by_name(self, name: str) -> entities.LLMModelInfo:  # deprecated
         """通过名称获取模型"""
         for model in self.model_list:
@@ -116,23 +156,44 @@ class ModelManager:
                 return model
         raise ValueError(f'无法确定模型 {name} 的信息')
 
-    async def get_model_by_uuid(self, uuid: str) -> entities.LLMModelInfo:
-        """通过uuid获取模型"""
+    async def get_model_by_uuid(self, uuid: str) -> requester.RuntimeLLMModel:
+        """通过uuid获取 LLM 模型"""
         for model in self.llm_models:
             if model.model_entity.uuid == uuid:
                 return model
-        raise ValueError(f'model {uuid} not found')
+        raise ValueError(f'LLM model {uuid} not found')
+
+    async def get_embedding_model_by_uuid(self, uuid: str) -> requester.RuntimeEmbeddingModel:
+        """通过uuid获取 Embedding 模型"""
+        for model in self.embedding_models:
+            if model.model_entity.uuid == uuid:
+                return model
+        raise ValueError(f'Embedding model {uuid} not found')
 
     async def remove_llm_model(self, model_uuid: str):
-        """移除模型"""
+        """移除 LLM 模型"""
         for model in self.llm_models:
             if model.model_entity.uuid == model_uuid:
                 self.llm_models.remove(model)
                 return
 
-    def get_available_requesters_info(self) -> list[dict]:
+    async def remove_embedding_model(self, model_uuid: str):
+        """移除 Embedding 模型"""
+        for model in self.embedding_models:
+            if model.model_entity.uuid == model_uuid:
+                self.embedding_models.remove(model)
+                return
+
+    def get_available_requesters_info(self, model_type: str) -> list[dict]:
         """获取所有可用的请求器"""
-        return [component.to_plain_dict() for component in self.requester_components]
+        if model_type != '':
+            return [
+                component.to_plain_dict()
+                for component in self.requester_components
+                if model_type in component.spec['support_type']
+            ]
+        else:
+            return [component.to_plain_dict() for component in self.requester_components]
 
     def get_available_requester_info_by_name(self, name: str) -> dict | None:
         """通过名称获取请求器信息"""

pkg/provider/modelmgr/requester.py

@@ -20,22 +20,45 @@ class RuntimeLLMModel:
     token_mgr: token.TokenManager
     """api key管理器"""
 
-    requester: LLMAPIRequester
+    requester: ProviderAPIRequester
     """请求器实例"""
 
     def __init__(
         self,
         model_entity: persistence_model.LLMModel,
         token_mgr: token.TokenManager,
-        requester: LLMAPIRequester,
+        requester: ProviderAPIRequester,
     ):
         self.model_entity = model_entity
         self.token_mgr = token_mgr
         self.requester = requester
 
 
-class LLMAPIRequester(metaclass=abc.ABCMeta):
-    """LLM API请求器"""
+class RuntimeEmbeddingModel:
+    """运行时 Embedding 模型"""
+
+    model_entity: persistence_model.EmbeddingModel
+    """模型数据"""
+
+    token_mgr: token.TokenManager
+    """api key管理器"""
+
+    requester: ProviderAPIRequester
+    """请求器实例"""
+
+    def __init__(
+        self,
+        model_entity: persistence_model.EmbeddingModel,
+        token_mgr: token.TokenManager,
+        requester: ProviderAPIRequester,
+    ):
+        self.model_entity = model_entity
+        self.token_mgr = token_mgr
+        self.requester = requester
+
+
+class ProviderAPIRequester(metaclass=abc.ABCMeta):
+    """Provider API请求器"""
 
     name: str = None
 
@@ -61,6 +84,7 @@ class LLMAPIRequester(metaclass=abc.ABCMeta):
         messages: typing.List[llm_entities.Message],
         funcs: typing.List[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         """调用API
 
@@ -69,8 +93,51 @@ class LLMAPIRequester(metaclass=abc.ABCMeta):
             messages (typing.List[llm_entities.Message]): 消息对象列表
             funcs (typing.List[tools_entities.LLMFunction], optional): 使用的工具函数列表. Defaults to None.
             extra_args (dict[str, typing.Any], optional): 额外的参数. Defaults to {}.
+            remove_think (bool, optional): 是否移思考中的消息. Defaults to False.
 
         Returns:
             llm_entities.Message: 返回消息对象
         """
         pass
+
+    async def invoke_llm_stream(
+        self,
+        query: core_entities.Query,
+        model: RuntimeLLMModel,
+        messages: typing.List[llm_entities.Message],
+        funcs: typing.List[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.MessageChunk:
+        """调用API
+
+        Args:
+            model (RuntimeLLMModel): 使用的模型信息
+            messages (typing.List[llm_entities.Message]): 消息对象列表
+            funcs (typing.List[tools_entities.LLMFunction], optional): 使用的工具函数列表. Defaults to None.
+            extra_args (dict[str, typing.Any], optional): 额外的参数. Defaults to {}.
+            remove_think (bool, optional): 是否移除思考中的消息. Defaults to False.
+
+        Returns:
+            typing.AsyncGenerator[llm_entities.MessageChunk]: 返回消息对象
+        """
+        pass
+
+    async def invoke_embedding(
+        self,
+        model: RuntimeEmbeddingModel,
+        input_text: list[str],
+        extra_args: dict[str, typing.Any] = {},
+    ) -> list[list[float]]:
+        """调用 Embedding API
+
+        Args:
+            query (core_entities.Query): 请求上下文
+            model (RuntimeEmbeddingModel): 使用的模型信息
+            input_text (list[str]): 输入文本
+            extra_args (dict[str, typing.Any], optional): 额外的参数. Defaults to {}.
+
+        Returns:
+            list[list[float]]: 返回的 embedding 向量
+        """
+        pass

pkg/provider/modelmgr/requesters/302aichatcmpl.yaml

@@ -22,6 +22,9 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
+    - text-embedding
 execution:
   python:
     path: ./302aichatcmpl.py

pkg/provider/modelmgr/requesters/anthropicmsgs.py

@@ -15,13 +15,13 @@ from ...tools import entities as tools_entities
 from ....utils import image
 
 
-class AnthropicMessages(requester.LLMAPIRequester):
+class AnthropicMessages(requester.ProviderAPIRequester):
     """Anthropic Messages API 请求器"""
 
     client: anthropic.AsyncAnthropic
 
     default_config: dict[str, typing.Any] = {
-        'base_url': 'https://api.anthropic.com/v1',
+        'base_url': 'https://api.anthropic.com',
         'timeout': 120,
     }
 
@@ -44,6 +44,7 @@ class AnthropicMessages(requester.LLMAPIRequester):
         self.client = anthropic.AsyncAnthropic(
             api_key='',
             http_client=httpx_client,
+            base_url=self.requester_cfg['base_url'],
         )
 
     async def invoke_llm(
@@ -53,6 +54,7 @@ class AnthropicMessages(requester.LLMAPIRequester):
         messages: typing.List[llm_entities.Message],
         funcs: typing.List[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         self.client.api_key = model.token_mgr.get_token()
 
@@ -89,7 +91,8 @@ class AnthropicMessages(requester.LLMAPIRequester):
                             {
                                 'type': 'tool_result',
                                 'tool_use_id': tool_call_id,
-                                'content': m.content,
+                                'is_error': False,
+                                'content': [{'type': 'text', 'text': m.content}],
                             }
                         ],
                     }
@@ -133,6 +136,9 @@ class AnthropicMessages(requester.LLMAPIRequester):
 
         args['messages'] = req_messages
 
+        if 'thinking' in args:
+            args['thinking'] = {'type': 'enabled', 'budget_tokens': 10000}
+
         if funcs:
             tools = await self.ap.tool_mgr.generate_tools_for_anthropic(funcs)
 
@@ -140,19 +146,17 @@ class AnthropicMessages(requester.LLMAPIRequester):
                 args['tools'] = tools
 
         try:
-            # print(json.dumps(args, indent=4, ensure_ascii=False))
             resp = await self.client.messages.create(**args)
 
             args = {
                 'content': '',
                 'role': resp.role,
             }
-
             assert type(resp) is anthropic.types.message.Message
 
             for block in resp.content:
-                if block.type == 'thinking':
-                    args['content'] = '<think>' + block.thinking + '</think>\n' + args['content']
+                if not remove_think and block.type == 'thinking':
+                    args['content'] = '<think>\n' + block.thinking + '\n</think>\n' + args['content']
                 elif block.type == 'text':
                     args['content'] += block.text
                 elif block.type == 'tool_use':
@@ -176,3 +180,191 @@ class AnthropicMessages(requester.LLMAPIRequester):
                 raise errors.RequesterError(f'模型无效: {e.message}')
             else:
                 raise errors.RequesterError(f'请求地址无效: {e.message}')
+
+    async def invoke_llm_stream(
+        self,
+        query: core_entities.Query,
+        model: requester.RuntimeLLMModel,
+        messages: typing.List[llm_entities.Message],
+        funcs: typing.List[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.Message:
+        self.client.api_key = model.token_mgr.get_token()
+
+        args = extra_args.copy()
+        args['model'] = model.model_entity.name
+        args['stream'] = True
+
+        # 处理消息
+
+        # system
+        system_role_message = None
+
+        for i, m in enumerate(messages):
+            if m.role == 'system':
+                system_role_message = m
+
+                break
+
+        if system_role_message:
+            messages.pop(i)
+
+        if isinstance(system_role_message, llm_entities.Message) and isinstance(system_role_message.content, str):
+            args['system'] = system_role_message.content
+
+        req_messages = []
+
+        for m in messages:
+            if m.role == 'tool':
+                tool_call_id = m.tool_call_id
+
+                req_messages.append(
+                    {
+                        'role': 'user',
+                        'content': [
+                            {
+                                'type': 'tool_result',
+                                'tool_use_id': tool_call_id,
+                                'is_error': False,  # 暂时直接写false
+                                'content': [
+                                    {'type': 'text', 'text': m.content}
+                                ],  # 这里要是list包裹，应该是多个返回的情况？type类型好像也可以填其他的，暂时只写text
+                            }
+                        ],
+                    }
+                )
+
+                continue
+
+            msg_dict = m.dict(exclude_none=True)
+
+            if isinstance(m.content, str) and m.content.strip() != '':
+                msg_dict['content'] = [{'type': 'text', 'text': m.content}]
+            elif isinstance(m.content, list):
+                for i, ce in enumerate(m.content):
+                    if ce.type == 'image_base64':
+                        image_b64, image_format = await image.extract_b64_and_format(ce.image_base64)
+
+                        alter_image_ele = {
+                            'type': 'image',
+                            'source': {
+                                'type': 'base64',
+                                'media_type': f'image/{image_format}',
+                                'data': image_b64,
+                            },
+                        }
+                        msg_dict['content'][i] = alter_image_ele
+            if isinstance(msg_dict['content'], str) and msg_dict['content'] == '':
+                msg_dict['content'] = []  # 这里不知道为什么会莫名有个空导致content为字符
+            if m.tool_calls:
+                for tool_call in m.tool_calls:
+                    msg_dict['content'].append(
+                        {
+                            'type': 'tool_use',
+                            'id': tool_call.id,
+                            'name': tool_call.function.name,
+                            'input': json.loads(tool_call.function.arguments),
+                        }
+                    )
+
+                del msg_dict['tool_calls']
+
+            req_messages.append(msg_dict)
+        if 'thinking' in args:
+            args['thinking'] = {'type': 'enabled', 'budget_tokens': 10000}
+
+        args['messages'] = req_messages
+
+        if funcs:
+            tools = await self.ap.tool_mgr.generate_tools_for_anthropic(funcs)
+
+            if tools:
+                args['tools'] = tools
+
+        try:
+            role = 'assistant'  # 默认角色
+            # chunk_idx = 0
+            think_started = False
+            think_ended = False
+            finish_reason = False
+            content = ''
+            tool_name = ''
+            tool_id = ''
+            async for chunk in await self.client.messages.create(**args):
+                tool_call = {'id': None, 'function': {'name': None, 'arguments': None}, 'type': 'function'}
+                if isinstance(
+                    chunk, anthropic.types.raw_content_block_start_event.RawContentBlockStartEvent
+                ):  # 记录开始
+                    if chunk.content_block.type == 'tool_use':
+                        if chunk.content_block.name is not None:
+                            tool_name = chunk.content_block.name
+                        if chunk.content_block.id is not None:
+                            tool_id = chunk.content_block.id
+
+                        tool_call['function']['name'] = tool_name
+                        tool_call['function']['arguments'] = ''
+                        tool_call['id'] = tool_id
+
+                    if not remove_think:
+                        if chunk.content_block.type == 'thinking' and not remove_think:
+                            think_started = True
+                        elif chunk.content_block.type == 'text' and chunk.index != 0 and not remove_think:
+                            think_ended = True
+                        continue
+                elif isinstance(chunk, anthropic.types.raw_content_block_delta_event.RawContentBlockDeltaEvent):
+                    if chunk.delta.type == 'thinking_delta':
+                        if think_started:
+                            think_started = False
+                            content = '<think>\n' + chunk.delta.thinking
+                        elif remove_think:
+                            continue
+                        else:
+                            content = chunk.delta.thinking
+                    elif chunk.delta.type == 'text_delta':
+                        if think_ended:
+                            think_ended = False
+                            content = '\n</think>\n' + chunk.delta.text
+                        else:
+                            content = chunk.delta.text
+                    elif chunk.delta.type == 'input_json_delta':
+                        tool_call['function']['arguments'] = chunk.delta.partial_json
+                        tool_call['function']['name'] = tool_name
+                        tool_call['id'] = tool_id
+                elif isinstance(chunk, anthropic.types.raw_content_block_stop_event.RawContentBlockStopEvent):
+                    continue  # 记录raw_content_block结束的
+
+                elif isinstance(chunk, anthropic.types.raw_message_delta_event.RawMessageDeltaEvent):
+                    if chunk.delta.stop_reason == 'end_turn':
+                        finish_reason = True
+                elif isinstance(chunk, anthropic.types.raw_message_stop_event.RawMessageStopEvent):
+                    continue  # 这个好像是完全结束
+                else:
+                    # print(chunk)
+                    self.ap.logger.debug(f'anthropic chunk: {chunk}')
+                    continue
+
+                args = {
+                    'content': content,
+                    'role': role,
+                    'is_final': finish_reason,
+                    'tool_calls': None if tool_call['id'] is None else [tool_call],
+                }
+                # if chunk_idx == 0:
+                #     chunk_idx += 1
+                #     continue
+
+                # assert type(chunk) is anthropic.types.message.Chunk
+
+                yield llm_entities.MessageChunk(**args)
+
+            # return llm_entities.Message(**args)
+        except anthropic.AuthenticationError as e:
+            raise errors.RequesterError(f'api-key 无效: {e.message}')
+        except anthropic.BadRequestError as e:
+            raise errors.RequesterError(str(e.message))
+        except anthropic.NotFoundError as e:
+            if 'model: ' in str(e):
+                raise errors.RequesterError(f'模型无效: {e.message}')
+            else:
+                raise errors.RequesterError(f'请求地址无效: {e.message}')

pkg/provider/modelmgr/requesters/anthropicmsgs.yaml

@@ -14,7 +14,7 @@ spec:
         zh_Hans: 基础 URL
       type: string
       required: true
-      default: "https://api.anthropic.com/v1"
+      default: "https://api.anthropic.com"
     - name: timeout
       label:
         en_US: Timeout
@@ -22,6 +22,8 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./anthropicmsgs.py

pkg/provider/modelmgr/requesters/bailianchatcmpl.yaml

@@ -22,6 +22,8 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./bailianchatcmpl.py

pkg/provider/modelmgr/requesters/chatcmpl.py

@@ -13,7 +13,7 @@ from ... import entities as llm_entities
 from ...tools import entities as tools_entities
 
 
-class OpenAIChatCompletions(requester.LLMAPIRequester):
+class OpenAIChatCompletions(requester.ProviderAPIRequester):
     """OpenAI ChatCompletion API 请求器"""
 
     client: openai.AsyncClient
@@ -38,9 +38,18 @@ class OpenAIChatCompletions(requester.LLMAPIRequester):
     ) -> chat_completion.ChatCompletion:
         return await self.client.chat.completions.create(**args, extra_body=extra_body)
 
+    async def _req_stream(
+        self,
+        args: dict,
+        extra_body: dict = {},
+    ):
+        async for chunk in await self.client.chat.completions.create(**args, extra_body=extra_body):
+            yield chunk
+
     async def _make_msg(
         self,
         chat_completion: chat_completion.ChatCompletion,
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         chatcmpl_message = chat_completion.choices[0].message.model_dump()
 
@@ -48,16 +57,192 @@ class OpenAIChatCompletions(requester.LLMAPIRequester):
         if 'role' not in chatcmpl_message or chatcmpl_message['role'] is None:
             chatcmpl_message['role'] = 'assistant'
 
-        reasoning_content = chatcmpl_message['reasoning_content'] if 'reasoning_content' in chatcmpl_message else None
+        # 处理思维链
+        content = chatcmpl_message.get('content', '')
+        reasoning_content = chatcmpl_message.get('reasoning_content', None)
 
-        # deepseek的reasoner模型
-        if reasoning_content is not None:
-            chatcmpl_message['content'] = '<think>\n' + reasoning_content + '\n</think>\n' + chatcmpl_message['content']
+        processed_content, _ = await self._process_thinking_content(
+            content=content, reasoning_content=reasoning_content, remove_think=remove_think
+        )
+
+        chatcmpl_message['content'] = processed_content
+
+        # 移除 reasoning_content 字段，避免传递给 Message
+        if 'reasoning_content' in chatcmpl_message:
+            del chatcmpl_message['reasoning_content']
 
         message = llm_entities.Message(**chatcmpl_message)
-
         return message
 
+    async def _process_thinking_content(
+        self,
+        content: str,
+        reasoning_content: str = None,
+        remove_think: bool = False,
+    ) -> tuple[str, str]:
+        """处理思维链内容
+
+        Args:
+            content: 原始内容
+            reasoning_content: reasoning_content 字段内容
+            remove_think: 是否移除思维链
+
+        Returns:
+            (处理后的内容, 提取的思维链内容)
+        """
+        thinking_content = ''
+
+        # 1. 从 reasoning_content 提取思维链
+        if reasoning_content:
+            thinking_content = reasoning_content
+
+        # 2. 从 content 中提取 <think> 标签内容
+        if content and '<think>' in content and '</think>' in content:
+            import re
+
+            think_pattern = r'<think>(.*?)</think>'
+            think_matches = re.findall(think_pattern, content, re.DOTALL)
+            if think_matches:
+                # 如果已有 reasoning_content，则追加
+                if thinking_content:
+                    thinking_content += '\n' + '\n'.join(think_matches)
+                else:
+                    thinking_content = '\n'.join(think_matches)
+                # 移除 content 中的 <think> 标签
+                content = re.sub(think_pattern, '', content, flags=re.DOTALL).strip()
+
+        # 3. 根据 remove_think 参数决定是否保留思维链
+        if remove_think:
+            return content, ''
+        else:
+            # 如果有思维链内容，将其以 <think> 格式添加到 content 开头
+            if thinking_content:
+                content = f'<think>\n{thinking_content}\n</think>\n{content}'.strip()
+            return content, thinking_content
+
+    async def _closure_stream(
+        self,
+        query: core_entities.Query,
+        req_messages: list[dict],
+        use_model: requester.RuntimeLLMModel,
+        use_funcs: list[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.MessageChunk:
+        self.client.api_key = use_model.token_mgr.get_token()
+
+        args = {}
+        args['model'] = use_model.model_entity.name
+
+        if use_funcs:
+            tools = await self.ap.tool_mgr.generate_tools_for_openai(use_funcs)
+            if tools:
+                args['tools'] = tools
+
+        # 设置此次请求中的messages
+        messages = req_messages.copy()
+
+        # 检查vision
+        for msg in messages:
+            if 'content' in msg and isinstance(msg['content'], list):
+                for me in msg['content']:
+                    if me['type'] == 'image_base64':
+                        me['image_url'] = {'url': me['image_base64']}
+                        me['type'] = 'image_url'
+                        del me['image_base64']
+
+        args['messages'] = messages
+        args['stream'] = True
+
+        # 流式处理状态
+        tool_calls_map: dict[str, llm_entities.ToolCall] = {}
+        chunk_idx = 0
+        thinking_started = False
+        thinking_ended = False
+        role = 'assistant'  # 默认角色
+        tool_id = ""
+        tool_name = ''
+        # accumulated_reasoning = ''  # 仅用于判断何时结束思维链
+
+        async for chunk in self._req_stream(args, extra_body=extra_args):
+            # 解析 chunk 数据
+
+            if hasattr(chunk, 'choices') and chunk.choices:
+                choice = chunk.choices[0]
+                delta = choice.delta.model_dump() if hasattr(choice, 'delta') else {}
+
+                finish_reason = getattr(choice, 'finish_reason', None)
+            else:
+                delta = {}
+                finish_reason = None
+            # 从第一个 chunk 获取 role，后续使用这个 role
+            if 'role' in delta and delta['role']:
+                role = delta['role']
+
+            # 获取增量内容
+            delta_content = delta.get('content', '')
+            reasoning_content = delta.get('reasoning_content', '')
+
+            # 处理 reasoning_content
+            if reasoning_content:
+                # accumulated_reasoning += reasoning_content
+                # 如果设置了 remove_think，跳过 reasoning_content
+                if remove_think:
+                    chunk_idx += 1
+                    continue
+
+                # 第一次出现 reasoning_content，添加 <think> 开始标签
+                if not thinking_started:
+                    thinking_started = True
+                    delta_content = '<think>\n' + reasoning_content
+                else:
+                    # 继续输出 reasoning_content
+                    delta_content = reasoning_content
+            elif thinking_started and not thinking_ended and delta_content:
+                # reasoning_content 结束，normal content 开始，添加 </think> 结束标签
+                thinking_ended = True
+                delta_content = '\n</think>\n' + delta_content
+
+            # 处理 content 中已有的 <think> 标签（如果需要移除）
+            # if delta_content and remove_think and '<think>' in delta_content:
+            #     import re
+            #
+            #     # 移除 <think> 标签及其内容
+            #     delta_content = re.sub(r'<think>.*?</think>', '', delta_content, flags=re.DOTALL)
+
+            # 处理工具调用增量
+            # delta_tool_calls = None
+            if delta.get('tool_calls'):
+                for tool_call in delta['tool_calls']:
+                    if tool_call['id'] and tool_call['function']['name']:
+                        tool_id = tool_call['id']
+                        tool_name = tool_call['function']['name']
+                    else:
+                        tool_call['id'] = tool_id
+                        tool_call['function']['name'] = tool_name
+                    if tool_call['type'] is None:
+                        tool_call['type'] = 'function'
+
+
+
+            # 跳过空的第一个 chunk（只有 role 没有内容）
+            if chunk_idx == 0 and not delta_content and not reasoning_content and not delta.get('tool_calls'):
+                chunk_idx += 1
+                continue
+            # 构建 MessageChunk - 只包含增量内容
+            chunk_data = {
+                'role': role,
+                'content': delta_content if delta_content else None,
+                'tool_calls': delta.get('tool_calls'),
+                'is_final': bool(finish_reason),
+            }
+
+            # 移除 None 值
+            chunk_data = {k: v for k, v in chunk_data.items() if v is not None}
+
+            yield llm_entities.MessageChunk(**chunk_data)
+            chunk_idx += 1
+
     async def _closure(
         self,
         query: core_entities.Query,
@@ -65,6 +250,7 @@ class OpenAIChatCompletions(requester.LLMAPIRequester):
         use_model: requester.RuntimeLLMModel,
         use_funcs: list[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         self.client.api_key = use_model.token_mgr.get_token()
 
@@ -92,10 +278,10 @@ class OpenAIChatCompletions(requester.LLMAPIRequester):
         args['messages'] = messages
 
         # 发送请求
-        resp = await self._req(args, extra_body=extra_args)
 
+        resp = await self._req(args, extra_body=extra_args)
         # 处理请求结果
-        message = await self._make_msg(resp)
+        message = await self._make_msg(resp, remove_think)
 
         return message
 
@@ -106,6 +292,7 @@ class OpenAIChatCompletions(requester.LLMAPIRequester):
         messages: typing.List[llm_entities.Message],
         funcs: typing.List[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         req_messages = []  # req_messages 仅用于类内，外部同步由 query.messages 进行
         for m in messages:
@@ -119,13 +306,90 @@ class OpenAIChatCompletions(requester.LLMAPIRequester):
             req_messages.append(msg_dict)
 
         try:
-            return await self._closure(
+            msg = await self._closure(
                 query=query,
                 req_messages=req_messages,
                 use_model=model,
                 use_funcs=funcs,
                 extra_args=extra_args,
+                remove_think=remove_think,
             )
+            return msg
+        except asyncio.TimeoutError:
+            raise errors.RequesterError('请求超时')
+        except openai.BadRequestError as e:
+            if 'context_length_exceeded' in e.message:
+                raise errors.RequesterError(f'上文过长，请重置会话: {e.message}')
+            else:
+                raise errors.RequesterError(f'请求参数错误: {e.message}')
+        except openai.AuthenticationError as e:
+            raise errors.RequesterError(f'无效的 api-key: {e.message}')
+        except openai.NotFoundError as e:
+            raise errors.RequesterError(f'请求路径错误: {e.message}')
+        except openai.RateLimitError as e:
+            raise errors.RequesterError(f'请求过于频繁或余额不足: {e.message}')
+        except openai.APIError as e:
+            raise errors.RequesterError(f'请求错误: {e.message}')
+
+    async def invoke_embedding(
+        self,
+        model: requester.RuntimeEmbeddingModel,
+        input_text: list[str],
+        extra_args: dict[str, typing.Any] = {},
+    ) -> list[list[float]]:
+        """调用 Embedding API"""
+        self.client.api_key = model.token_mgr.get_token()
+
+        args = {
+            'model': model.model_entity.name,
+            'input': input_text,
+        }
+
+        if model.model_entity.extra_args:
+            args.update(model.model_entity.extra_args)
+
+        args.update(extra_args)
+
+        try:
+            resp = await self.client.embeddings.create(**args)
+
+            return [d.embedding for d in resp.data]
+        except asyncio.TimeoutError:
+            raise errors.RequesterError('请求超时')
+        except openai.BadRequestError as e:
+            raise errors.RequesterError(f'请求参数错误: {e.message}')
+
+    async def invoke_llm_stream(
+        self,
+        query: core_entities.Query,
+        model: requester.RuntimeLLMModel,
+        messages: typing.List[llm_entities.Message],
+        funcs: typing.List[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.MessageChunk:
+        req_messages = []  # req_messages 仅用于类内，外部同步由 query.messages 进行
+        for m in messages:
+            msg_dict = m.dict(exclude_none=True)
+            content = msg_dict.get('content')
+            if isinstance(content, list):
+                # 检查 content 列表中是否每个部分都是文本
+                if all(isinstance(part, dict) and part.get('type') == 'text' for part in content):
+                    # 将所有文本部分合并为一个字符串
+                    msg_dict['content'] = '\n'.join(part['text'] for part in content)
+            req_messages.append(msg_dict)
+
+        try:
+            async for item in self._closure_stream(
+                query=query,
+                req_messages=req_messages,
+                use_model=model,
+                use_funcs=funcs,
+                extra_args=extra_args,
+                remove_think=remove_think,
+            ):
+                yield item
+
         except asyncio.TimeoutError:
             raise errors.RequesterError('请求超时')
         except openai.BadRequestError as e:

pkg/provider/modelmgr/requesters/chatcmpl.yaml

@@ -22,6 +22,9 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
+    - text-embedding
 execution:
   python:
     path: ./chatcmpl.py

pkg/provider/modelmgr/requesters/compsharechatcmpl.yaml

@@ -22,6 +22,8 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./compsharechatcmpl.py

pkg/provider/modelmgr/requesters/deepseekchatcmpl.py

@@ -24,6 +24,7 @@ class DeepseekChatCompletions(chatcmpl.OpenAIChatCompletions):
         use_model: requester.RuntimeLLMModel,
         use_funcs: list[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         self.client.api_key = use_model.token_mgr.get_token()
 
@@ -49,10 +50,11 @@ class DeepseekChatCompletions(chatcmpl.OpenAIChatCompletions):
         # 发送请求
         resp = await self._req(args, extra_body=extra_args)
 
+        # print(resp)
+
         if resp is None:
             raise errors.RequesterError('接口返回为空，请确定模型提供商服务是否正常')
-
         # 处理请求结果
-        message = await self._make_msg(resp)
+        message = await self._make_msg(resp, remove_think)
 
         return message

pkg/provider/modelmgr/requesters/deepseekchatcmpl.yaml

@@ -4,7 +4,7 @@ metadata:
   name: deepseek-chat-completions
   label:
     en_US: DeepSeek
-    zh_Hans: 深度求索
+    zh_Hans: DeepSeek
   icon: deepseek.svg
 spec:
   config:
@@ -22,6 +22,8 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./deepseekchatcmpl.py

pkg/provider/modelmgr/requesters/geminichatcmpl.py

@@ -4,6 +4,13 @@ import typing
 
 from . import chatcmpl
 
+import uuid
+
+from .. import errors, requester
+from ....core import entities as core_entities
+from ... import entities as llm_entities
+from ...tools import entities as tools_entities
+
 
 class GeminiChatCompletions(chatcmpl.OpenAIChatCompletions):
     """Google Gemini API 请求器"""
@@ -12,3 +19,127 @@ class GeminiChatCompletions(chatcmpl.OpenAIChatCompletions):
         'base_url': 'https://generativelanguage.googleapis.com/v1beta/openai',
         'timeout': 120,
     }
+
+
+    async def _closure_stream(
+        self,
+        query: core_entities.Query,
+        req_messages: list[dict],
+        use_model: requester.RuntimeLLMModel,
+        use_funcs: list[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.MessageChunk:
+        self.client.api_key = use_model.token_mgr.get_token()
+
+        args = {}
+        args['model'] = use_model.model_entity.name
+
+        if use_funcs:
+            tools = await self.ap.tool_mgr.generate_tools_for_openai(use_funcs)
+            if tools:
+                args['tools'] = tools
+
+        # 设置此次请求中的messages
+        messages = req_messages.copy()
+
+        # 检查vision
+        for msg in messages:
+            if 'content' in msg and isinstance(msg['content'], list):
+                for me in msg['content']:
+                    if me['type'] == 'image_base64':
+                        me['image_url'] = {'url': me['image_base64']}
+                        me['type'] = 'image_url'
+                        del me['image_base64']
+
+        args['messages'] = messages
+        args['stream'] = True
+
+        # 流式处理状态
+        tool_calls_map: dict[str, llm_entities.ToolCall] = {}
+        chunk_idx = 0
+        thinking_started = False
+        thinking_ended = False
+        role = 'assistant'  # 默认角色
+        tool_id = ""
+        tool_name = ''
+        # accumulated_reasoning = ''  # 仅用于判断何时结束思维链
+
+        async for chunk in self._req_stream(args, extra_body=extra_args):
+            # 解析 chunk 数据
+
+            if hasattr(chunk, 'choices') and chunk.choices:
+                choice = chunk.choices[0]
+                delta = choice.delta.model_dump() if hasattr(choice, 'delta') else {}
+
+                finish_reason = getattr(choice, 'finish_reason', None)
+            else:
+                delta = {}
+                finish_reason = None
+            # 从第一个 chunk 获取 role，后续使用这个 role
+            if 'role' in delta and delta['role']:
+                role = delta['role']
+
+            # 获取增量内容
+            delta_content = delta.get('content', '')
+            reasoning_content = delta.get('reasoning_content', '')
+
+            # 处理 reasoning_content
+            if reasoning_content:
+                # accumulated_reasoning += reasoning_content
+                # 如果设置了 remove_think，跳过 reasoning_content
+                if remove_think:
+                    chunk_idx += 1
+                    continue
+
+                # 第一次出现 reasoning_content，添加 <think> 开始标签
+                if not thinking_started:
+                    thinking_started = True
+                    delta_content = '<think>\n' + reasoning_content
+                else:
+                    # 继续输出 reasoning_content
+                    delta_content = reasoning_content
+            elif thinking_started and not thinking_ended and delta_content:
+                # reasoning_content 结束，normal content 开始，添加 </think> 结束标签
+                thinking_ended = True
+                delta_content = '\n</think>\n' + delta_content
+
+            # 处理 content 中已有的 <think> 标签（如果需要移除）
+            # if delta_content and remove_think and '<think>' in delta_content:
+            #     import re
+            #
+            #     # 移除 <think> 标签及其内容
+            #     delta_content = re.sub(r'<think>.*?</think>', '', delta_content, flags=re.DOTALL)
+
+            # 处理工具调用增量
+            # delta_tool_calls = None
+            if delta.get('tool_calls'):
+                for tool_call in delta['tool_calls']:
+                    if tool_call['id'] == '' and tool_id == '':
+                        tool_id = str(uuid.uuid4())
+                    if  tool_call['function']['name']:
+                        tool_name = tool_call['function']['name']
+                    tool_call['id'] = tool_id
+                    tool_call['function']['name'] = tool_name
+                    if tool_call['type'] is None:
+                        tool_call['type'] = 'function'
+
+
+
+            # 跳过空的第一个 chunk（只有 role 没有内容）
+            if chunk_idx == 0 and not delta_content and not reasoning_content and not delta.get('tool_calls'):
+                chunk_idx += 1
+                continue
+            # 构建 MessageChunk - 只包含增量内容
+            chunk_data = {
+                'role': role,
+                'content': delta_content if delta_content else None,
+                'tool_calls': delta.get('tool_calls'),
+                'is_final': bool(finish_reason),
+            }
+
+            # 移除 None 值
+            chunk_data = {k: v for k, v in chunk_data.items() if v is not None}
+
+            yield llm_entities.MessageChunk(**chunk_data)
+            chunk_idx += 1

pkg/provider/modelmgr/requesters/geminichatcmpl.yaml

@@ -22,6 +22,8 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./geminichatcmpl.py

pkg/provider/modelmgr/requesters/giteeaichatcmpl.py

@@ -3,14 +3,16 @@ from __future__ import annotations
 
 import typing
 
-from . import chatcmpl
+from . import ppiochatcmpl
 from .. import requester
 from ....core import entities as core_entities
 from ... import entities as llm_entities
 from ...tools import entities as tools_entities
+import re
+import openai.types.chat.chat_completion as chat_completion
 
 
-class GiteeAIChatCompletions(chatcmpl.OpenAIChatCompletions):
+class GiteeAIChatCompletions(ppiochatcmpl.PPIOChatCompletions):
     """Gitee AI ChatCompletions API 请求器"""
 
     default_config: dict[str, typing.Any] = {
@@ -18,34 +20,3 @@ class GiteeAIChatCompletions(chatcmpl.OpenAIChatCompletions):
         'timeout': 120,
     }
 
-    async def _closure(
-        self,
-        query: core_entities.Query,
-        req_messages: list[dict],
-        use_model: requester.RuntimeLLMModel,
-        use_funcs: list[tools_entities.LLMFunction] = None,
-        extra_args: dict[str, typing.Any] = {},
-    ) -> llm_entities.Message:
-        self.client.api_key = use_model.token_mgr.get_token()
-
-        args = {}
-        args['model'] = use_model.model_entity.name
-
-        if use_funcs:
-            tools = await self.ap.tool_mgr.generate_tools_for_openai(use_funcs)
-
-            if tools:
-                args['tools'] = tools
-
-        # gitee 不支持多模态，把content都转换成纯文字
-        for m in req_messages:
-            if 'content' in m and isinstance(m['content'], list):
-                m['content'] = ' '.join([c['text'] for c in m['content']])
-
-        args['messages'] = req_messages
-
-        resp = await self._req(args, extra_body=extra_args)
-
-        message = await self._make_msg(resp)
-
-        return message

pkg/provider/modelmgr/requesters/giteeaichatcmpl.yaml

@@ -22,6 +22,9 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
+    - text-embedding
 execution:
   python:
     path: ./giteeaichatcmpl.py

pkg/provider/modelmgr/requesters/lmstudiochatcmpl.yaml

@@ -22,6 +22,9 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
+    - text-embedding
 execution:
   python:
     path: ./lmstudiochatcmpl.py

pkg/provider/modelmgr/requesters/modelscopechatcmpl.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import asyncio
+import json
 import typing
 
 import openai
@@ -14,7 +15,7 @@ from ... import entities as llm_entities
 from ...tools import entities as tools_entities
 
 
-class ModelScopeChatCompletions(requester.LLMAPIRequester):
+class ModelScopeChatCompletions(requester.ProviderAPIRequester):
     """ModelScope ChatCompletion API 请求器"""
 
     client: openai.AsyncClient
@@ -34,9 +35,11 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
 
     async def _req(
         self,
+        query: core_entities.Query,
         args: dict,
         extra_body: dict = {},
-    ) -> chat_completion.ChatCompletion:
+        remove_think: bool = False,
+    ) -> list[dict[str, typing.Any]]:
         args['stream'] = True
 
         chunk = None
@@ -47,78 +50,75 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
 
         resp_gen: openai.AsyncStream = await self.client.chat.completions.create(**args, extra_body=extra_body)
 
+        chunk_idx = 0
+        thinking_started = False
+        thinking_ended = False
+        tool_id = ''
+        tool_name = ''
+        message_delta = {}
         async for chunk in resp_gen:
-            # print(chunk)
             if not chunk or not chunk.id or not chunk.choices or not chunk.choices[0] or not chunk.choices[0].delta:
                 continue
 
-            if chunk.choices[0].delta.content is not None:
-                pending_content += chunk.choices[0].delta.content
+            delta = chunk.choices[0].delta.model_dump() if hasattr(chunk.choices[0], 'delta') else {}
+            reasoning_content = delta.get('reasoning_content')
+            # 处理 reasoning_content
+            if reasoning_content:
+                # accumulated_reasoning += reasoning_content
+                # 如果设置了 remove_think，跳过 reasoning_content
+                if remove_think:
+                    chunk_idx += 1
+                    continue
 
-            if chunk.choices[0].delta.tool_calls is not None:
-                for tool_call in chunk.choices[0].delta.tool_calls:
-                    if tool_call.function.arguments is None:
+                # 第一次出现 reasoning_content，添加 <think> 开始标签
+                if not thinking_started:
+                    thinking_started = True
+                    pending_content += '<think>\n' + reasoning_content
+                else:
+                    # 继续输出 reasoning_content
+                    pending_content += reasoning_content
+            elif thinking_started and not thinking_ended and delta.get('content'):
+                # reasoning_content 结束，normal content 开始，添加 </think> 结束标签
+                thinking_ended = True
+                pending_content += '\n</think>\n' + delta.get('content')
+
+            if delta.get('content') is not None:
+                pending_content += delta.get('content')
+
+            if delta.get('tool_calls') is not None:
+                for tool_call in delta.get('tool_calls'):
+                    if tool_call['id'] != '':
+                        tool_id = tool_call['id']
+                    if tool_call['function']['name'] is not None:
+                        tool_name = tool_call['function']['name']
+                    if tool_call['function']['arguments'] is None:
                         continue
+                    tool_call['id'] = tool_id
+                    tool_call['name'] = tool_name
                     for tc in tool_calls:
-                        if tc.index == tool_call.index:
-                            tc.function.arguments += tool_call.function.arguments
+                        if tc['index'] == tool_call['index']:
+                            tc['function']['arguments'] += tool_call['function']['arguments']
                             break
                     else:
                         tool_calls.append(tool_call)
 
             if chunk.choices[0].finish_reason is not None:
                 break
+        message_delta['content'] = pending_content
+        message_delta['role'] = 'assistant'
 
-        real_tool_calls = []
-
-        for tc in tool_calls:
-            function = chat_completion_message_tool_call.Function(
-                name=tc.function.name, arguments=tc.function.arguments
-            )
-            real_tool_calls.append(
-                chat_completion_message_tool_call.ChatCompletionMessageToolCall(
-                    id=tc.id, function=function, type='function'
-                )
-            )
-
-        return (
-            chat_completion.ChatCompletion(
-                id=chunk.id,
-                object='chat.completion',
-                created=chunk.created,
-                choices=[
-                    chat_completion.Choice(
-                        index=0,
-                        message=chat_completion.ChatCompletionMessage(
-                            role='assistant',
-                            content=pending_content,
-                            tool_calls=real_tool_calls if len(real_tool_calls) > 0 else None,
-                        ),
-                        finish_reason=chunk.choices[0].finish_reason
-                        if hasattr(chunk.choices[0], 'finish_reason') and chunk.choices[0].finish_reason is not None
-                        else 'stop',
-                        logprobs=chunk.choices[0].logprobs,
-                    )
-                ],
-                model=chunk.model,
-                service_tier=chunk.service_tier if hasattr(chunk, 'service_tier') else None,
-                system_fingerprint=chunk.system_fingerprint if hasattr(chunk, 'system_fingerprint') else None,
-                usage=chunk.usage if hasattr(chunk, 'usage') else None,
-            )
-            if chunk
-            else None
-        )
+        message_delta['tool_calls'] = tool_calls if tool_calls else None
+        return [message_delta]
 
     async def _make_msg(
         self,
-        chat_completion: chat_completion.ChatCompletion,
+        chat_completion: list[dict[str, typing.Any]],
     ) -> llm_entities.Message:
-        chatcmpl_message = chat_completion.choices[0].message.dict()
+        chatcmpl_message = chat_completion[0]
 
         # 确保 role 字段存在且不为 None
         if 'role' not in chatcmpl_message or chatcmpl_message['role'] is None:
             chatcmpl_message['role'] = 'assistant'
-
         message = llm_entities.Message(**chatcmpl_message)
 
         return message
@@ -130,6 +130,7 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
         use_model: requester.RuntimeLLMModel,
         use_funcs: list[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think:bool = False,
     ) -> llm_entities.Message:
         self.client.api_key = use_model.token_mgr.get_token()
 
@@ -157,13 +158,146 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
         args['messages'] = messages
 
         # 发送请求
-        resp = await self._req(args, extra_body=extra_args)
+        resp = await self._req(query, args, extra_body=extra_args, remove_think=remove_think)
 
         # 处理请求结果
         message = await self._make_msg(resp)
 
         return message
 
+    async def _req_stream(
+        self,
+        args: dict,
+        extra_body: dict = {},
+    ) -> chat_completion.ChatCompletion:
+        async for chunk in await self.client.chat.completions.create(**args, extra_body=extra_body):
+            yield chunk
+
+
+    async def _closure_stream(
+        self,
+        query: core_entities.Query,
+        req_messages: list[dict],
+        use_model: requester.RuntimeLLMModel,
+        use_funcs: list[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.Message | typing.AsyncGenerator[llm_entities.MessageChunk, None]:
+        self.client.api_key = use_model.token_mgr.get_token()
+
+        args = {}
+        args['model'] = use_model.model_entity.name
+
+        if use_funcs:
+            tools = await self.ap.tool_mgr.generate_tools_for_openai(use_funcs)
+
+            if tools:
+                args['tools'] = tools
+
+        # 设置此次请求中的messages
+        messages = req_messages.copy()
+
+        # 检查vision
+        for msg in messages:
+            if 'content' in msg and isinstance(msg['content'], list):
+                for me in msg['content']:
+                    if me['type'] == 'image_base64':
+                        me['image_url'] = {'url': me['image_base64']}
+                        me['type'] = 'image_url'
+                        del me['image_base64']
+
+        args['messages'] = messages
+        args['stream'] = True
+
+
+        # 流式处理状态
+        tool_calls_map: dict[str, llm_entities.ToolCall] = {}
+        chunk_idx = 0
+        thinking_started = False
+        thinking_ended = False
+        role = 'assistant'  # 默认角色
+        # accumulated_reasoning = ''  # 仅用于判断何时结束思维链
+
+        async for chunk in self._req_stream(args, extra_body=extra_args):
+            # 解析 chunk 数据
+            if hasattr(chunk, 'choices') and chunk.choices:
+                choice = chunk.choices[0]
+                delta = choice.delta.model_dump() if hasattr(choice, 'delta') else {}
+                finish_reason = getattr(choice, 'finish_reason', None)
+            else:
+                delta = {}
+                finish_reason = None
+
+            # 从第一个 chunk 获取 role，后续使用这个 role
+            if 'role' in delta and delta['role']:
+                role = delta['role']
+
+            # 获取增量内容
+            delta_content = delta.get('content', '')
+            reasoning_content = delta.get('reasoning_content', '')
+
+            # 处理 reasoning_content
+            if reasoning_content:
+                # accumulated_reasoning += reasoning_content
+                # 如果设置了 remove_think，跳过 reasoning_content
+                if remove_think:
+                    chunk_idx += 1
+                    continue
+
+                # 第一次出现 reasoning_content，添加 <think> 开始标签
+                if not thinking_started:
+                    thinking_started = True
+                    delta_content = '<think>\n' + reasoning_content
+                else:
+                    # 继续输出 reasoning_content
+                    delta_content = reasoning_content
+            elif thinking_started and not thinking_ended and delta_content:
+                # reasoning_content 结束，normal content 开始，添加 </think> 结束标签
+                thinking_ended = True
+                delta_content = '\n</think>\n' + delta_content
+
+            # 处理 content 中已有的 <think> 标签（如果需要移除）
+            # if delta_content and remove_think and '<think>' in delta_content:
+            #     import re
+            #
+            #     # 移除 <think> 标签及其内容
+            #     delta_content = re.sub(r'<think>.*?</think>', '', delta_content, flags=re.DOTALL)
+
+            # 处理工具调用增量
+            if delta.get('tool_calls'):
+                for tool_call in delta['tool_calls']:
+                    if tool_call['id'] != '':
+                        tool_id = tool_call['id']
+                    if tool_call['function']['name'] is not None:
+                        tool_name = tool_call['function']['name']
+
+                    if tool_call['type'] is None:
+                        tool_call['type'] = 'function'
+                    tool_call['id'] = tool_id
+                    tool_call['function']['name'] = tool_name
+                    tool_call['function']['arguments'] = "" if tool_call['function']['arguments'] is None else tool_call['function']['arguments']
+
+
+            # 跳过空的第一个 chunk（只有 role 没有内容）
+            if chunk_idx == 0 and not delta_content and not reasoning_content and not delta.get('tool_calls'):
+                chunk_idx += 1
+                continue
+
+            # 构建 MessageChunk - 只包含增量内容
+            chunk_data = {
+                'role': role,
+                'content': delta_content if delta_content else None,
+                'tool_calls': delta.get('tool_calls'),
+                'is_final': bool(finish_reason),
+            }
+
+            # 移除 None 值
+            chunk_data = {k: v for k, v in chunk_data.items() if v is not None}
+
+            yield llm_entities.MessageChunk(**chunk_data)
+            chunk_idx += 1
+            # return
+
     async def invoke_llm(
         self,
         query: core_entities.Query,
@@ -171,6 +305,7 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
         messages: typing.List[llm_entities.Message],
         funcs: typing.List[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         req_messages = []  # req_messages 仅用于类内，外部同步由 query.messages 进行
         for m in messages:
@@ -185,7 +320,7 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
 
         try:
             return await self._closure(
-                query=query, req_messages=req_messages, use_model=model, use_funcs=funcs, extra_args=extra_args
+                query=query, req_messages=req_messages, use_model=model, use_funcs=funcs, extra_args=extra_args, remove_think=remove_think
             )
         except asyncio.TimeoutError:
             raise errors.RequesterError('请求超时')
@@ -202,3 +337,50 @@ class ModelScopeChatCompletions(requester.LLMAPIRequester):
             raise errors.RequesterError(f'请求过于频繁或余额不足: {e.message}')
         except openai.APIError as e:
             raise errors.RequesterError(f'请求错误: {e.message}')
+
+    async def invoke_llm_stream(
+        self,
+        query: core_entities.Query,
+        model: requester.RuntimeLLMModel,
+        messages: typing.List[llm_entities.Message],
+        funcs: typing.List[tools_entities.LLMFunction] = None,
+        extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
+    ) -> llm_entities.MessageChunk:
+        req_messages = []  # req_messages 仅用于类内，外部同步由 query.messages 进行
+        for m in messages:
+            msg_dict = m.dict(exclude_none=True)
+            content = msg_dict.get('content')
+            if isinstance(content, list):
+                # 检查 content 列表中是否每个部分都是文本
+                if all(isinstance(part, dict) and part.get('type') == 'text' for part in content):
+                    # 将所有文本部分合并为一个字符串
+                    msg_dict['content'] = '\n'.join(part['text'] for part in content)
+            req_messages.append(msg_dict)
+
+        try:
+            async for item in self._closure_stream(
+                query=query,
+                req_messages=req_messages,
+                use_model=model,
+                use_funcs=funcs,
+                extra_args=extra_args,
+                remove_think=remove_think,
+            ):
+                yield item
+
+        except asyncio.TimeoutError:
+            raise errors.RequesterError('请求超时')
+        except openai.BadRequestError as e:
+            if 'context_length_exceeded' in e.message:
+                raise errors.RequesterError(f'上文过长，请重置会话: {e.message}')
+            else:
+                raise errors.RequesterError(f'请求参数错误: {e.message}')
+        except openai.AuthenticationError as e:
+            raise errors.RequesterError(f'无效的 api-key: {e.message}')
+        except openai.NotFoundError as e:
+            raise errors.RequesterError(f'请求路径错误: {e.message}')
+        except openai.RateLimitError as e:
+            raise errors.RequesterError(f'请求过于频繁或余额不足: {e.message}')
+        except openai.APIError as e:
+            raise errors.RequesterError(f'请求错误: {e.message}')

pkg/provider/modelmgr/requesters/modelscopechatcmpl.yaml

@@ -29,6 +29,8 @@ spec:
       type: int
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./modelscopechatcmpl.py

pkg/provider/modelmgr/requesters/moonshotchatcmpl.py

@@ -25,6 +25,7 @@ class MoonshotChatCompletions(chatcmpl.OpenAIChatCompletions):
         use_model: requester.RuntimeLLMModel,
         use_funcs: list[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         self.client.api_key = use_model.token_mgr.get_token()
 
@@ -54,6 +55,6 @@ class MoonshotChatCompletions(chatcmpl.OpenAIChatCompletions):
         resp = await self._req(args, extra_body=extra_args)
 
         # 处理请求结果
-        message = await self._make_msg(resp)
+        message = await self._make_msg(resp, remove_think)
 
         return message

pkg/provider/modelmgr/requesters/moonshotchatcmpl.yaml

@@ -14,7 +14,7 @@ spec:
         zh_Hans: 基础 URL
       type: string
       required: true
-      default: "https://api.moonshot.com/v1"
+      default: "https://api.moonshot.ai/v1"
     - name: timeout
       label:
         en_US: Timeout
@@ -22,6 +22,8 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
 execution:
   python:
     path: ./moonshotchatcmpl.py

pkg/provider/modelmgr/requesters/newapichatcmpl.py

@@ -0,0 +1,17 @@
+from __future__ import annotations
+
+import typing
+import openai
+
+from . import chatcmpl
+
+
+class NewAPIChatCompletions(chatcmpl.OpenAIChatCompletions):
+    """New API ChatCompletion API 请求器"""
+
+    client: openai.AsyncClient
+
+    default_config: dict[str, typing.Any] = {
+        'base_url': 'http://localhost:3000/v1',
+        'timeout': 120,
+    }

pkg/provider/modelmgr/requesters/newapichatcmpl.yaml

@@ -0,0 +1,31 @@
+apiVersion: v1
+kind: LLMAPIRequester
+metadata:
+  name: new-api-chat-completions
+  label:
+    en_US: New API
+    zh_Hans: New API
+  icon: newapi.png
+spec:
+  config:
+    - name: base_url
+      label:
+        en_US: Base URL
+        zh_Hans: 基础 URL
+      type: string
+      required: true
+      default: "http://localhost:3000/v1"
+    - name: timeout
+      label:
+        en_US: Timeout
+        zh_Hans: 超时时间
+      type: integer
+      required: true
+      default: 120
+  support_type:
+    - llm
+    - text-embedding
+execution:
+  python:
+    path: ./newapichatcmpl.py
+    attr: NewAPIChatCompletions

pkg/provider/modelmgr/requesters/ollamachat.py

@@ -17,7 +17,7 @@ from ....core import entities as core_entities
 REQUESTER_NAME: str = 'ollama-chat'
 
 
-class OllamaChatCompletions(requester.LLMAPIRequester):
+class OllamaChatCompletions(requester.ProviderAPIRequester):
     """Ollama平台 ChatCompletion API请求器"""
 
     client: ollama.AsyncClient
@@ -44,6 +44,7 @@ class OllamaChatCompletions(requester.LLMAPIRequester):
         use_model: requester.RuntimeLLMModel,
         use_funcs: list[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         args = extra_args.copy()
         args['model'] = use_model.model_entity.name
@@ -110,6 +111,7 @@ class OllamaChatCompletions(requester.LLMAPIRequester):
         messages: typing.List[llm_entities.Message],
         funcs: typing.List[tools_entities.LLMFunction] = None,
         extra_args: dict[str, typing.Any] = {},
+        remove_think: bool = False,
     ) -> llm_entities.Message:
         req_messages: list = []
         for m in messages:
@@ -126,6 +128,19 @@ class OllamaChatCompletions(requester.LLMAPIRequester):
                 use_model=model,
                 use_funcs=funcs,
                 extra_args=extra_args,
+                remove_think=remove_think,
             )
         except asyncio.TimeoutError:
             raise errors.RequesterError('请求超时')
+
+    async def invoke_embedding(
+        self,
+        model: requester.RuntimeEmbeddingModel,
+        input_text: list[str],
+        extra_args: dict[str, typing.Any] = {},
+    ) -> list[list[float]]:
+        return (await self.client.embed(
+            model=model.model_entity.name,
+            input=input_text,
+            **extra_args,
+        )).embeddings

pkg/provider/modelmgr/requesters/ollamachat.yaml

@@ -22,6 +22,9 @@ spec:
       type: integer
       required: true
       default: 120
+  support_type:
+    - llm
+    - text-embedding
 execution:
   python:
     path: ./ollamachat.py