feat: implement retry mechanism for model requests in anthropic and openai providers

chore: ruff format
chore: bump version to 4.17.1
2026-02-17 13:18:00 +08:00 · 2026-02-17 02:31:38 +08:00 · 2026-02-17 02:30:13 +08:00 · 2026-02-17 02:27:07 +08:00 · 2026-02-17 02:19:36 +08:00 · 2026-02-16 17:06:56 +08:00
140 changed files with 7905 additions and 1139 deletions
@@ -16,7 +16,7 @@ jobs:
      - name: Setup Node.js
        uses: actions/setup-node@v6
        with:
-          node-version: 'latest'
+          node-version: '24.13.0'

      - name: npm install, build
        run: |
@@ -52,4 +52,4 @@ jobs:
          repo: astrbot-release-harbour
          body: "Automated release from commit ${{ github.sha }}"
          token: ${{ secrets.ASTRBOT_HARBOUR_TOKEN }}
-          artifacts: "dashboard/dist.zip"
+          artifacts: "dashboard/dist.zip"
@@ -57,7 +57,7 @@ jobs:
      - name: Setup Node.js
        uses: actions/setup-node@v6
        with:
-          node-version: 20
+          node-version: '24.13.0'
          cache: "pnpm"
          cache-dependency-path: dashboard/pnpm-lock.yaml

@@ -175,7 +175,7 @@ jobs:
      - name: Setup Node.js
        uses: actions/setup-node@v6
        with:
-          node-version: 20
+          node-version: '24.13.0'
          cache: "pnpm"
          cache-dependency-path: |
            dashboard/pnpm-lock.yaml
@@ -15,17 +15,17 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
    curl \
    gnupg \
    git \
+    && curl -fsSL https://deb.nodesource.com/setup_lts.x | bash - \
+    && apt-get install -y --no-install-recommends nodejs \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*

-RUN apt-get update && apt-get install -y curl gnupg \
-    && curl -fsSL https://deb.nodesource.com/setup_lts.x | bash - \
-    && apt-get install -y nodejs
-
 RUN python -m pip install uv \
-    && echo "3.12" > .python-version
-RUN uv pip install -r requirements.txt --no-cache-dir --system
-RUN uv pip install socksio uv pilk --no-cache-dir --system
+    && echo "3.12" > .python-version \
+    && uv lock \
+    && uv export --format requirements.txt --output-file requirements.txt --frozen \
+    && uv pip install -r requirements.txt --no-cache-dir --system \
+    && uv pip install socksio uv pilk --no-cache-dir --system

 EXPOSE 6185

@@ -0,0 +1,14 @@
+## Welcome to AstrBot
+
+🌟 Thank you for using AstrBot!
+
+AstrBot is an Agentic AI assistant for personal and group chats, with support for multiple IM platforms and a wide range of built-in features. We hope it brings you an efficient and enjoyable experience. ❤️
+
+Important notice:
+
+AstrBot is a **free and open-source software project** protected by the AGPLv3 license. You can find the full source code and related resources on our [**official website**](https://astrbot.app) and [**GitHub**](https://github.com/astrbotdevs/astrbot).
+As of now, AstrBot has **no commercial services of any kind**, and the official team **will never charge users any fees** under any name.
+
+If anyone asks you to pay while using AstrBot, **you are likely being scammed**. Please request a refund immediately and report it to us by email.
+
+📮 Official email: [community@astrbot.app](mailto:community@astrbot.app)
@@ -0,0 +1,14 @@
+## 欢迎使用 AstrBot
+
+🌟 感谢您使用 AstrBot！
+
+AstrBot 是一款可接入多种 IM 平台的 Agentic AI 个人 / 群聊助手，内置多项强大功能，希望能为您带来高效、愉快的使用体验。❤️
+
+我们想特别说明：
+
+AstrBot 是受 AGPLv3 开源协议保护的**免费开源软件项目**，您可以在[**官方网站**](https://astrbot.app)、[**GitHub**](https://github.com/astrbotdevs/astrbot) 上找到 AstrBot 的全部源代码及相关资源。
+截至目前，AstrBot 项目**未开展任何形式的商业化服务**，官方**不会以任何名义向用户收取费用**。
+
+如果您在使用 AstrBot 的过程中被要求付费，**表明您已经遭遇诈骗行为**。请立即向相关方申请退款，并及时通过邮件向我们反馈。
+
+📮 官方邮箱：[community@astrbot.app](mailto:community@astrbot.app)
@@ -2,7 +2,6 @@

 <div align="center">

-
 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
@@ -41,14 +40,14 @@ AstrBot 是一个开源的一站式 Agentic 个人和群聊助手，可在 QQ、
 ## 主要功能

 1. 💯 免费 & 开源。
-1. ✨ AI 大模型对话，多模态，Agent，MCP，Skills，知识库，人格设定，自动压缩对话。
-2. 🤖 支持接入 Dify、阿里云百炼、Coze 等智能体平台。
-2. 🌐 多平台，支持 QQ、企业微信、飞书、钉钉、微信公众号、Telegram、Slack 以及[更多](#支持的消息平台)。
-3. 📦 插件扩展，已有近 800 个插件可一键安装。
-5. 🛡️ [Agent Sandbox](https://docs.astrbot.app/use/astrbot-agent-sandbox.html) 隔离化环境，安全地执行任何代码、调用 Shell、会话级资源复用。
-6. 💻 WebUI 支持。
-7. 🌈 Web ChatUI 支持，ChatUI 内置代理沙盒、网页搜索等。
-8. 🌐 国际化（i18n）支持。
+2. ✨ AI 大模型对话，多模态，Agent，MCP，Skills，知识库，人格设定，自动压缩对话。
+3. 🤖 支持接入 Dify、阿里云百炼、Coze 等智能体平台。
+4. 🌐 多平台，支持 QQ、企业微信、飞书、钉钉、微信公众号、Telegram、Slack 以及[更多](#支持的消息平台)。
+5. 📦 插件扩展，已有近 800 个插件可一键安装。
+6. 🛡️ [Agent Sandbox](https://docs.astrbot.app/use/astrbot-agent-sandbox.html) 隔离化环境，安全地执行任何代码、调用 Shell、会话级资源复用。
+7. 💻 WebUI 支持。
+8. 🌈 Web ChatUI 支持，ChatUI 内置代理沙盒、网页搜索等。
+9. 🌐 国际化（i18n）支持。

 <br>

@@ -78,9 +77,14 @@ AstrBot 是一个开源的一站式 Agentic 个人和群聊助手，可在 QQ、
 #### uv 部署

 ```bash
-uvx astrbot
+uv tool install astrbot
+astrbot
 ```

+#### 启动器一键部署（AstrBot Launcher）
+
+进入 [AstrBot Launcher](https://github.com/Raven95676/astrbot-launcher) 仓库，在 Releases 页最新版本下找到对应的系统安装包安装即可。
+
 #### 宝塔面板部署

 AstrBot 与宝塔面板合作，已上架至宝塔面板。
@@ -132,6 +136,16 @@ uv run main.py

 或者请参阅官方文档 [通过源码部署 AstrBot](https://astrbot.app/deploy/astrbot/cli.html) 。

+#### 系统包管理器安装
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# 或者使用 paru
+paru -S astrbot-git
+```
+
 #### 桌面端 Electron 打包

 桌面端（Electron 打包，`pnpm` 工作流）构建流程请参阅：[`desktop/README.md`](desktop/README.md)。
@@ -264,8 +278,6 @@ pre-commit install

 </div>

-</details>
-
 <div align="center">

 _陪伴与能力从来不应该是对立面。我们希望创造的是一个既能理解情绪、给予陪伴，也能可靠完成工作的机器人。_
@@ -273,3 +285,5 @@ _陪伴与能力从来不应该是对立面。我们希望创造的是一个既
 _私は、高性能ですから!_

 <img src="https://files.astrbot.app/watashiwa-koseino-desukara.gif" width="100"/>
+
+</div>
@@ -3,7 +3,6 @@
 <div align="center">

 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
 <a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a> ｜
@@ -52,6 +51,23 @@ AstrBot is an open-source all-in-one Agent chatbot platform that integrates with
 8. 🌈 Web ChatUI Support with built-in agent sandbox and web search.
 9. 🌐 Internationalization (i18n) Support.

+<br>
+
+<table align="center">
+  <tr align="center">
+    <th>💙 Role-playing & Emotional Companionship</th>
+    <th>✨ Proactive Agent</th>
+    <th>🚀 General Agentic Capabilities</th>
+    <th>🧩 900+ Community Plugins</th>
+  </tr>
+  <tr>
+    <td align="center"><p align="center"><img width="984" height="1746" alt="99b587c5d35eea09d84f33e6cf6cfd4f" src="https://github.com/user-attachments/assets/89196061-3290-458d-b51f-afa178049f84" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1612" alt="c449acd838c41d0915cc08a3824025b1" src="https://github.com/user-attachments/assets/f75368b4-e022-41dc-a9e0-131c3e73e32e" /></p></td>
+    <td align="center"><p align="center"><img width="974" height="1732" alt="image" src="https://github.com/user-attachments/assets/e22a3968-87d7-4708-a7cd-e7f198c7c32e" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1734" alt="image" src="https://github.com/user-attachments/assets/0952b395-6b4a-432a-8a50-c294b7f89750" /></p></td>
+  </tr>
+</table>
+
 ## Quick Start

 #### Docker Deployment (Recommended 🥳)
@@ -63,7 +79,18 @@ Please refer to the official documentation: [Deploy AstrBot with Docker](https:/
 #### uv Deployment

 ```bash
-uvx astrbot
+uv tool install astrbot
+astrbot
+```
+
+#### System Package Manager Installation
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# or use paru
+paru -S astrbot-git
 ```

 #### BT-Panel Deployment
@@ -117,6 +144,16 @@ uv run main.py

 Or refer to the official documentation: [Deploy AstrBot from Source](https://astrbot.app/deploy/astrbot/cli.html).

+#### System Package Manager Installation 
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# or use paru
+paru -S astrbot-git
+```
+
 #### Desktop Electron Build

 For desktop build steps (Electron packaging, `pnpm` workflow), see [`desktop/README.md`](desktop/README.md).
@@ -159,7 +196,7 @@ For desktop build steps (Electron packaging, `pnpm` workflow), see [`desktop/REA
 - [CompShare](https://www.compshare.cn/?ytag=GPU_YY-gh_astrbot&referral_code=FV7DcGowN4hB5UuXKgpE74)
 - [302.AI](https://share.302.ai/rr1M3l)
 - [TokenPony](https://www.tokenpony.cn/3YPyf)
- [SiliconFlow](https://docs.siliconflow.cn/cn/usecases/use-siliconcloud-in-astrbot)
+- [SiliconFlow](https://docs.siliconflow.cn/cn/usercases/use-siliconcloud-in-astrbot)
 - [PPIO Cloud](https://ppio.com/user/register?invited_by=AIOONE)
 - ModelScope
 - OneAPI
@@ -249,10 +286,10 @@ Additionally, the birth of this project would not have been possible without the

 </div>

-</details>
-
 <div align="center">

+_Companionship and capability should never be at odds. What we aim to create is a robot that can understand emotions, provide genuine companionship, and reliably accomplish tasks._
+
 _私は、高性能ですから!_

 <img src="https://files.astrbot.app/watashiwa-koseino-desukara.gif" width="100"/>
@@ -1,9 +1,13 @@
 ![AstrBot-Logo-Simplified](https://github.com/user-attachments/assets/ffd99b6b-3272-4682-beaa-6fe74250f7d9)

-</p>
-
 <div align="center">

+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ru.md">Русский</a>
+
 <br>

 <div>
@@ -14,22 +18,17 @@
 <br>

 <div>
-<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?style=for-the-badge&color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
-<img src="https://img.shields.io/badge/python-3.10+-blue.svg?style=for-the-badge&color=76bad9" alt="python">
-<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?style=for-the-badge&color=76bad9"/></a>
-<a href="https://qm.qq.com/cgi-bin/qm/qr?k=wtbaNx7EioxeaqS9z7RQWVXPIxg2zYr7&jump_from=webapi&authKey=vlqnv/AV2DbJEvGIcxdlNSpfxVy+8vVqijgreRdnVKOaydpc+YSw4MctmEbr0k5"><img alt="QQ_community" src="https://img.shields.io/badge/QQ群-775869627-purple?style=for-the-badge&color=76bad9"></a>
-<a href="https://t.me/+hAsD2Ebl5as3NmY1"><img alt="Telegram_community" src="https://img.shields.io/badge/Telegram-AstrBot-purple?style=for-the-badge&color=76bad9"></a>
-<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%20plugins&style=for-the-badge&label=Marketplace&cacheSeconds=3600">
+<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
+<img src="https://img.shields.io/badge/python-3.10+-blue.svg" alt="python">
+<img src="https://deepwiki.com/badge.svg" href="https://deepwiki.com/AstrBotDevs/AstrBot">
+<a href="https://zread.ai/AstrBotDevs/AstrBot" target="_blank"><img src="https://img.shields.io/badge/Ask_Zread-_.svg?style=flat&color=00b0aa&labelColor=000000&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB3aWR0aD0iMTYiIGhlaWdodD0iMTYiIHZpZXdCb3g9IjAgMCAxNiAxNiIgZmlsbD0ibm9uZSIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KPHBhdGggZD0iTTQuOTYxNTYgMS42MDAxSDIuMjQxNTZDMS44ODgxIDEuNjAwMSAxLjYwMTU2IDEuODg2NjQgMS42MDE1NiAyLjI0MDFWNC45NjAxQzEuNjAxNTYgNS4zMTM1NiAxLjg4ODEgNS42MDAxIDIuMjQxNTYgNS42MDAxSDQuOTYxNTZDNS4zMTUwMiA1LjYwMDEgNS42MDE1NiA1LjMxMzU2IDUuNjAxNTYgNC45NjAxVjIuMjQwMUM1LjYwMTU2IDEuODg2NjQgNS4zMTUwMiAxLjYwMDEgNC45NjE1NiAxLjYwMDFaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00Ljk2MTU2IDEwLjM5OTlIMi4yNDE1NkMxLjg4ODEgMTAuMzk5OSAxLjYwMTU2IDEwLjY4NjQgMS42MDE1NiAxMS4wMzk5VjEzLjc1OTlDMS42MDE1NiAxNC4xMTM0IDEuODg4MSAxNC4zOTk5IDIuMjQxNTYgMTQuMzk5OUg0Ljk2MTU2QzUuMzE1MDIgMTQuMzk5OSA1LjYwMTU2IDE0LjExMzQgNS42MDE1NiAxMy43NTk5VjExLjAzOTlDNS42MDE1NiAxMC42ODY0IDUuMzE1MDIgMTAuMzk5OSA0Ljk2MTU2IDEwLjM5OTlaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik0xMy43NTg0IDEuNjAwMUgxMS4wMzg0QzEwLjY4NSAxLjYwMDEgMTAuMzk4NCAxLjg4NjY0IDEwLjM5ODQgMi4yNDAxVjQuOTYwMUMxMC4zOTg0IDUuMzEzNTYgMTAuNjg1IDUuNjAwMSAxMS4wMzg0IDUuNjAwMUgxMy43NTg0QzE0LjExMTkgNS42MDAxIDE0LjM5ODQgNS4zMTM1NiAxNC4zOTg0IDQuOTYwMVYyLjI0MDFDMTQuMzk4NCAxLjg4NjY0IDE0LjExMTkgMS42MDAxIDEzLjc1ODQgMS42MDAxWiIgZmlsbD0iI2ZmZiIvPgo8cGF0aCBkPSJNNCAxMkwxMiA0TDQgMTJaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00IDEyTDEyIDQiIHN0cm9rZT0iI2ZmZiIgc3Ryb2tlLXdpZHRoPSIxLjUiIHN0cm9rZS1saW5lY2FwPSJyb3VuZCIvPgo8L3N2Zz4K&logoColor=ffffff" alt="zread"/></a>
+<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?color=76bad9"/></a>
+<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%E4%B8%AA&label=%E6%8F%92%E4%BB%B6%E5%B8%82%E5%9C%BA&cacheSeconds=3600">
+<img src="https://gitcode.com/Soulter/AstrBot/star/badge.svg" href="https://gitcode.com/Soulter/AstrBot">
 </div>

 <br>

-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ru.md">Русский</a>
-
 <a href="https://astrbot.app/">Documentation</a> ｜
 <a href="https://blog.astrbot.app/">Blog</a> ｜
 <a href="https://astrbot.featurebase.app/roadmap">Feuille de route</a> ｜
@@ -43,12 +42,31 @@ AstrBot est une plateforme de chatbot Agent tout-en-un open source qui s'intègr
 ## Fonctionnalités principales

 1. 💯 Gratuit & Open Source.
-2. ✨ Conversations avec LLM IA, Multimodal, Agent, MCP, Base de connaissances, Paramètres de personnalité.
-3. 🤖 Prise en charge de l'intégration avec Dify, Alibaba Cloud Bailian, Coze et autres plateformes d'agents.
-4. 🌐 Multi-plateforme : QQ, WeChat Work, Feishu, DingTalk, Comptes officiels WeChat, Telegram, Slack, et [plus encore](#plateformes-de-messagerie-prises-en-charge).
-5. 📦 Extensions de plugins avec près de 800 plugins disponibles pour une installation en un clic.
-6. 💻 Support WebUI.
-7. 🌐 Support de l'internationalisation (i18n).
+2. ✨ Dialogue avec de grands modèles d'IA, multimodal, Agent, MCP, Skills, Base de connaissances, Paramétrage de personnalité, compression automatique des dialogues.
+3. 🤖 Prise en charge de l'accès aux plateformes d'Agents telles que Dify, Alibaba Cloud Bailian, Coze, etc.
+4. 🌐 Multiplateforme : supporte QQ, WeChat Enterprise, Feishu, DingTalk, Comptes officiels WeChat, Telegram, Slack et [plus encore](#plateformes-de-messagerie-prises-en-charge).
+5. 📦 Extension par plugins, avec près de 800 plugins déjà disponibles pour une installation en un clic.
+6. 🛡️  Environnement isolé [Agent Sandbox](https://docs.astrbot.app/use/astrbot-agent-sandbox.html) : exécution sécurisée de code, appels Shell et réutilisation des ressources au niveau de la session.
+7. 💻 Support WebUI.
+8. 🌈 Support Web ChatUI, avec sandbox d'agent intégrée, recherche web, etc.
+9. 🌐 Support de l'internationalisation (i18n).
+
+<br>
+
+<table align="center">
+  <tr align="center">
+    <th>💙 Jeux de rôle & Accompagnement émotionnel</th>
+    <th>✨ Agent proactif</th>
+    <th>🚀 Capacités agentiques générales</th>
+    <th>🧩 900+ Plugins de communauté</th>
+  </tr>
+  <tr>
+    <td align="center"><p align="center"><img width="984" height="1746" alt="99b587c5d35eea09d84f33e6cf6cfd4f" src="https://github.com/user-attachments/assets/89196061-3290-458d-b51f-afa178049f84" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1612" alt="c449acd838c41d0915cc08a3824025b1" src="https://github.com/user-attachments/assets/f75368b4-e022-41dc-a9e0-131c3e73e32e" /></p></td>
+    <td align="center"><p align="center"><img width="974" height="1732" alt="image" src="https://github.com/user-attachments/assets/e22a3968-87d7-4708-a7cd-e7f198c7c32e" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1734" alt="image" src="https://github.com/user-attachments/assets/0952b395-6b4a-432a-8a50-c294b7f89750" /></p></td>
+  </tr>
+</table>

 ## Démarrage rapide

@@ -61,7 +79,18 @@ Veuillez consulter la documentation officielle : [Déployer AstrBot avec Docker]
 #### Déploiement uv

 ```bash
-uvx astrbot
+uv tool install astrbot
+astrbot
+```
+
+#### Installation via le gestionnaire de paquets du système
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# ou utiliser paru
+paru -S astrbot-git
 ```

 #### Déploiement BT-Panel
@@ -115,6 +144,16 @@ uv run main.py

 Ou consultez la documentation officielle : [Déployer AstrBot depuis les sources](https://astrbot.app/deploy/astrbot/cli.html).

+#### Установка через системный пакетный менеджер
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# или используйте paru
+paru -S astrbot-git
+```
+
 ## Plateformes de messagerie prises en charge

 **Maintenues officiellement**
@@ -153,7 +192,7 @@ Ou consultez la documentation officielle : [Déployer AstrBot depuis les sources
 - [CompShare](https://www.compshare.cn/?ytag=GPU_YY-gh_astrbot&referral_code=FV7DcGowN4hB5UuXKgpE74)
 - [302.AI](https://share.302.ai/rr1M3l)
 - [TokenPony](https://www.tokenpony.cn/3YPyf)
- [SiliconFlow](https://docs.siliconflow.cn/cn/usecases/use-siliconcloud-in-astrbot)
+- [SiliconFlow](https://docs.siliconflow.cn/cn/usercases/use-siliconcloud-in-astrbot)
 - [PPIO Cloud](https://ppio.com/user/register?invited_by=AIOONE)
 - ModelScope
 - OneAPI
@@ -241,7 +280,12 @@ De plus, la naissance de ce projet n'aurait pas été possible sans l'aide des p

 </div>

-</details>
+<div align="center">
+
+_La compagnie et la capacité ne devraient jamais être des opposés. Nous souhaitons créer un robot capable à la fois de comprendre les émotions, d'offrir de la présence, et d'accomplir des tâches de manière fiable._

 _私は、高性能ですから!_

+<img src="https://files.astrbot.app/watashiwa-koseino-desukara.gif" width="100"/>
+
+</div>
@@ -1,9 +1,13 @@
 ![AstrBot-Logo-Simplified](https://github.com/user-attachments/assets/ffd99b6b-3272-4682-beaa-6fe74250f7d9)

-</p>
-
 <div align="center">

+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ru.md">Русский</a>
+
 <br>

 <div>
@@ -14,22 +18,17 @@
 <br>

 <div>
-<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?style=for-the-badge&color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
-<img src="https://img.shields.io/badge/python-3.10+-blue.svg?style=for-the-badge&color=76bad9" alt="python">
-<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?style=for-the-badge&color=76bad9"/></a>
-<a href="https://qm.qq.com/cgi-bin/qm/qr?k=wtbaNx7EioxeaqS9z7RQWVXPIxg2zYr7&jump_from=webapi&authKey=vlqnv/AV2DbJEvGIcxdlNSpfxVy+8vVqijgreRdnVKOaydpc+YSw4MctmEbr0k5"><img alt="QQ_community" src="https://img.shields.io/badge/QQ群-775869627-purple?style=for-the-badge&color=76bad9"></a>
-<a href="https://t.me/+hAsD2Ebl5as3NmY1"><img alt="Telegram_community" src="https://img.shields.io/badge/Telegram-AstrBot-purple?style=for-the-badge&color=76bad9"></a>
-<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%E5%80%8B&style=for-the-badge&label=%E3%83%97%E3%83%A9%E3%82%B0%E3%82%A4%E3%83%B3&cacheSeconds=3600">
+<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
+<img src="https://img.shields.io/badge/python-3.10+-blue.svg" alt="python">
+<img src="https://deepwiki.com/badge.svg" href="https://deepwiki.com/AstrBotDevs/AstrBot">
+<a href="https://zread.ai/AstrBotDevs/AstrBot" target="_blank"><img src="https://img.shields.io/badge/Ask_Zread-_.svg?style=flat&color=00b0aa&labelColor=000000&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB3aWR0aD0iMTYiIGhlaWdodD0iMTYiIHZpZXdCb3g9IjAgMCAxNiAxNiIgZmlsbD0ibm9uZSIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KPHBhdGggZD0iTTQuOTYxNTYgMS42MDAxSDIuMjQxNTZDMS44ODgxIDEuNjAwMSAxLjYwMTU2IDEuODg2NjQgMS42MDE1NiAyLjI0MDFWNC45NjAxQzEuNjAxNTYgNS4zMTM1NiAxLjg4ODEgNS42MDAxIDIuMjQxNTYgNS42MDAxSDQuOTYxNTZDNS4zMTUwMiA1LjYwMDEgNS42MDE1NiA1LjMxMzU2IDUuNjAxNTYgNC45NjAxVjIuMjQwMUM1LjYwMTU2IDEuODg2NjQgNS4zMTUwMiAxLjYwMDEgNC45NjE1NiAxLjYwMDFaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00Ljk2MTU2IDEwLjM5OTlIMi4yNDE1NkMxLjg4ODEgMTAuMzk5OSAxLjYwMTU2IDEwLjY4NjQgMS42MDE1NiAxMS4wMzk5VjEzLjc1OTlDMS42MDE1NiAxNC4xMTM0IDEuODg4MSAxNC4zOTk5IDIuMjQxNTYgMTQuMzk5OUg0Ljk2MTU2QzUuMzE1MDIgMTQuMzk5OSA1LjYwMTU2IDE0LjExMzQgNS42MDE1NiAxMy43NTk5VjExLjAzOTlDNS42MDE1NiAxMC42ODY0IDUuMzE1MDIgMTAuMzk5OSA0Ljk2MTU2IDEwLjM5OTlaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik0xMy43NTg0IDEuNjAwMUgxMS4wMzg0QzEwLjY4NSAxLjYwMDEgMTAuMzk4NCAxLjg4NjY0IDEwLjM5ODQgMi4yNDAxVjQuOTYwMUMxMC4zOTg0IDUuMzEzNTYgMTAuNjg1IDUuNjAwMSAxMS4wMzg0IDUuNjAwMUgxMy43NTg0QzE0LjExMTkgNS42MDAxIDE0LjM5ODQgNS4zMTM1NiAxNC4zOTg0IDQuOTYwMVYyLjI0MDFDMTQuMzk4NCAxLjg4NjY0IDE0LjExMTkgMS42MDAxIDEzLjc1ODQgMS42MDAxWiIgZmlsbD0iI2ZmZiIvPgo8cGF0aCBkPSJNNCAxMkwxMiA0TDQgMTJaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00IDEyTDEyIDQiIHN0cm9rZT0iI2ZmZiIgc3Ryb2tlLXdpZHRoPSIxLjUiIHN0cm9rZS1saW5lY2FwPSJyb3VuZCIvPgo8L3N2Zz4K&logoColor=ffffff" alt="zread"/></a>
+<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?color=76bad9"/></a>
+<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%E4%B8%AA&label=%E6%8F%92%E4%BB%B6%E5%B8%82%E5%9C%BA&cacheSeconds=3600">
+<img src="https://gitcode.com/Soulter/AstrBot/star/badge.svg" href="https://gitcode.com/Soulter/AstrBot">
 </div>

 <br>

-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ru.md">Русский</a>
-
 <a href="https://astrbot.app/">ドキュメント</a> ｜
 <a href="https://blog.astrbot.app/">Blog</a> ｜
 <a href="https://astrbot.featurebase.app/roadmap">ロードマップ</a> ｜
@@ -43,12 +42,31 @@ AstrBot は、主要なインスタントメッセージングアプリと統合
 ## 主な機能

 1. 💯 無料 & オープンソース。
-2. ✨ AI 大規模言語モデル対話、マルチモーダル、Agent、MCP、ナレッジベース、ペルソナ設定。
-3. 🤖 Dify、Alibaba Cloud 百炼、Coze などの Agent プラットフォームとの統合をサポート。
-4. 🌐 マルチプラットフォーム：QQ、WeChat Work、Feishu、DingTalk、WeChat 公式アカウント、Telegram、Slack、[その他](#サポートされているメッセージプラットフォーム)。
-5. 📦 約800個のプラグインをワンクリックでインストール可能なプラグイン拡張機能。
-6. 💻 WebUI サポート。
-7. 🌐 国際化（i18n）サポート。
+2. ✨ AI大規模言語モデル対話、マルチモーダル、Agent、MCP、Skills、ナレッジベース、ペルソナ設定、対話の自動圧縮。
+3. 🤖 Dify、Alibaba Cloud Bailian（百煉）、Coze などのAgentプラットフォームへの接続をサポート。
+4. 🌐 マルチプラットフォーム：QQ、企業微信（WeCom）、飛書（Lark）、釘釘（DingTalk）、WeChat公式アカウント、Telegram、Slack、[その他](#サポートされているメッセージプラットフォーム)に対応。
+5. 📦 プラグイン拡張：800近い既存プラグインをワンクリックでインストール可能。
+6. 🛡️  隔離環境[Agent Sandbox](https://docs.astrbot.app/use/astrbot-agent-sandbox.html)：コードの安全な実行、Shell呼び出し、セッションレベルのリソース再利用。
+7. 💻 WebUI 対応。
+8. 🌈 Web ChatUI 対応：ChatUI内にAgent Sandboxやウェブ検索などを内蔵。
+9. 🌐 多言語対応（i18n）。
+
+<br>
+
+<table align="center">
+  <tr align="center">
+    <th>💙 ロールプレイ & 感情的な対話</th>
+    <th>✨ プロアクティブ・エージェント (Proactive Agent)</th>
+    <th>🚀 汎用 エージェント的能力</th>
+    <th>🧩 900+ コミュニティプラグイン</th>
+  </tr>
+  <tr>
+    <td align="center"><p align="center"><img width="984" height="1746" alt="99b587c5d35eea09d84f33e6cf6cfd4f" src="https://github.com/user-attachments/assets/89196061-3290-458d-b51f-afa178049f84" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1612" alt="c449acd838c41d0915cc08a3824025b1" src="https://github.com/user-attachments/assets/f75368b4-e022-41dc-a9e0-131c3e73e32e" /></p></td>
+    <td align="center"><p align="center"><img width="974" height="1732" alt="image" src="https://github.com/user-attachments/assets/e22a3968-87d7-4708-a7cd-e7f198c7c32e" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1734" alt="image" src="https://github.com/user-attachments/assets/0952b395-6b4a-432a-8a50-c294b7f89750" /></p></td>
+  </tr>
+</table>

 ## クイックスタート

@@ -61,7 +79,18 @@ Docker / Docker Compose を使用した AstrBot のデプロイを推奨しま
 #### uv デプロイ

 ```bash
-uvx astrbot
+uv tool install astrbot
+astrbot
+```
+
+#### システムパッケージマネージャーでのインストール
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# または paru を使用
+paru -S astrbot-git
 ```

 #### 宝塔パネルデプロイ
@@ -115,6 +144,16 @@ uv run main.py

 または、公式ドキュメント [ソースコードから AstrBot をデプロイ](https://astrbot.app/deploy/astrbot/cli.html) をご参照ください。

+#### Установка через системный пакетный менеджер
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# или используйте paru
+paru -S astrbot-git
+```
+
 ## サポートされているメッセージプラットフォーム

 **公式メンテナンス**
@@ -242,6 +281,12 @@ AstrBot への貢献をしていただいたすべてのコントリビュータ

 </div>

-</details>
+<div align="center">
+
+_共感力と能力は決して対立するものではありません。私たちが目指すのは、感情を理解し、心の支えとなるだけでなく、確実に仕事をこなせるロボットの創造です。_

 _私は、高性能ですから!_
+
+<img src="https://files.astrbot.app/watashiwa-koseino-desukara.gif" width="100"/>
+
+</div>
@@ -1,9 +1,13 @@
 ![AstrBot-Logo-Simplified](https://github.com/user-attachments/assets/ffd99b6b-3272-4682-beaa-6fe74250f7d9)

-</p>
-
 <div align="center">

+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a>
+
 <br>

 <div>
@@ -14,22 +18,17 @@
 <br>

 <div>
-<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?style=for-the-badge&color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
-<img src="https://img.shields.io/badge/python-3.10+-blue.svg?style=for-the-badge&color=76bad9" alt="python">
-<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?style=for-the-badge&color=76bad9"/></a>
-<a href="https://qm.qq.com/cgi-bin/qm/qr?k=wtbaNx7EioxeaqS9z7RQWVXPIxg2zYr7&jump_from=webapi&authKey=vlqnv/AV2DbJEvGIcxdlNSpfxVy+8vVqijgreRdnVKOaydpc+YSw4MctmEbr0k5"><img alt="QQ_community" src="https://img.shields.io/badge/QQ群-775869627-purple?style=for-the-badge&color=76bad9"></a>
-<a href="https://t.me/+hAsD2Ebl5as3NmY1"><img alt="Telegram_community" src="https://img.shields.io/badge/Telegram-AstrBot-purple?style=for-the-badge&color=76bad9"></a>
-<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%20%D0%BF%D0%BB%D0%B0%D0%B3%D0%B8%D0%BD%D0%BE%D0%B2&style=for-the-badge&label=%D0%9C%D0%B0%D0%B3%D0%B0%D0%B7%D0%B8%D0%BD&cacheSeconds=3600">
+<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
+<img src="https://img.shields.io/badge/python-3.10+-blue.svg" alt="python">
+<img src="https://deepwiki.com/badge.svg" href="https://deepwiki.com/AstrBotDevs/AstrBot">
+<a href="https://zread.ai/AstrBotDevs/AstrBot" target="_blank"><img src="https://img.shields.io/badge/Ask_Zread-_.svg?style=flat&color=00b0aa&labelColor=000000&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB3aWR0aD0iMTYiIGhlaWdodD0iMTYiIHZpZXdCb3g9IjAgMCAxNiAxNiIgZmlsbD0ibm9uZSIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KPHBhdGggZD0iTTQuOTYxNTYgMS42MDAxSDIuMjQxNTZDMS44ODgxIDEuNjAwMSAxLjYwMTU2IDEuODg2NjQgMS42MDE1NiAyLjI0MDFWNC45NjAxQzEuNjAxNTYgNS4zMTM1NiAxLjg4ODEgNS42MDAxIDIuMjQxNTYgNS42MDAxSDQuOTYxNTZDNS4zMTUwMiA1LjYwMDEgNS42MDE1NiA1LjMxMzU2IDUuNjAxNTYgNC45NjAxVjIuMjQwMUM1LjYwMTU2IDEuODg2NjQgNS4zMTUwMiAxLjYwMDEgNC45NjE1NiAxLjYwMDFaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00Ljk2MTU2IDEwLjM5OTlIMi4yNDE1NkMxLjg4ODEgMTAuMzk5OSAxLjYwMTU2IDEwLjY4NjQgMS42MDE1NiAxMS4wMzk5VjEzLjc1OTlDMS42MDE1NiAxNC4xMTM0IDEuODg4MSAxNC4zOTk5IDIuMjQxNTYgMTQuMzk5OUg0Ljk2MTU2QzUuMzE1MDIgMTQuMzk5OSA1LjYwMTU2IDE0LjExMzQgNS42MDE1NiAxMy43NTk5VjExLjAzOTlDNS42MDE1NiAxMC42ODY0IDUuMzE1MDIgMTAuMzk5OSA0Ljk2MTU2IDEwLjM5OTlaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik0xMy43NTg0IDEuNjAwMUgxMS4wMzg0QzEwLjY4NSAxLjYwMDEgMTAuMzk4NCAxLjg4NjY0IDEwLjM5ODQgMi4yNDAxVjQuOTYwMUMxMC4zOTg0IDUuMzEzNTYgMTAuNjg1IDUuNjAwMSAxMS4wMzg0IDUuNjAwMUgxMy43NTg0QzE0LjExMTkgNS42MDAxIDE0LjM5ODQgNS4zMTM1NiAxNC4zOTg0IDQuOTYwMVYyLjI0MDFDMTQuMzk4NCAxLjg4NjY0IDE0LjExMTkgMS42MDAxIDEzLjc1ODQgMS42MDAxWiIgZmlsbD0iI2ZmZiIvPgo8cGF0aCBkPSJNNCAxMkwxMiA0TDQgMTJaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00IDEyTDEyIDQiIHN0cm9rZT0iI2ZmZiIgc3Ryb2tlLXdpZHRoPSIxLjUiIHN0cm9rZS1saW5lY2FwPSJyb3VuZCIvPgo8L3N2Zz4K&logoColor=ffffff" alt="zread"/></a>
+<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?color=76bad9"/></a>
+<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%E4%B8%AA&label=%E6%8F%92%E4%BB%B6%E5%B8%82%E5%9C%BA&cacheSeconds=3600">
+<img src="https://gitcode.com/Soulter/AstrBot/star/badge.svg" href="https://gitcode.com/Soulter/AstrBot">
 </div>

 <br>

-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_zh-TW.md">繁體中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a>
-
 <a href="https://astrbot.app/">Документация</a> ｜
 <a href="https://blog.astrbot.app/">Блог</a> ｜
 <a href="https://astrbot.featurebase.app/roadmap">Дорожная карта</a> ｜
@@ -42,13 +41,32 @@ AstrBot — это универсальная платформа Agent-чатб

 ## Основные возможности

-1. 💯 Бесплатно и с открытым исходным кодом.
-2. ✨ ИИ-диалоги с LLM, мультимодальность, Agent, MCP, база знаний, настройки личности.
-3. 🤖 Поддержка интеграции с Dify, Alibaba Cloud Bailian, Coze и другими платформами агентов.
-4. 🌐 Мультиплатформенность: QQ, WeChat Work, Feishu, DingTalk, официальные аккаунты WeChat, Telegram, Slack и [другие](#поддерживаемые-платформы-обмена-сообщениями).
-5. 📦 Расширения плагинов с почти 800 плагинами, доступными для установки в один клик.
-6. 💻 Поддержка WebUI.
-7. 🌐 Поддержка интернационализации (i18n).
+1. 💯 Бесплатно & Открытый исходный код.
+2. ✨ Диалоги с ИИ-моделями, мультимодальность, Agent, MCP, Skills, База знаний, Настройка личности, автоматическое сжатие диалогов.
+3. 🤖 Поддержка интеграции с платформами Agents, такими как Dify, Alibaba Cloud Bailian, Coze и др.
+4. 🌐 Мультиплатформенность: поддержка QQ, WeChat для предприятий, Feishu, DingTalk, публичных аккаунтов WeChat, Telegram, Slack и [других](#Поддерживаемые-платформы-обмена-сообщениями).
+5. 📦 Расширение плагинами: доступно почти 800 плагинов для установки в один клик.
+6. 🛡️  Изолированная среда[Agent Sandbox](https://docs.astrbot.app/use/astrbot-agent-sandbox.html): безопасное выполнение любого кода, вызов Shell, повторное использование ресурсов на уровне сессии.
+7. 💻 Поддержка WebUI.
+8. 🌈 Поддержка Web ChatUI: встроенная песочница агента, веб-поиск и др.
+9. 🌐 Поддержка интернационализации (i18n).
+
+<br>
+
+<table align="center">
+  <tr align="center">
+    <th>💙 Ролевые игры & Эмоциональная поддержка</th>
+    <th>✨ Проактивный Агент(Agent)</th>
+    <th>🚀 Универсальные Агентные возможности</th>
+    <th>🧩 Универсальные Агентные (Agentic) возможности</th>
+  </tr>
+  <tr>
+    <td align="center"><p align="center"><img width="984" height="1746" alt="99b587c5d35eea09d84f33e6cf6cfd4f" src="https://github.com/user-attachments/assets/89196061-3290-458d-b51f-afa178049f84" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1612" alt="c449acd838c41d0915cc08a3824025b1" src="https://github.com/user-attachments/assets/f75368b4-e022-41dc-a9e0-131c3e73e32e" /></p></td>
+    <td align="center"><p align="center"><img width="974" height="1732" alt="image" src="https://github.com/user-attachments/assets/e22a3968-87d7-4708-a7cd-e7f198c7c32e" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1734" alt="image" src="https://github.com/user-attachments/assets/0952b395-6b4a-432a-8a50-c294b7f89750" /></p></td>
+  </tr>
+</table>

 ## Быстрый старт

@@ -61,7 +79,8 @@ AstrBot — это универсальная платформа Agent-чатб
 #### Развёртывание uv

 ```bash
-uvx astrbot
+uv tool install astrbot
+astrbot
 ```

 #### Развёртывание BT-Panel
@@ -115,6 +134,16 @@ uv run main.py

 Или см. официальную документацию: [Развёртывание AstrBot из исходного кода](https://astrbot.app/deploy/astrbot/cli.html).

+#### Установка через системный пакетный менеджер
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# или используйте paru
+paru -S astrbot-git
+```
+
 ## Поддерживаемые платформы обмена сообщениями

 **Официально поддерживаемые**
@@ -153,7 +182,7 @@ uv run main.py
 - [CompShare](https://www.compshare.cn/?ytag=GPU_YY-gh_astrbot&referral_code=FV7DcGowN4hB5UuXKgpE74)
 - [302.AI](https://share.302.ai/rr1M3l)
 - [TokenPony](https://www.tokenpony.cn/3YPyf)
- [SiliconFlow](https://docs.siliconflow.cn/cn/usecases/use-siliconcloud-in-astrbot)
+- [SiliconFlow](https://docs.siliconflow.cn/cn/usercases/use-siliconcloud-in-astrbot)
 - [PPIO Cloud](https://ppio.com/user/register?invited_by=AIOONE)
 - ModelScope
 - OneAPI
@@ -235,13 +264,19 @@ pre-commit install
 > [!TIP]
 > Если этот проект помог вам в жизни или работе, или если вас интересует его будущее развитие, пожалуйста, поставьте проекту звезду. Это движущая сила поддержки этого проекта с открытым исходным кодом <3

+
 <div align="center">

 [![Star History Chart](https://api.star-history.com/svg?repos=astrbotdevs/astrbot&type=Date)](https://star-history.com/#astrbotdevs/astrbot&Date)

 </div>

-</details>
+<div align="center">
+
+_Сопровождение и способности никогда не должны быть противоположностями. Мы стремимся создать робота, который сможет как понимать эмоции, оказывать душевную поддержку, так и надёжно выполнять работу._

 _私は、高性能ですから!_

+<img src="https://files.astrbot.app/watashiwa-koseino-desukara.gif" width="100"/>
+
+</div>
@@ -1,9 +1,13 @@
 ![AstrBot-Logo-Simplified](https://github.com/user-attachments/assets/ffd99b6b-3272-4682-beaa-6fe74250f7d9)

-</p>
-
 <div align="center">

+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">简体中文</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a> ｜
+<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ru.md">Русский</a>
+
 <br>

 <div>
@@ -14,22 +18,17 @@
 <br>

 <div>
-<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?style=for-the-badge&color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
-<img src="https://img.shields.io/badge/python-3.10+-blue.svg?style=for-the-badge&color=76bad9" alt="python">
-<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?style=for-the-badge&color=76bad9"/></a>
-<a href="https://qm.qq.com/cgi-bin/qm/qr?k=wtbaNx7EioxeaqS9z7RQWVXPIxg2zYr7&jump_from=webapi&authKey=vlqnv/AV2DbJEvGIcxdlNSpfxVy+8vVqijgreRdnVKOaydpc+YSw4MctmEbr0k5"><img alt="QQ_community" src="https://img.shields.io/badge/QQ群-775869627-purple?style=for-the-badge&color=76bad9"></a>
-<a href="https://t.me/+hAsD2Ebl5as3NmY1"><img alt="Telegram_community" src="https://img.shields.io/badge/Telegram-AstrBot-purple?style=for-the-badge&color=76bad9"></a>
-<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%E5%80%8B&style=for-the-badge&label=%E6%8F%92%E4%BB%B6%E5%B8%82%E5%A0%B4&cacheSeconds=3600">
+<img src="https://img.shields.io/github/v/release/AstrBotDevs/AstrBot?color=76bad9" href="https://github.com/AstrBotDevs/AstrBot/releases/latest">
+<img src="https://img.shields.io/badge/python-3.10+-blue.svg" alt="python">
+<img src="https://deepwiki.com/badge.svg" href="https://deepwiki.com/AstrBotDevs/AstrBot">
+<a href="https://zread.ai/AstrBotDevs/AstrBot" target="_blank"><img src="https://img.shields.io/badge/Ask_Zread-_.svg?style=flat&color=00b0aa&labelColor=000000&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB3aWR0aD0iMTYiIGhlaWdodD0iMTYiIHZpZXdCb3g9IjAgMCAxNiAxNiIgZmlsbD0ibm9uZSIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KPHBhdGggZD0iTTQuOTYxNTYgMS42MDAxSDIuMjQxNTZDMS44ODgxIDEuNjAwMSAxLjYwMTU2IDEuODg2NjQgMS42MDE1NiAyLjI0MDFWNC45NjAxQzEuNjAxNTYgNS4zMTM1NiAxLjg4ODEgNS42MDAxIDIuMjQxNTYgNS42MDAxSDQuOTYxNTZDNS4zMTUwMiA1LjYwMDEgNS42MDE1NiA1LjMxMzU2IDUuNjAxNTYgNC45NjAxVjIuMjQwMUM1LjYwMTU2IDEuODg2NjQgNS4zMTUwMiAxLjYwMDEgNC45NjE1NiAxLjYwMDFaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00Ljk2MTU2IDEwLjM5OTlIMi4yNDE1NkMxLjg4ODEgMTAuMzk5OSAxLjYwMTU2IDEwLjY4NjQgMS42MDE1NiAxMS4wMzk5VjEzLjc1OTlDMS42MDE1NiAxNC4xMTM0IDEuODg4MSAxNC4zOTk5IDIuMjQxNTYgMTQuMzk5OUg0Ljk2MTU2QzUuMzE1MDIgMTQuMzk5OSA1LjYwMTU2IDE0LjExMzQgNS42MDE1NiAxMy43NTk5VjExLjAzOTlDNS42MDE1NiAxMC42ODY0IDUuMzE1MDIgMTAuMzk5OSA0Ljk2MTU2IDEwLjM5OTlaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik0xMy43NTg0IDEuNjAwMUgxMS4wMzg0QzEwLjY4NSAxLjYwMDEgMTAuMzk4NCAxLjg4NjY0IDEwLjM5ODQgMi4yNDAxVjQuOTYwMUMxMC4zOTg0IDUuMzEzNTYgMTAuNjg1IDUuNjAwMSAxMS4wMzg0IDUuNjAwMUgxMy43NTg0QzE0LjExMTkgNS42MDAxIDE0LjM5ODQgNS4zMTM1NiAxNC4zOTg0IDQuOTYwMVYyLjI0MDFDMTQuMzk4NCAxLjg4NjY0IDE0LjExMTkgMS42MDAxIDEzLjc1ODQgMS42MDAxWiIgZmlsbD0iI2ZmZiIvPgo8cGF0aCBkPSJNNCAxMkwxMiA0TDQgMTJaIiBmaWxsPSIjZmZmIi8%2BCjxwYXRoIGQ9Ik00IDEyTDEyIDQiIHN0cm9rZT0iI2ZmZiIgc3Ryb2tlLXdpZHRoPSIxLjUiIHN0cm9rZS1saW5lY2FwPSJyb3VuZCIvPgo8L3N2Zz4K&logoColor=ffffff" alt="zread"/></a>
+<a href="https://hub.docker.com/r/soulter/astrbot"><img alt="Docker pull" src="https://img.shields.io/docker/pulls/soulter/astrbot.svg?color=76bad9"/></a>
+<img src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fapi.soulter.top%2Fastrbot%2Fplugin-num&query=%24.result&suffix=%E5%80%8B&label=%E6%8F%92%E4%BB%B6%E5%B8%82%E5%A0%B4&cacheSeconds=3600">
+<img src="https://gitcode.com/Soulter/AstrBot/star/badge.svg" href="https://gitcode.com/Soulter/AstrBot">
 </div>

 <br>

-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README.md">简体中文</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_en.md">English</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ja.md">日本語</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_fr.md">Français</a> ｜
-<a href="https://github.com/AstrBotDevs/AstrBot/blob/master/README_ru.md">Русский</a>
-
 <a href="https://astrbot.app/">文件</a> ｜
 <a href="https://blog.astrbot.app/">Blog</a> ｜
 <a href="https://astrbot.featurebase.app/roadmap">路線圖</a> ｜
@@ -43,12 +42,31 @@ AstrBot 是一個開源的一站式 Agent 聊天機器人平台，可接入主
 ## 主要功能

 1. 💯 免費 & 開源。
-2. ✨ AI 大型模型對話，多模態，Agent，MCP，知識庫，人格設定。
-3. 🤖 支援接入 Dify、阿里雲百煉、Coze 等智慧體平台。
-4. 🌐 多平台：QQ、企業微信、飛書、釘釘、微信公眾號、Telegram、Slack 以及[更多](#支援的訊息平台)。
-5. 📦 外掛擴充，已有近 800 個外掛可一鍵安裝。
-6. 💻 WebUI 支援。
-7. 🌐 國際化（i18n）支援。
+2. ✨ AI 大模型對話，多模態，Agent，MCP，Skills，知識庫，人格設定，自動壓縮對話。
+3. 🤖 支援接入 Dify、阿里雲百煉、Coze 等智慧體 (Agent) 平台。
+4. 🌐 多平台，支援 QQ、企業微信、飛書、釘釘、微信公眾號、Telegram、Slack 以及[更多](#支援的訊息平台)。
+5. 📦 插件擴展，已有近 800 個插件可一鍵安裝。
+6. 🛡️ [Agent Sandbox](https://docs.astrbot.app/use/astrbot-agent-sandbox.html) 隔離化環境，安全地執行任何代碼、調用 Shell、會話級資源複用。
+7. 💻 WebUI 支援。
+8. 🌈 Web ChatUI 支援，ChatUI 內置代理沙盒 (Agent Sandbox)、網頁搜尋等。
+9. 🌐 國際化（i18n）支援。
+
+<br>
+
+<table align="center">
+  <tr align="center">
+    <th>💙 角色扮演 & 情感陪伴</th>
+    <th>✨ 主動式 Agent</th>
+    <th>🚀 通用 Agentic 能力</th>
+    <th>🧩 900+ 社區外掛程式</th>
+  </tr>
+  <tr>
+    <td align="center"><p align="center"><img width="984" height="1746" alt="99b587c5d35eea09d84f33e6cf6cfd4f" src="https://github.com/user-attachments/assets/89196061-3290-458d-b51f-afa178049f84" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1612" alt="c449acd838c41d0915cc08a3824025b1" src="https://github.com/user-attachments/assets/f75368b4-e022-41dc-a9e0-131c3e73e32e" /></p></td>
+    <td align="center"><p align="center"><img width="974" height="1732" alt="image" src="https://github.com/user-attachments/assets/e22a3968-87d7-4708-a7cd-e7f198c7c32e" /></p></td>
+    <td align="center"><p align="center"><img width="976" height="1734" alt="image" src="https://github.com/user-attachments/assets/0952b395-6b4a-432a-8a50-c294b7f89750" /></p></td>
+  </tr>
+</table>

 ## 快速開始

@@ -61,7 +79,8 @@ AstrBot 是一個開源的一站式 Agent 聊天機器人平台，可接入主
 #### uv 部署

 ```bash
-uvx astrbot
+uv tool install astrbot
+astrbot
 ```

 #### 寶塔面板部署
@@ -115,6 +134,16 @@ uv run main.py

 或者請參閱官方文件 [透過原始碼部署 AstrBot](https://astrbot.app/deploy/astrbot/cli.html)。

+#### 系統套件管理員安裝
+
+##### Arch Linux
+
+```bash
+yay -S astrbot-git
+# 或者使用 paru
+paru -S astrbot-git
+```
+
 ## 支援的訊息平台

 **官方維護**
@@ -241,7 +270,12 @@ pre-commit install

 </div>

-</details>
+<div align="center">
+
+_陪伴與能力從來不應該是對立面。我們希望創造的是一個既能理解情緒、給予陪伴，也能可靠完成工作的機器人。_

 _私は、高性能ですから!_

+<img src="https://files.astrbot.app/watashiwa-koseino-desukara.gif" width="100"/>
+
+</div>
@@ -1 +1 @@
-__version__ = "4.15.0"
+__version__ = "4.17.2"
@@ -10,7 +10,7 @@ from astrbot.core.provider.entities import (
    LLMResponse,
    ProviderRequest,
 )
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_file

 from ...hooks import BaseAgentRunHooks
@@ -291,8 +291,8 @@ class DifyAgentRunner(BaseAgentRunner[TContext]):
                    return Comp.Image(file=item["url"], url=item["url"])
                case "audio":
                    # 仅支持 wav
-                    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-                    path = os.path.join(temp_dir, f"{item['filename']}.wav")
+                    temp_dir = get_astrbot_temp_path()
+                    path = os.path.join(temp_dir, f"dify_{item['filename']}.wav")
                    await download_file(item["url"], path)
                    return Comp.Image(file=item["url"], url=item["url"])
                case "video":
@@ -91,6 +91,7 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
        custom_token_counter: TokenCounter | None = None,
        custom_compressor: ContextCompressor | None = None,
        tool_schema_mode: str | None = "full",
+        fallback_providers: list[Provider] | None = None,
        **kwargs: T.Any,
    ) -> None:
        self.req = request
@@ -120,6 +121,17 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
        self.context_manager = ContextManager(self.context_config)

        self.provider = provider
+        self.fallback_providers: list[Provider] = []
+        seen_provider_ids: set[str] = {str(provider.provider_config.get("id", ""))}
+        for fallback_provider in fallback_providers or []:
+            fallback_id = str(fallback_provider.provider_config.get("id", ""))
+            if fallback_provider is provider:
+                continue
+            if fallback_id and fallback_id in seen_provider_ids:
+                continue
+            self.fallback_providers.append(fallback_provider)
+            if fallback_id:
+                seen_provider_ids.add(fallback_id)
        self.final_llm_resp = None
        self._state = AgentState.IDLE
        self.tool_executor = tool_executor
@@ -166,16 +178,19 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
        self.stats = AgentStats()
        self.stats.start_time = time.time()

-    async def _iter_llm_responses(self) -> T.AsyncGenerator[LLMResponse, None]:
+    async def _iter_llm_responses(
+        self, *, include_model: bool = True
+    ) -> T.AsyncGenerator[LLMResponse, None]:
        """Yields chunks *and* a final LLMResponse."""
        payload = {
            "contexts": self.run_context.messages,  # list[Message]
            "func_tool": self.req.func_tool,
-            "model": self.req.model,  # NOTE: in fact, this arg is None in most cases
            "session_id": self.req.session_id,
            "extra_user_content_parts": self.req.extra_user_content_parts,  # list[ContentPart]
        }
-
+        if include_model:
+            # For primary provider we keep explicit model selection if provided.
+            payload["model"] = self.req.model
        if self.streaming:
            stream = self.provider.text_chat_stream(**payload)
            async for resp in stream:  # type: ignore
@@ -183,6 +198,83 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
        else:
            yield await self.provider.text_chat(**payload)

+    async def _iter_llm_responses_with_fallback(
+        self,
+    ) -> T.AsyncGenerator[LLMResponse, None]:
+        """Wrap _iter_llm_responses with provider fallback handling."""
+        candidates = [self.provider, *self.fallback_providers]
+        total_candidates = len(candidates)
+        last_exception: Exception | None = None
+        last_err_response: LLMResponse | None = None
+
+        for idx, candidate in enumerate(candidates):
+            candidate_id = candidate.provider_config.get("id", "<unknown>")
+            is_last_candidate = idx == total_candidates - 1
+            if idx > 0:
+                logger.warning(
+                    "Switched from %s to fallback chat provider: %s",
+                    self.provider.provider_config.get("id", "<unknown>"),
+                    candidate_id,
+                )
+            self.provider = candidate
+            has_stream_output = False
+            try:
+                async for resp in self._iter_llm_responses(include_model=idx == 0):
+                    if resp.is_chunk:
+                        has_stream_output = True
+                        yield resp
+                        continue
+
+                    if (
+                        resp.role == "err"
+                        and not has_stream_output
+                        and (not is_last_candidate)
+                    ):
+                        last_err_response = resp
+                        logger.warning(
+                            "Chat Model %s returns error response, trying fallback to next provider.",
+                            candidate_id,
+                        )
+                        break
+
+                    yield resp
+                    return
+
+                if has_stream_output:
+                    return
+            except Exception as exc:  # noqa: BLE001
+                last_exception = exc
+                logger.warning(
+                    "Chat Model %s request error: %s",
+                    candidate_id,
+                    exc,
+                    exc_info=True,
+                )
+                continue
+
+        if last_err_response:
+            yield last_err_response
+            return
+        if last_exception:
+            yield LLMResponse(
+                role="err",
+                completion_text=(
+                    "All chat models failed: "
+                    f"{type(last_exception).__name__}: {last_exception}"
+                ),
+            )
+            return
+        yield LLMResponse(
+            role="err",
+            completion_text="All available chat models are unavailable.",
+        )
+
+    def _simple_print_message_role(self, tag: str = ""):
+        roles = []
+        for message in self.run_context.messages:
+            roles.append(message.role)
+        logger.debug(f"{tag} RunCtx.messages -> [{len(roles)}] {','.join(roles)}")
+
    @override
    async def step(self):
        """Process a single step of the agent.
@@ -203,11 +295,13 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):

        # do truncate and compress
        token_usage = self.req.conversation.token_usage if self.req.conversation else 0
+        self._simple_print_message_role("[BefCompact]")
        self.run_context.messages = await self.context_manager.process(
            self.run_context.messages, trusted_token_usage=token_usage
        )
+        self._simple_print_message_role("[AftCompact]")

-        async for llm_response in self._iter_llm_responses():
+        async for llm_response in self._iter_llm_responses_with_fallback():
            if llm_response.is_chunk:
                # update ttft
                if self.stats.time_to_first_token == 0:
@@ -42,6 +42,7 @@ from astrbot.core.message.components import File, Image, Reply
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.provider import Provider
 from astrbot.core.provider.entities import ProviderRequest
+from astrbot.core.provider.manager import llm_tools
 from astrbot.core.skills.skill_manager import SkillManager, build_skills_prompt
 from astrbot.core.star.context import Context
 from astrbot.core.star.star_handler import star_map
@@ -52,6 +53,17 @@ from astrbot.core.tools.cron_tools import (
 )
 from astrbot.core.utils.file_extract import extract_file_moonshotai
 from astrbot.core.utils.llm_metadata import LLM_METADATAS
+from astrbot.core.utils.quoted_message.settings import (
+    SETTINGS as DEFAULT_QUOTED_MESSAGE_SETTINGS,
+)
+from astrbot.core.utils.quoted_message.settings import (
+    QuotedMessageParserSettings,
+)
+from astrbot.core.utils.quoted_message_parser import (
+    extract_quoted_message_images,
+    extract_quoted_message_text,
+)
+from astrbot.core.utils.string_utils import normalize_and_dedupe_strings


@dataclass(slots=True)
@@ -108,6 +120,8 @@ class MainAgentBuildConfig:
    provider_settings: dict = field(default_factory=dict)
    subagent_orchestrator: dict = field(default_factory=dict)
    timezone: str | None = None
+    max_quoted_fallback_images: int = 20
+    """Maximum number of images injected from quoted-message fallback extraction."""


@dataclass(slots=True)
@@ -470,11 +484,29 @@ async def _ensure_img_caption(
        logger.error("处理图片描述失败: %s", exc)


+def _append_quoted_image_attachment(req: ProviderRequest, image_path: str) -> None:
+    req.extra_user_content_parts.append(
+        TextPart(text=f"[Image Attachment in quoted message: path {image_path}]")
+    )
+
+
+def _get_quoted_message_parser_settings(
+    provider_settings: dict[str, object] | None,
+) -> QuotedMessageParserSettings:
+    if not isinstance(provider_settings, dict):
+        return DEFAULT_QUOTED_MESSAGE_SETTINGS
+    overrides = provider_settings.get("quoted_message_parser")
+    if not isinstance(overrides, dict):
+        return DEFAULT_QUOTED_MESSAGE_SETTINGS
+    return DEFAULT_QUOTED_MESSAGE_SETTINGS.with_overrides(overrides)
+
+
 async def _process_quote_message(
    event: AstrMessageEvent,
    req: ProviderRequest,
    img_cap_prov_id: str,
    plugin_context: Context,
+    quoted_message_settings: QuotedMessageParserSettings = DEFAULT_QUOTED_MESSAGE_SETTINGS,
 ) -> None:
    quote = None
    for comp in event.message_obj.message:
@@ -486,7 +518,15 @@ async def _process_quote_message(

    content_parts = []
    sender_info = f"({quote.sender_nickname}): " if quote.sender_nickname else ""
-    message_str = quote.message_str or "[Empty Text]"
+    message_str = (
+        await extract_quoted_message_text(
+            event,
+            quote,
+            settings=quoted_message_settings,
+        )
+        or quote.message_str
+        or "[Empty Text]"
+    )
    content_parts.append(f"{sender_info}{message_str}")

    image_seg = None
@@ -592,11 +632,13 @@ async def _decorate_llm_request(
            )

    img_cap_prov_id = cfg.get("default_image_caption_provider_id") or ""
+    quoted_message_settings = _get_quoted_message_parser_settings(cfg)
    await _process_quote_message(
        event,
        req,
        img_cap_prov_id,
        plugin_context,
+        quoted_message_settings,
    )

    tz = config.timezone
@@ -728,6 +770,14 @@ def _plugin_tool_fix(event: AstrMessageEvent, req: ProviderRequest) -> None:
            if plugin.name in event.plugins_name or plugin.reserved:
                new_tool_set.add_tool(tool)
        req.func_tool = new_tool_set
+    else:
+        # mcp tools
+        tool_set = req.func_tool
+        if not tool_set:
+            tool_set = ToolSet()
+        for tool in llm_tools.func_list:
+            if isinstance(tool, MCPTool):
+                tool_set.add_tool(tool)


 async def _handle_webchat(
@@ -829,6 +879,41 @@ def _get_compress_provider(
    return provider


+def _get_fallback_chat_providers(
+    provider: Provider, plugin_context: Context, provider_settings: dict
+) -> list[Provider]:
+    fallback_ids = provider_settings.get("fallback_chat_models", [])
+    if not isinstance(fallback_ids, list):
+        logger.warning(
+            "fallback_chat_models setting is not a list, skip fallback providers."
+        )
+        return []
+
+    provider_id = str(provider.provider_config.get("id", ""))
+    seen_provider_ids: set[str] = {provider_id} if provider_id else set()
+    fallbacks: list[Provider] = []
+
+    for fallback_id in fallback_ids:
+        if not isinstance(fallback_id, str) or not fallback_id:
+            continue
+        if fallback_id in seen_provider_ids:
+            continue
+        fallback_provider = plugin_context.get_provider_by_id(fallback_id)
+        if fallback_provider is None:
+            logger.warning("Fallback chat provider `%s` not found, skip.", fallback_id)
+            continue
+        if not isinstance(fallback_provider, Provider):
+            logger.warning(
+                "Fallback chat provider `%s` is invalid type: %s, skip.",
+                fallback_id,
+                type(fallback_provider),
+            )
+            continue
+        fallbacks.append(fallback_provider)
+        seen_provider_ids.add(fallback_id)
+    return fallbacks
+
+
 async def build_main_agent(
    *,
    event: AstrMessageEvent,
@@ -867,6 +952,8 @@ async def build_main_agent(
                return None

            req.prompt = event.message_str[len(config.provider_wake_prefix) :]
+
+            # media files attachments
            for comp in event.message_obj.message:
                if isinstance(comp, Image):
                    image_path = await comp.convert_to_file_path()
@@ -882,6 +969,81 @@ async def build_main_agent(
                            text=f"[File Attachment: name {file_name}, path {file_path}]"
                        )
                    )
+            # quoted message attachments
+            reply_comps = [
+                comp for comp in event.message_obj.message if isinstance(comp, Reply)
+            ]
+            quoted_message_settings = _get_quoted_message_parser_settings(
+                config.provider_settings
+            )
+            fallback_quoted_image_count = 0
+            for comp in reply_comps:
+                has_embedded_image = False
+                if comp.chain:
+                    for reply_comp in comp.chain:
+                        if isinstance(reply_comp, Image):
+                            has_embedded_image = True
+                            image_path = await reply_comp.convert_to_file_path()
+                            req.image_urls.append(image_path)
+                            _append_quoted_image_attachment(req, image_path)
+                        elif isinstance(reply_comp, File):
+                            file_path = await reply_comp.get_file()
+                            file_name = reply_comp.name or os.path.basename(file_path)
+                            req.extra_user_content_parts.append(
+                                TextPart(
+                                    text=(
+                                        f"[File Attachment in quoted message: "
+                                        f"name {file_name}, path {file_path}]"
+                                    )
+                                )
+                            )
+
+                # Fallback quoted image extraction for reply-id-only payloads, or when
+                # embedded reply chain only contains placeholders (e.g. [Forward Message], [Image]).
+                if not has_embedded_image:
+                    try:
+                        fallback_images = normalize_and_dedupe_strings(
+                            await extract_quoted_message_images(
+                                event,
+                                comp,
+                                settings=quoted_message_settings,
+                            )
+                        )
+                        remaining_limit = max(
+                            config.max_quoted_fallback_images
+                            - fallback_quoted_image_count,
+                            0,
+                        )
+                        if remaining_limit <= 0 and fallback_images:
+                            logger.warning(
+                                "Skip quoted fallback images due to limit=%d for umo=%s",
+                                config.max_quoted_fallback_images,
+                                event.unified_msg_origin,
+                            )
+                            continue
+                        if len(fallback_images) > remaining_limit:
+                            logger.warning(
+                                "Truncate quoted fallback images for umo=%s, reply_id=%s from %d to %d",
+                                event.unified_msg_origin,
+                                getattr(comp, "id", None),
+                                len(fallback_images),
+                                remaining_limit,
+                            )
+                            fallback_images = fallback_images[:remaining_limit]
+                        for image_ref in fallback_images:
+                            if image_ref in req.image_urls:
+                                continue
+                            req.image_urls.append(image_ref)
+                            fallback_quoted_image_count += 1
+                            _append_quoted_image_attachment(req, image_ref)
+                    except Exception as exc:  # noqa: BLE001
+                        logger.warning(
+                            "Failed to resolve fallback quoted images for umo=%s, reply_id=%s: %s",
+                            event.unified_msg_origin,
+                            getattr(comp, "id", None),
+                            exc,
+                            exc_info=True,
+                        )

            conversation = await _get_session_conv(event, plugin_context)
            req.conversation = conversation
@@ -890,6 +1052,7 @@ async def build_main_agent(

    if isinstance(req.contexts, str):
        req.contexts = json.loads(req.contexts)
+    req.image_urls = normalize_and_dedupe_strings(req.image_urls)

    if config.file_extract_enabled:
        try:
@@ -974,6 +1137,9 @@ async def build_main_agent(
        truncate_turns=config.dequeue_context_length,
        enforce_max_turns=config.max_context_length,
        tool_schema_mode=config.tool_schema_mode,
+        fallback_providers=_get_fallback_chat_providers(
+            provider, plugin_context, config.provider_settings
+        ),
    )

    if apply_reset:
@@ -1,6 +1,7 @@
 import base64
 import json
 import os
+import uuid

 from pydantic import Field
 from pydantic.dataclasses import dataclass
@@ -240,7 +241,9 @@ class SendMessageToUserTool(FunctionTool[AstrAgentContext]):
            if "_&exists_" in json.dumps(result):
                # Download the file from sandbox
                name = os.path.basename(path)
-                local_path = os.path.join(get_astrbot_temp_path(), name)
+                local_path = os.path.join(
+                    get_astrbot_temp_path(), f"sandbox_{uuid.uuid4().hex[:4]}_{name}"
+                )
                await sb.download_file(path, local_path)
                logger.info(f"Downloaded file from sandbox: {path} -> {local_path}")
                return local_path, True
@@ -352,11 +355,11 @@ class SendMessageToUserTool(FunctionTool[AstrAgentContext]):
            MessageChain(chain=components),
        )

-        if file_from_sandbox:
-            try:
-                os.remove(local_path)
-            except Exception as e:
-                logger.error(f"Error removing temp file {local_path}: {e}")
+        # if file_from_sandbox:
+        #     try:
+        #         os.remove(local_path)
+        #     except Exception as e:
+        #         logger.error(f"Error removing temp file {local_path}: {e}")

        return f"Message sent to session {target_session}"

@@ -11,6 +11,7 @@ from astrbot.core.db.po import (
    CommandConflict,
    ConversationV2,
    Persona,
+    PersonaFolder,
    PlatformMessageHistory,
    PlatformSession,
    PlatformStat,
@@ -39,6 +40,7 @@ MAIN_DB_MODELS: dict[str, type[SQLModel]] = {
    "platform_stats": PlatformStat,
    "conversations": ConversationV2,
    "personas": Persona,
+    "persona_folders": PersonaFolder,
    "preferences": Preference,
    "platform_message_history": PlatformMessageHistory,
    "platform_sessions": PlatformSession,
@@ -1,4 +1,5 @@
 import os
+import uuid
 from dataclasses import dataclass, field

 from astrbot.api import FunctionTool, logger
@@ -167,7 +168,9 @@ class FileDownloadTool(FunctionTool):
        try:
            name = os.path.basename(remote_path)

-            local_path = os.path.join(get_astrbot_temp_path(), name)
+            local_path = os.path.join(
+                get_astrbot_temp_path(), f"sandbox_{uuid.uuid4().hex[:4]}_{name}"
+            )

            # Download file from sandbox
            await sb.download_file(remote_path, local_path)
@@ -183,12 +186,12 @@ class FileDownloadTool(FunctionTool):
                    logger.error(f"Error sending file message: {e}")

                # remove
-                try:
-                    os.remove(local_path)
-                except Exception as e:
-                    logger.error(f"Error removing temp file {local_path}: {e}")
+                # try:
+                #     os.remove(local_path)
+                # except Exception as e:
+                #     logger.error(f"Error removing temp file {local_path}: {e}")

-                return f"File downloaded successfully to {local_path} and sent to user. The file has been removed from local storage."
+                return f"File downloaded successfully to {local_path} and sent to user."

            return f"File downloaded successfully to {local_path}"
        except Exception as e:
@@ -5,7 +5,7 @@ from typing import Any, TypedDict

 from astrbot.core.utils.astrbot_path import get_astrbot_data_path

-VERSION = "4.15.0"
+VERSION = "4.17.2"
 DB_PATH = os.path.join(get_astrbot_data_path(), "data_v4.db")

 WEBHOOK_SUPPORTED_PLATFORMS = [
@@ -15,6 +15,7 @@ WEBHOOK_SUPPORTED_PLATFORMS = [
    "wecom_ai_bot",
    "slack",
    "lark",
+    "line",
 ]

 # 默认配置
@@ -67,6 +68,7 @@ DEFAULT_CONFIG = {
    "provider_settings": {
        "enable": True,
        "default_provider_id": "",
+        "fallback_chat_models": [],
        "default_image_caption_provider_id": "",
        "image_caption_prompt": "Please describe the image using Chinese.",
        "provider_pool": ["*"],  # "*" 表示使用所有可用的提供者
@@ -99,6 +101,13 @@ DEFAULT_CONFIG = {
        "streaming_response": False,
        "show_tool_use_status": False,
        "sanitize_context_by_modalities": False,
+        "max_quoted_fallback_images": 20,
+        "quoted_message_parser": {
+            "max_component_chain_depth": 4,
+            "max_forward_node_depth": 6,
+            "max_forward_fetch": 32,
+            "warn_on_action_failure": False,
+        },
        "agent_runner_type": "local",
        "dify_agent_runner_provider_id": "",
        "coze_agent_runner_provider_id": "",
@@ -187,6 +196,12 @@ DEFAULT_CONFIG = {
        "host": "0.0.0.0",
        "port": 6185,
        "disable_access_log": True,
+        "ssl": {
+            "enable": False,
+            "cert_file": "",
+            "key_file": "",
+            "ca_certs": "",
+        },
    },
    "platform": [],
    "platform_specific": {
@@ -203,6 +218,7 @@ DEFAULT_CONFIG = {
    "log_file_enable": False,
    "log_file_path": "logs/astrbot.log",
    "log_file_max_mb": 20,
+    "temp_dir_max_size": 1024,
    "trace_enable": False,
    "trace_log_enable": False,
    "trace_log_path": "logs/astrbot.trace.log",
@@ -407,6 +423,7 @@ CONFIG_METADATA_2 = {
                        "slack_webhook_port": 6197,
                        "slack_webhook_path": "/astrbot-slack-webhook/callback",
                    },
+                    # LINE's config is located in line_adapter.py
                    "Satori": {
                        "id": "satori",
                        "type": "satori",
@@ -2197,6 +2214,10 @@ CONFIG_METADATA_2 = {
                    "default_provider_id": {
                        "type": "string",
                    },
+                    "fallback_chat_models": {
+                        "type": "list",
+                        "items": {"type": "string"},
+                    },
                    "wake_prefix": {
                        "type": "string",
                    },
@@ -2391,9 +2412,23 @@ CONFIG_METADATA_2 = {
                "type": "string",
                "options": ["DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL"],
            },
+            "dashboard.ssl.enable": {"type": "bool"},
+            "dashboard.ssl.cert_file": {
+                "type": "string",
+                "condition": {"dashboard.ssl.enable": True},
+            },
+            "dashboard.ssl.key_file": {
+                "type": "string",
+                "condition": {"dashboard.ssl.enable": True},
+            },
+            "dashboard.ssl.ca_certs": {
+                "type": "string",
+                "condition": {"dashboard.ssl.enable": True},
+            },
            "log_file_enable": {"type": "bool"},
            "log_file_path": {"type": "string", "condition": {"log_file_enable": True}},
            "log_file_max_mb": {"type": "int", "condition": {"log_file_enable": True}},
+            "temp_dir_max_size": {"type": "int"},
            "trace_log_enable": {"type": "bool"},
            "trace_log_path": {
                "type": "string",
@@ -2493,15 +2528,22 @@ CONFIG_METADATA_3 = {
            },
            "ai": {
                "description": "模型",
-                "hint": "当使用非内置 Agent 执行器时，默认聊天模型和默认图片转述模型可能会无效，但某些插件会依赖此配置项来调用 AI 能力。",
+                "hint": "当使用非内置 Agent 执行器时，默认对话模型和默认图片转述模型可能会无效，但某些插件会依赖此配置项来调用 AI 能力。",
                "type": "object",
                "items": {
                    "provider_settings.default_provider_id": {
-                        "description": "默认聊天模型",
+                        "description": "默认对话模型",
                        "type": "string",
                        "_special": "select_provider",
                        "hint": "留空时使用第一个模型",
                    },
+                    "provider_settings.fallback_chat_models": {
+                        "description": "回退对话模型列表",
+                        "type": "list",
+                        "items": {"type": "string"},
+                        "_special": "select_providers",
+                        "hint": "主聊天模型请求失败时，按顺序切换到这些模型。",
+                    },
                    "provider_settings.default_image_caption_provider_id": {
                        "description": "默认图片转述模型",
                        "type": "string",
@@ -2906,6 +2948,46 @@ CONFIG_METADATA_3 = {
                            "provider_settings.agent_runner_type": "local",
                        },
                    },
+                    "provider_settings.max_quoted_fallback_images": {
+                        "description": "引用图片回退解析上限",
+                        "type": "int",
+                        "hint": "引用/转发消息回退解析图片时的最大注入数量，超出会截断。",
+                        "condition": {
+                            "provider_settings.agent_runner_type": "local",
+                        },
+                    },
+                    "provider_settings.quoted_message_parser.max_component_chain_depth": {
+                        "description": "引用解析组件链深度",
+                        "type": "int",
+                        "hint": "解析 Reply 组件链时允许的最大递归深度。",
+                        "condition": {
+                            "provider_settings.agent_runner_type": "local",
+                        },
+                    },
+                    "provider_settings.quoted_message_parser.max_forward_node_depth": {
+                        "description": "引用解析转发节点深度",
+                        "type": "int",
+                        "hint": "解析合并转发节点时允许的最大递归深度。",
+                        "condition": {
+                            "provider_settings.agent_runner_type": "local",
+                        },
+                    },
+                    "provider_settings.quoted_message_parser.max_forward_fetch": {
+                        "description": "引用解析转发拉取上限",
+                        "type": "int",
+                        "hint": "递归拉取 get_forward_msg 的最大次数。",
+                        "condition": {
+                            "provider_settings.agent_runner_type": "local",
+                        },
+                    },
+                    "provider_settings.quoted_message_parser.warn_on_action_failure": {
+                        "description": "引用解析 action 失败告警",
+                        "type": "bool",
+                        "hint": "开启后，get_msg/get_forward_msg 全部尝试失败时输出 warning 日志。",
+                        "condition": {
+                            "provider_settings.agent_runner_type": "local",
+                        },
+                    },
                    "provider_settings.max_agent_step": {
                        "description": "工具调用轮数上限",
                        "type": "int",
@@ -3357,6 +3439,29 @@ CONFIG_METADATA_3_SYSTEM = {
                        "hint": "控制台输出日志的级别。",
                        "options": ["DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL"],
                    },
+                    "dashboard.ssl.enable": {
+                        "description": "启用 WebUI HTTPS",
+                        "type": "bool",
+                        "hint": "启用后，WebUI 将直接使用 HTTPS 提供服务。",
+                    },
+                    "dashboard.ssl.cert_file": {
+                        "description": "SSL 证书文件路径",
+                        "type": "string",
+                        "hint": "证书文件路径（PEM）。支持绝对路径和相对路径（相对于当前工作目录）。",
+                        "condition": {"dashboard.ssl.enable": True},
+                    },
+                    "dashboard.ssl.key_file": {
+                        "description": "SSL 私钥文件路径",
+                        "type": "string",
+                        "hint": "私钥文件路径（PEM）。支持绝对路径和相对路径（相对于当前工作目录）。",
+                        "condition": {"dashboard.ssl.enable": True},
+                    },
+                    "dashboard.ssl.ca_certs": {
+                        "description": "SSL CA 证书文件路径",
+                        "type": "string",
+                        "hint": "可选。用于指定 CA 证书文件路径。",
+                        "condition": {"dashboard.ssl.enable": True},
+                    },
                    "log_file_enable": {
                        "description": "启用文件日志",
                        "type": "bool",
@@ -3372,6 +3477,11 @@ CONFIG_METADATA_3_SYSTEM = {
                        "type": "int",
                        "hint": "超过大小后自动轮转，默认 20MB。",
                    },
+                    "temp_dir_max_size": {
+                        "description": "临时目录大小上限 (MB)",
+                        "type": "int",
+                        "hint": "用于限制 data/temp 目录总大小，单位为 MB。系统每 10 分钟检查一次，超限时按文件修改时间从旧到新删除，释放约 30% 当前体积。",
+                    },
                    "trace_log_enable": {
                        "description": "启用 Trace 文件日志",
                        "type": "bool",
@@ -37,6 +37,7 @@ from astrbot.core.umop_config_router import UmopConfigRouter
 from astrbot.core.updator import AstrBotUpdator
 from astrbot.core.utils.llm_metadata import update_llm_metadata
 from astrbot.core.utils.migra_helper import migra
+from astrbot.core.utils.temp_dir_cleaner import TempDirCleaner

 from . import astrbot_config, html_renderer
 from .event_bus import EventBus
@@ -57,6 +58,7 @@ class AstrBotCoreLifecycle:

        self.subagent_orchestrator: SubAgentOrchestrator | None = None
        self.cron_manager: CronJobManager | None = None
+        self.temp_dir_cleaner: TempDirCleaner | None = None

        # 设置代理
        proxy_config = self.astrbot_config.get("http_proxy", "")
@@ -125,6 +127,12 @@ class AstrBotCoreLifecycle:
            ucr=self.umop_config_router,
            sp=sp,
        )
+        self.temp_dir_cleaner = TempDirCleaner(
+            max_size_getter=lambda: self.astrbot_config_mgr.default_conf.get(
+                TempDirCleaner.CONFIG_KEY,
+                TempDirCleaner.DEFAULT_MAX_SIZE,
+            ),
+        )

        # apply migration
        try:
@@ -238,6 +246,12 @@ class AstrBotCoreLifecycle:
                self.cron_manager.start(self.star_context),
                name="cron_manager",
            )
+        temp_dir_cleaner_task = None
+        if self.temp_dir_cleaner:
+            temp_dir_cleaner_task = asyncio.create_task(
+                self.temp_dir_cleaner.run(),
+                name="temp_dir_cleaner",
+            )

        # 把插件中注册的所有协程函数注册到事件总线中并执行
        extra_tasks = []
@@ -247,6 +261,8 @@ class AstrBotCoreLifecycle:
        tasks_ = [event_bus_task, *(extra_tasks if extra_tasks else [])]
        if cron_task:
            tasks_.append(cron_task)
+        if temp_dir_cleaner_task:
+            tasks_.append(temp_dir_cleaner_task)
        for task in tasks_:
            self.curr_tasks.append(
                asyncio.create_task(self._task_wrapper(task), name=task.get_name()),
@@ -298,6 +314,9 @@ class AstrBotCoreLifecycle:

    async def stop(self) -> None:
        """停止 AstrBot 核心生命周期管理类, 取消所有当前任务并终止各个管理器."""
+        if self.temp_dir_cleaner:
+            await self.temp_dir_cleaner.stop()
+
        # 请求停止所有正在运行的异步任务
        for task in self.curr_tasks:
            task.cancel()
@@ -1,24 +1,4 @@
-"""日志系统, 用于支持核心组件和插件的日志记录, 提供了日志订阅功能
-
-const:
-    CACHED_SIZE: 日志缓存大小, 用于限制缓存的日志数量
-    log_color_config: 日志颜色配置, 定义了不同日志级别的颜色
-
-class:
-    LogBroker: 日志代理类, 用于缓存和分发日志消息
-    LogQueueHandler: 日志处理器, 用于将日志消息发送到 LogBroker
-    LogManager: 日志管理器, 用于创建和配置日志记录器
-
-function:
-    is_plugin_path: 检查文件路径是否来自插件目录
-    get_short_level_name: 将日志级别名称转换为四个字母的缩写
-
-工作流程:
-1. 通过 LogManager.GetLogger() 获取日志器, 配置了控制台输出和多个格式化过滤器
-2. 通过 set_queue_handler() 设置日志处理器, 将日志消息发送到 LogBroker
-3. logBroker 维护一个订阅者列表, 负责将日志分发给所有订阅者
-4. 订阅者可以使用 register() 方法注册到 LogBroker, 订阅日志流
-"""
+"""日志系统，统一将标准 logging 输出转发到 loguru。"""

 import asyncio
 import logging
@@ -27,54 +7,59 @@ import sys
 import time
 from asyncio import Queue
 from collections import deque
-from logging.handlers import RotatingFileHandler
+from typing import TYPE_CHECKING

-import colorlog
+from loguru import logger as _raw_loguru_logger

 from astrbot.core.config.default import VERSION
 from astrbot.core.utils.astrbot_path import get_astrbot_data_path

-# 日志缓存大小
 CACHED_SIZE = 500
-# 日志颜色配置
-log_color_config = {
-    "DEBUG": "green",
-    "INFO": "bold_cyan",
-    "WARNING": "bold_yellow",
-    "ERROR": "red",
-    "CRITICAL": "bold_red",
-    "RESET": "reset",
-    "asctime": "green",
-}
+
+if TYPE_CHECKING:
+    from loguru import Record


-def is_plugin_path(pathname):
-    """检查文件路径是否来自插件目录
+class _RecordEnricherFilter(logging.Filter):
+    """为 logging.LogRecord 注入 AstrBot 日志字段。"""

-    Args:
-        pathname (str): 文件路径
+    def filter(self, record: logging.LogRecord) -> bool:
+        record.plugin_tag = "[Plug]" if _is_plugin_path(record.pathname) else "[Core]"
+        record.short_levelname = _get_short_level_name(record.levelname)
+        record.astrbot_version_tag = (
+            f" [v{VERSION}]" if record.levelno >= logging.WARNING else ""
+        )
+        record.source_file = _build_source_file(record.pathname)
+        record.source_line = record.lineno
+        record.is_trace = record.name == "astrbot.trace"
+        return True

-    Returns:
-        bool: 如果路径来自插件目录，则返回 True，否则返回 False

-    """
+class _QueueAnsiColorFilter(logging.Filter):
+    """Attach ANSI color prefix for WebUI console rendering."""
+
+    _LEVEL_COLOR = {
+        "DEBUG": "\u001b[1;34m",
+        "INFO": "\u001b[1;36m",
+        "WARNING": "\u001b[1;33m",
+        "ERROR": "\u001b[31m",
+        "CRITICAL": "\u001b[1;31m",
+    }
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        record.ansi_prefix = self._LEVEL_COLOR.get(record.levelname, "\u001b[0m")
+        record.ansi_reset = "\u001b[0m"
+        return True
+
+
+def _is_plugin_path(pathname: str | None) -> bool:
    if not pathname:
        return False
-
    norm_path = os.path.normpath(pathname)
    return ("data/plugins" in norm_path) or ("astrbot/builtin_stars/" in norm_path)


-def get_short_level_name(level_name):
-    """将日志级别名称转换为四个字母的缩写
-
-    Args:
-        level_name (str): 日志级别名称, 如 "DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL"
-
-    Returns:
-        str: 四个字母的日志级别缩写
-
-    """
+def _get_short_level_name(level_name: str) -> str:
    level_map = {
        "DEBUG": "DBUG",
        "INFO": "INFO",
@@ -85,44 +70,75 @@ def get_short_level_name(level_name):
    return level_map.get(level_name, level_name[:4].upper())


-class LogBroker:
-    """日志代理类, 用于缓存和分发日志消息
+def _build_source_file(pathname: str | None) -> str:
+    if not pathname:
+        return "unknown"
+    dirname = os.path.dirname(pathname)
+    return (
+        os.path.basename(dirname) + "." + os.path.basename(pathname).replace(".py", "")
+    )

-    发布-订阅模式
-    """
+
+def _patch_record(record: "Record") -> None:
+    extra = record["extra"]
+    extra.setdefault("plugin_tag", "[Core]")
+    extra.setdefault("short_levelname", _get_short_level_name(record["level"].name))
+    level_no = record["level"].no
+    extra.setdefault("astrbot_version_tag", f" [v{VERSION}]" if level_no >= 30 else "")
+    extra.setdefault("source_file", _build_source_file(record["file"].path))
+    extra.setdefault("source_line", record["line"])
+    extra.setdefault("is_trace", False)
+
+
+_loguru = _raw_loguru_logger.patch(_patch_record)
+
+
+class _LoguruInterceptHandler(logging.Handler):
+    """将 logging 记录转发到 loguru。"""
+
+    def emit(self, record: logging.LogRecord) -> None:
+        try:
+            level: str | int = _loguru.level(record.levelname).name
+        except ValueError:
+            level = record.levelno
+
+        payload = {
+            "plugin_tag": getattr(record, "plugin_tag", "[Core]"),
+            "short_levelname": getattr(
+                record,
+                "short_levelname",
+                _get_short_level_name(record.levelname),
+            ),
+            "astrbot_version_tag": getattr(record, "astrbot_version_tag", ""),
+            "source_file": getattr(
+                record, "source_file", _build_source_file(record.pathname)
+            ),
+            "source_line": getattr(record, "source_line", record.lineno),
+            "is_trace": getattr(record, "is_trace", record.name == "astrbot.trace"),
+        }
+
+        _loguru.bind(**payload).opt(exception=record.exc_info).log(
+            level,
+            record.getMessage(),
+        )
+
+
+class LogBroker:
+    """日志代理类，用于缓存和分发日志消息。"""

    def __init__(self) -> None:
-        self.log_cache = deque(maxlen=CACHED_SIZE)  # 环形缓冲区, 保存最近的日志
-        self.subscribers: list[Queue] = []  # 订阅者列表
+        self.log_cache = deque(maxlen=CACHED_SIZE)
+        self.subscribers: list[Queue] = []

    def register(self) -> Queue:
-        """注册新的订阅者, 并给每个订阅者返回一个带有日志缓存的队列
-
-        Returns:
-            Queue: 订阅者的队列, 可用于接收日志消息
-
-        """
        q = Queue(maxsize=CACHED_SIZE + 10)
        self.subscribers.append(q)
        return q

    def unregister(self, q: Queue) -> None:
-        """取消订阅
-
-        Args:
-            q (Queue): 需要取消订阅的队列
-
-        """
        self.subscribers.remove(q)

    def publish(self, log_entry: dict) -> None:
-        """发布新日志到所有订阅者, 使用非阻塞方式投递, 避免一个订阅者阻塞整个系统
-
-        Args:
-            log_entry (dict): 日志消息, 包含日志级别和日志内容.
-                example: {"level": "INFO", "data": "This is a log message.", "time": "2023-10-01 12:00:00"}
-
-        """
        self.log_cache.append(log_entry)
        for q in self.subscribers:
            try:
@@ -132,23 +148,13 @@ class LogBroker:


 class LogQueueHandler(logging.Handler):
-    """日志处理器, 用于将日志消息发送到 LogBroker
-
-    继承自 logging.Handler
-    """
+    """日志处理器，用于将日志消息发送到 LogBroker。"""

    def __init__(self, log_broker: LogBroker) -> None:
        super().__init__()
        self.log_broker = log_broker

-    def emit(self, record) -> None:
-        """日志处理的入口方法, 接受一个日志记录, 转换为字符串后由 LogBroker 发布
-        这个方法会在每次日志记录时被调用
-
-        Args:
-            record (logging.LogRecord): 日志记录对象, 包含日志信息
-
-        """
+    def emit(self, record: logging.LogRecord) -> None:
        log_entry = self.format(record)
        self.log_broker.publish(
            {
@@ -160,117 +166,16 @@ class LogQueueHandler(logging.Handler):


 class LogManager:
-    """日志管理器, 用于创建和配置日志记录器
+    _LOGGER_HANDLER_FLAG = "_astrbot_loguru_handler"
+    _ENRICH_FILTER_FLAG = "_astrbot_enrich_filter"

-    提供了获取默认日志记录器logger和设置队列处理器的方法
-    """
-
-    _FILE_HANDLER_FLAG = "_astrbot_file_handler"
-    _TRACE_FILE_HANDLER_FLAG = "_astrbot_trace_file_handler"
-
-    @classmethod
-    def GetLogger(cls, log_name: str = "default"):
-        """获取指定名称的日志记录器logger
-
-        Args:
-            log_name (str): 日志记录器的名称, 默认为 "default"
-
-        Returns:
-            logging.Logger: 返回配置好的日志记录器
-
-        """
-        logger = logging.getLogger(log_name)
-        # 检查该logger或父级logger是否已经有处理器, 如果已经有处理器, 直接返回该logger, 避免重复配置
-        if logger.hasHandlers():
-            return logger
-        # 如果logger没有处理器
-        console_handler = logging.StreamHandler(
-            sys.stdout,
-        )  # 创建一个StreamHandler用于控制台输出
-        console_handler.setLevel(
-            logging.DEBUG,
-        )  # 将日志级别设置为DEBUG(最低级别, 显示所有日志), *如果插件没有设置级别, 默认为DEBUG
-
-        # 创建彩色日志格式化器, 输出日志格式为: [时间] [插件标签] [日志级别] [文件名:行号]: 日志消息
-        console_formatter = colorlog.ColoredFormatter(
-            fmt="%(log_color)s [%(asctime)s] %(plugin_tag)s [%(short_levelname)-4s]%(astrbot_version_tag)s [%(filename)s:%(lineno)d]: %(message)s %(reset)s",
-            datefmt="%H:%M:%S",
-            log_colors=log_color_config,
-        )
-
-        class PluginFilter(logging.Filter):
-            """插件过滤器类, 用于标记日志来源是插件还是核心组件"""
-
-            def filter(self, record) -> bool:
-                record.plugin_tag = (
-                    "[Plug]" if is_plugin_path(record.pathname) else "[Core]"
-                )
-                return True
-
-        class FileNameFilter(logging.Filter):
-            """文件名过滤器类, 用于修改日志记录的文件名格式
-            例如: 将文件路径 /path/to/file.py 转换为 file.<file> 格式
-            """
-
-            # 获取这个文件和父文件夹的名字：<folder>.<file> 并且去除 .py
-            def filter(self, record) -> bool:
-                dirname = os.path.dirname(record.pathname)
-                record.filename = (
-                    os.path.basename(dirname)
-                    + "."
-                    + os.path.basename(record.pathname).replace(".py", "")
-                )
-                return True
-
-        class LevelNameFilter(logging.Filter):
-            """短日志级别名称过滤器类, 用于将日志级别名称转换为四个字母的缩写"""
-
-            # 添加短日志级别名称
-            def filter(self, record) -> bool:
-                record.short_levelname = get_short_level_name(record.levelname)
-                return True
-
-        class AstrBotVersionTagFilter(logging.Filter):
-            """在 WARNING 及以上级别日志后追加当前 AstrBot 版本号。"""
-
-            def filter(self, record) -> bool:
-                if record.levelno >= logging.WARNING:
-                    record.astrbot_version_tag = f" [v{VERSION}]"
-                else:
-                    record.astrbot_version_tag = ""
-                return True
-
-        console_handler.setFormatter(console_formatter)  # 设置处理器的格式化器
-        logger.addFilter(PluginFilter())  # 添加插件过滤器
-        logger.addFilter(FileNameFilter())  # 添加文件名过滤器
-        logger.addFilter(LevelNameFilter())  # 添加级别名称过滤器
-        logger.addFilter(AstrBotVersionTagFilter())  # 追加版本号（WARNING 及以上）
-        logger.setLevel(logging.DEBUG)  # 设置日志级别为DEBUG
-        logger.addHandler(console_handler)  # 添加处理器到logger
-
-        return logger
-
-    @classmethod
-    def set_queue_handler(cls, logger: logging.Logger, log_broker: LogBroker) -> None:
-        """设置队列处理器, 用于将日志消息发送到 LogBroker
-
-        Args:
-            logger (logging.Logger): 日志记录器
-            log_broker (LogBroker): 日志代理类, 用于缓存和分发日志消息
-
-        """
-        handler = LogQueueHandler(log_broker)
-        handler.setLevel(logging.DEBUG)
-        if logger.handlers:
-            handler.setFormatter(logger.handlers[0].formatter)
-        else:
-            # 为队列处理器设置相同格式的formatter
-            handler.setFormatter(
-                logging.Formatter(
-                    "[%(asctime)s] [%(short_levelname)s] %(plugin_tag)s[%(filename)s:%(lineno)d]: %(message)s",
-                ),
-            )
-        logger.addHandler(handler)
+    _configured = False
+    _console_sink_id: int | None = None
+    _file_sink_id: int | None = None
+    _trace_sink_id: int | None = None
+    _NOISY_LOGGER_LEVELS: dict[str, int] = {
+        "aiosqlite": logging.WARNING,
+    }

    @classmethod
    def _default_log_path(cls) -> str:
@@ -285,79 +190,147 @@ class LogManager:
        return os.path.join(get_astrbot_data_path(), configured_path)

    @classmethod
-    def _get_file_handlers(cls, logger: logging.Logger) -> list[logging.Handler]:
-        return [
-            handler
-            for handler in logger.handlers
-            if getattr(handler, cls._FILE_HANDLER_FLAG, False)
-        ]
+    def _setup_loguru(cls) -> None:
+        if cls._configured:
+            return

-    @classmethod
-    def _get_trace_file_handlers(cls, logger: logging.Logger) -> list[logging.Handler]:
-        return [
-            handler
-            for handler in logger.handlers
-            if getattr(handler, cls._TRACE_FILE_HANDLER_FLAG, False)
-        ]
-
-    @classmethod
-    def _remove_file_handlers(cls, logger: logging.Logger) -> None:
-        for handler in cls._get_file_handlers(logger):
-            logger.removeHandler(handler)
-            try:
-                handler.close()
-            except Exception:
-                pass
-
-    @classmethod
-    def _remove_trace_file_handlers(cls, logger: logging.Logger) -> None:
-        for handler in cls._get_trace_file_handlers(logger):
-            logger.removeHandler(handler)
-            try:
-                handler.close()
-            except Exception:
-                pass
-
-    @classmethod
-    def _add_file_handler(
-        cls,
-        logger: logging.Logger,
-        file_path: str,
-        max_mb: int | None = None,
-        backup_count: int = 3,
-        trace: bool = False,
-    ) -> None:
-        os.makedirs(os.path.dirname(file_path) or ".", exist_ok=True)
-        max_bytes = 0
-        if max_mb and max_mb > 0:
-            max_bytes = max_mb * 1024 * 1024
-        if max_bytes > 0:
-            file_handler = RotatingFileHandler(
-                file_path,
-                maxBytes=max_bytes,
-                backupCount=backup_count,
-                encoding="utf-8",
-            )
-        else:
-            file_handler = logging.FileHandler(file_path, encoding="utf-8")
-        file_handler.setLevel(logger.level)
-        if trace:
-            formatter = logging.Formatter(
-                "[%(asctime)s] %(message)s",
-                datefmt="%Y-%m-%d %H:%M:%S",
-            )
-        else:
-            formatter = logging.Formatter(
-                "[%(asctime)s] %(plugin_tag)s [%(short_levelname)s]%(astrbot_version_tag)s [%(filename)s:%(lineno)d]: %(message)s",
-                datefmt="%Y-%m-%d %H:%M:%S",
-            )
-        file_handler.setFormatter(formatter)
-        setattr(
-            file_handler,
-            cls._TRACE_FILE_HANDLER_FLAG if trace else cls._FILE_HANDLER_FLAG,
-            True,
+        _loguru.remove()
+        cls._console_sink_id = _loguru.add(
+            sys.stdout,
+            level="DEBUG",
+            colorize=True,
+            filter=lambda record: not record["extra"].get("is_trace", False),
+            format=(
+                "<green>[{time:HH:mm:ss.SSS}]</green> {extra[plugin_tag]} "
+                "<level>[{extra[short_levelname]}]</level>{extra[astrbot_version_tag]} "
+                "[{extra[source_file]}:{extra[source_line]}]: <level>{message}</level>"
+            ),
+        )
+        cls._configured = True
+
+    @classmethod
+    def _setup_root_bridge(cls) -> None:
+        root_logger = logging.getLogger()
+
+        has_handler = any(
+            getattr(handler, cls._LOGGER_HANDLER_FLAG, False)
+            for handler in root_logger.handlers
+        )
+        if not has_handler:
+            handler = _LoguruInterceptHandler()
+            setattr(handler, cls._LOGGER_HANDLER_FLAG, True)
+            root_logger.addHandler(handler)
+        root_logger.setLevel(logging.DEBUG)
+        for name, level in cls._NOISY_LOGGER_LEVELS.items():
+            logging.getLogger(name).setLevel(level)
+
+    @classmethod
+    def _ensure_logger_enricher_filter(cls, logger: logging.Logger) -> None:
+        has_filter = any(
+            getattr(existing_filter, cls._ENRICH_FILTER_FLAG, False)
+            for existing_filter in logger.filters
+        )
+        if not has_filter:
+            enrich_filter = _RecordEnricherFilter()
+            setattr(enrich_filter, cls._ENRICH_FILTER_FLAG, True)
+            logger.addFilter(enrich_filter)
+
+    @classmethod
+    def _ensure_logger_intercept_handler(cls, logger: logging.Logger) -> None:
+        has_handler = any(
+            getattr(handler, cls._LOGGER_HANDLER_FLAG, False)
+            for handler in logger.handlers
+        )
+        if not has_handler:
+            handler = _LoguruInterceptHandler()
+            setattr(handler, cls._LOGGER_HANDLER_FLAG, True)
+            logger.addHandler(handler)
+
+    @classmethod
+    def GetLogger(cls, log_name: str = "default") -> logging.Logger:
+        cls._setup_loguru()
+        cls._setup_root_bridge()
+
+        logger = logging.getLogger(log_name)
+        cls._ensure_logger_enricher_filter(logger)
+        cls._ensure_logger_intercept_handler(logger)
+        logger.setLevel(logging.DEBUG)
+        logger.propagate = False
+        return logger
+
+    @classmethod
+    def set_queue_handler(cls, logger: logging.Logger, log_broker: LogBroker) -> None:
+        cls._ensure_logger_enricher_filter(logger)
+
+        for handler in logger.handlers:
+            if isinstance(handler, LogQueueHandler):
+                return
+
+        handler = LogQueueHandler(log_broker)
+        handler.setLevel(logging.DEBUG)
+        handler.addFilter(_QueueAnsiColorFilter())
+        handler.setFormatter(
+            logging.Formatter(
+                "%(ansi_prefix)s[%(asctime)s.%(msecs)03d] %(plugin_tag)s [%(short_levelname)s]%(astrbot_version_tag)s "
+                "[%(source_file)s:%(source_line)d]: %(message)s%(ansi_reset)s",
+                datefmt="%Y-%m-%d %H:%M:%S",
+            ),
+        )
+        logger.addHandler(handler)
+
+    @classmethod
+    def _remove_sink(cls, sink_id: int | None) -> None:
+        if sink_id is None:
+            return
+        try:
+            _loguru.remove(sink_id)
+        except ValueError:
+            pass
+
+    @classmethod
+    def _add_file_sink(
+        cls,
+        *,
+        file_path: str,
+        level: int,
+        max_mb: int | None,
+        backup_count: int,
+        trace: bool,
+    ) -> int:
+        os.makedirs(os.path.dirname(file_path) or ".", exist_ok=True)
+        rotation = f"{max_mb} MB" if max_mb and max_mb > 0 else None
+        retention = (
+            backup_count if rotation and backup_count and backup_count > 0 else None
+        )
+        if trace:
+            return _loguru.add(
+                file_path,
+                level="INFO",
+                format="[{time:YYYY-MM-DD HH:mm:ss.SSS}] {message}",
+                encoding="utf-8",
+                rotation=rotation,
+                retention=retention,
+                enqueue=True,
+                filter=lambda record: record["extra"].get("is_trace", False),
+            )
+
+        logging_level_name = logging.getLevelName(level)
+        if isinstance(logging_level_name, int):
+            logging_level_name = "INFO"
+        return _loguru.add(
+            file_path,
+            level=logging_level_name,
+            format=(
+                "[{time:YYYY-MM-DD HH:mm:ss.SSS}] {extra[plugin_tag]} "
+                "[{extra[short_levelname]}]{extra[astrbot_version_tag]} "
+                "[{extra[source_file]}:{extra[source_line]}]: {message}"
+            ),
+            encoding="utf-8",
+            rotation=rotation,
+            retention=retention,
+            enqueue=True,
+            filter=lambda record: not record["extra"].get("is_trace", False),
        )
-        logger.addHandler(file_handler)

    @classmethod
    def configure_logger(
@@ -366,13 +339,6 @@ class LogManager:
        config: dict | None,
        override_level: str | None = None,
    ) -> None:
-        """根据配置设置日志级别和文件日志。
-
-        Args:
-            logger: 需要配置的 logger
-            config: 配置字典
-            override_level: 若提供，将覆盖配置中的日志级别
-        """
        if not config:
            return

@@ -383,7 +349,6 @@ class LogManager:
            except Exception:
                logger.setLevel(logging.INFO)

-        # 兼容旧版嵌套配置
        if "log_file" in config:
            file_conf = config.get("log_file") or {}
            enable_file = bool(file_conf.get("enable", False))
@@ -394,27 +359,25 @@ class LogManager:
            file_path = config.get("log_file_path")
            max_mb = config.get("log_file_max_mb")

-        file_path = cls._resolve_log_path(file_path)
+        cls._remove_sink(cls._file_sink_id)
+        cls._file_sink_id = None

-        existing = cls._get_file_handlers(logger)
        if not enable_file:
-            cls._remove_file_handlers(logger)
            return

-        # 如果已有文件处理器且路径一致，则仅同步级别
-        if existing:
-            handler = existing[0]
-            base = getattr(handler, "baseFilename", "")
-            if base and os.path.abspath(base) == os.path.abspath(file_path):
-                handler.setLevel(logger.level)
-                return
-            cls._remove_file_handlers(logger)
-
-        cls._add_file_handler(logger, file_path, max_mb=max_mb)
+        try:
+            cls._file_sink_id = cls._add_file_sink(
+                file_path=cls._resolve_log_path(file_path),
+                level=logger.level,
+                max_mb=max_mb,
+                backup_count=3,
+                trace=False,
+            )
+        except Exception as e:
+            logger.error(f"Failed to add file sink: {e}")

    @classmethod
    def configure_trace_logger(cls, config: dict | None) -> None:
-        """为 trace 事件配置独立的文件日志，不向控制台输出。"""
        if not config:
            return

@@ -429,28 +392,22 @@ class LogManager:
            path = path or legacy.get("trace_path")
            max_mb = max_mb or legacy.get("trace_max_mb")

-        if not enable:
-            trace_logger = logging.getLogger("astrbot.trace")
-            cls._remove_trace_file_handlers(trace_logger)
-            return
-
-        file_path = cls._resolve_log_path(path or "logs/astrbot.trace.log")
        trace_logger = logging.getLogger("astrbot.trace")
+        cls._ensure_logger_enricher_filter(trace_logger)
+        cls._ensure_logger_intercept_handler(trace_logger)
        trace_logger.setLevel(logging.INFO)
        trace_logger.propagate = False

-        existing = cls._get_trace_file_handlers(trace_logger)
-        if existing:
-            handler = existing[0]
-            base = getattr(handler, "baseFilename", "")
-            if base and os.path.abspath(base) == os.path.abspath(file_path):
-                handler.setLevel(trace_logger.level)
-                return
-            cls._remove_trace_file_handlers(trace_logger)
+        cls._remove_sink(cls._trace_sink_id)
+        cls._trace_sink_id = None

-        cls._add_file_handler(
-            trace_logger,
-            file_path,
+        if not enable:
+            return
+
+        cls._trace_sink_id = cls._add_file_sink(
+            file_path=cls._resolve_log_path(path or "logs/astrbot.trace.log"),
+            level=logging.INFO,
            max_mb=max_mb,
+            backup_count=3,
            trace=True,
        )
@@ -31,7 +31,7 @@ from enum import Enum
 from pydantic.v1 import BaseModel

 from astrbot.core import astrbot_config, file_token_service, logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_file, download_image_by_url, file_to_base64


@@ -156,8 +156,9 @@ class Record(BaseMessageComponent):
        if self.file.startswith("base64://"):
            bs64_data = self.file.removeprefix("base64://")
            image_bytes = base64.b64decode(bs64_data)
-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-            file_path = os.path.join(temp_dir, f"{uuid.uuid4()}.jpg")
+            file_path = os.path.join(
+                get_astrbot_temp_path(), f"recordseg_{uuid.uuid4()}.jpg"
+            )
            with open(file_path, "wb") as f:
                f.write(image_bytes)
            return os.path.abspath(file_path)
@@ -245,8 +246,9 @@ class Video(BaseMessageComponent):
        if url and url.startswith("file:///"):
            return url[8:]
        if url and url.startswith("http"):
-            download_dir = os.path.join(get_astrbot_data_path(), "temp")
-            video_file_path = os.path.join(download_dir, f"{uuid.uuid4().hex}")
+            video_file_path = os.path.join(
+                get_astrbot_temp_path(), f"videoseg_{uuid.uuid4().hex}"
+            )
            await download_file(url, video_file_path)
            if os.path.exists(video_file_path):
                return os.path.abspath(video_file_path)
@@ -445,8 +447,9 @@ class Image(BaseMessageComponent):
        if url.startswith("base64://"):
            bs64_data = url.removeprefix("base64://")
            image_bytes = base64.b64decode(bs64_data)
-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-            image_file_path = os.path.join(temp_dir, f"{uuid.uuid4()}.jpg")
+            image_file_path = os.path.join(
+                get_astrbot_temp_path(), f"imgseg_{uuid.uuid4()}.jpg"
+            )
            with open(image_file_path, "wb") as f:
                f.write(image_bytes)
            return os.path.abspath(image_file_path)
@@ -725,13 +728,12 @@ class File(BaseMessageComponent):
        """下载文件"""
        if not self.url:
            raise ValueError("Download failed: No URL provided in File component.")
-        download_dir = os.path.join(get_astrbot_data_path(), "temp")
-        os.makedirs(download_dir, exist_ok=True)
+        download_dir = get_astrbot_temp_path()
        if self.name:
            name, ext = os.path.splitext(self.name)
-            filename = f"{name}_{uuid.uuid4().hex[:8]}{ext}"
+            filename = f"fileseg_{name}_{uuid.uuid4().hex[:8]}{ext}"
        else:
-            filename = f"{uuid.uuid4().hex}"
+            filename = f"fileseg_{uuid.uuid4().hex}"
        file_path = os.path.join(download_dir, filename)
        await download_file(self.url, file_path)
        self.file_ = os.path.abspath(file_path)
@@ -123,6 +123,7 @@ class InternalAgentSubStage(Stage):
            provider_settings=settings,
            subagent_orchestrator=conf.get("subagent_orchestrator", {}),
            timezone=self.ctx.plugin_manager.context.get_config().get("timezone"),
+            max_quoted_fallback_images=settings.get("max_quoted_fallback_images", 20),
        )

    async def process(
@@ -149,6 +150,7 @@ class InternalAgentSubStage(Stage):

            logger.debug("ready to request llm provider")

+            await event.send_typing()
            await call_event_hook(event, EventType.OnWaitingLLMRequestEvent)

            async with session_lock_manager.acquire_lock(event.unified_msg_origin):
@@ -190,6 +192,8 @@ class InternalAgentSubStage(Stage):
                )

                if await call_event_hook(event, EventType.OnLLMRequestEvent, req):
+                    if reset_coro:
+                        reset_coro.close()
                    return

                # apply reset
@@ -61,16 +61,17 @@ class RespondStage(Stage):
        self.log_base = float(
            ctx.astrbot_config["platform_settings"]["segmented_reply"]["log_base"],
        )
-        interval_str: str = ctx.astrbot_config["platform_settings"]["segmented_reply"][
-            "interval"
-        ]
-        interval_str_ls = interval_str.replace(" ", "").split(",")
-        try:
-            self.interval = [float(t) for t in interval_str_ls]
-        except BaseException as e:
-            logger.error(f"解析分段回复的间隔时间失败。{e}")
-            self.interval = [1.5, 3.5]
-        logger.info(f"分段回复间隔时间：{self.interval}")
+        self.interval = [1.5, 3.5]
+        if self.enable_seg:
+            interval_str: str = ctx.astrbot_config["platform_settings"][
+                "segmented_reply"
+            ]["interval"]
+            interval_str_ls = interval_str.replace(" ", "").split(",")
+            try:
+                self.interval = [float(t) for t in interval_str_ls]
+            except BaseException as e:
+                logger.error(f"解析分段回复的间隔时间失败。{e}")
+            logger.info(f"分段回复间隔时间：{self.interval}")

    async def _word_cnt(self, text: str) -> int:
        """分段回复 统计字数"""
@@ -244,6 +244,12 @@ class AstrMessageEvent(abc.ABC):
        )
        self._has_send_oper = True

+    async def send_typing(self) -> None:
+        """发送输入中状态。
+
+        默认实现为空，由具体平台按需重写。
+        """
+
    async def _pre_send(self) -> None:
        """调度器会在执行 send() 前调用该方法 deprecated in v3.5.18"""

@@ -1,6 +1,7 @@
 import asyncio
 import traceback
 from asyncio import Queue
+from dataclasses import dataclass

 from astrbot.core import logger
 from astrbot.core.config.astrbot_config import AstrBotConfig
@@ -12,12 +13,19 @@ from .register import platform_cls_map
 from .sources.webchat.webchat_adapter import WebChatAdapter


+@dataclass
+class PlatformTasks:
+    run: asyncio.Task
+    wrapper: asyncio.Task
+
+
 class PlatformManager:
    def __init__(self, config: AstrBotConfig, event_queue: Queue) -> None:
        self.platform_insts: list[Platform] = []
        """加载的 Platform 的实例"""

        self._inst_map: dict[str, dict] = {}
+        self._platform_tasks: dict[str, PlatformTasks] = {}

        self.astrbot_config = config
        self.platforms_config = config["platform"]
@@ -38,6 +46,44 @@ class PlatformManager:
        sanitized = platform_id.replace(":", "_").replace("!", "_")
        return sanitized, sanitized != platform_id

+    def _start_platform_task(self, task_name: str, inst: Platform) -> None:
+        run_task = asyncio.create_task(inst.run(), name=task_name)
+        wrapper_task = asyncio.create_task(
+            self._task_wrapper(run_task, platform=inst),
+            name=f"{task_name}_wrapper",
+        )
+        self._platform_tasks[inst.client_self_id] = PlatformTasks(
+            run=run_task,
+            wrapper=wrapper_task,
+        )
+
+    async def _stop_platform_task(self, client_id: str) -> None:
+        tasks = self._platform_tasks.pop(client_id, None)
+        if not tasks:
+            return
+        for task in (tasks.run, tasks.wrapper):
+            if not task.done():
+                task.cancel()
+        await asyncio.gather(tasks.run, tasks.wrapper, return_exceptions=True)
+
+    async def _terminate_inst_and_tasks(self, inst: Platform) -> None:
+        client_id = inst.client_self_id
+        try:
+            if getattr(inst, "terminate", None):
+                try:
+                    await inst.terminate()
+                except asyncio.CancelledError:
+                    raise
+                except Exception as e:
+                    logger.error(
+                        "终止平台适配器失败: client_id=%s, error=%s",
+                        client_id,
+                        e,
+                    )
+                    logger.error(traceback.format_exc())
+        finally:
+            await self._stop_platform_task(client_id)
+
    async def initialize(self) -> None:
        """初始化所有平台适配器"""
        for platform in self.platforms_config:
@@ -51,12 +97,7 @@ class PlatformManager:
        # 网页聊天
        webchat_inst = WebChatAdapter({}, self.settings, self.event_queue)
        self.platform_insts.append(webchat_inst)
-        asyncio.create_task(
-            self._task_wrapper(
-                asyncio.create_task(webchat_inst.run(), name="webchat"),
-                platform=webchat_inst,
-            ),
-        )
+        self._start_platform_task("webchat", webchat_inst)

    async def load_platform(self, platform_config: dict) -> None:
        """实例化一个平台"""
@@ -135,6 +176,10 @@ class PlatformManager:
                    from .sources.satori.satori_adapter import (
                        SatoriPlatformAdapter,  # noqa: F401
                    )
+                case "line":
+                    from .sources.line.line_adapter import (
+                        LinePlatformAdapter,  # noqa: F401
+                    )
        except (ImportError, ModuleNotFoundError) as e:
            logger.error(
                f"加载平台适配器 {platform_config['type']} 失败，原因：{e}。请检查依赖库是否安装。提示：可以在 管理面板->平台日志->安装Pip库 中安装依赖库。",
@@ -154,15 +199,9 @@ class PlatformManager:
            "client_id": inst.client_self_id,
        }
        self.platform_insts.append(inst)
-
-        asyncio.create_task(
-            self._task_wrapper(
-                asyncio.create_task(
-                    inst.run(),
-                    name=f"platform_{platform_config['type']}_{platform_config['id']}",
-                ),
-                platform=inst,
-            ),
+        self._start_platform_task(
+            f"platform_{platform_config['type']}_{platform_config['id']}",
+            inst,
        )
        handlers = star_handlers_registry.get_handlers_by_event_type(
            EventType.OnPlatformLoadedEvent,
@@ -230,13 +269,25 @@ class PlatformManager:
            except Exception:
                logger.warning(f"可能未完全移除 {platform_id} 平台适配器")

-            if getattr(inst, "terminate", None):
-                await inst.terminate()
+            await self._terminate_inst_and_tasks(inst)

    async def terminate(self) -> None:
-        for inst in self.platform_insts:
-            if getattr(inst, "terminate", None):
-                await inst.terminate()
+        terminated_client_ids: set[str] = set()
+        for platform_id in list(self._inst_map.keys()):
+            info = self._inst_map.get(platform_id)
+            if info:
+                terminated_client_ids.add(info["client_id"])
+            await self.terminate_platform(platform_id)
+
+        for inst in list(self.platform_insts):
+            client_id = inst.client_self_id
+            if client_id in terminated_client_ids:
+                continue
+            await self._terminate_inst_and_tasks(inst)
+
+        self.platform_insts.clear()
+        self._inst_map.clear()
+        self._platform_tasks.clear()

    def get_insts(self):
        return self.platform_insts
@@ -24,3 +24,14 @@ class PlatformMetadata:

    module_path: str | None = None
    """注册该适配器的模块路径，用于插件热重载时清理"""
+    i18n_resources: dict[str, dict] | None = None
+    """国际化资源数据，如 {"zh-CN": {...}, "en-US": {...}}
+
+    参考 https://github.com/AstrBotDevs/AstrBot/pull/5045
+    """
+
+    config_metadata: dict | None = None
+    """配置项元数据，用于 WebUI 生成表单。对应 config_metadata.json 的内容
+
+    参考 https://github.com/AstrBotDevs/AstrBot/pull/5045
+    """
@@ -15,11 +15,14 @@ def register_platform_adapter(
    adapter_display_name: str | None = None,
    logo_path: str | None = None,
    support_streaming_message: bool = True,
+    i18n_resources: dict[str, dict] | None = None,
+    config_metadata: dict | None = None,
 ):
    """用于注册平台适配器的带参装饰器。

    default_config_tmpl 指定了平台适配器的默认配置模板。用户填写好后将会作为 platform_config 传入你的 Platform 类的实现类。
    logo_path 指定了平台适配器的 logo 文件路径，是相对于插件目录的路径。
+    config_metadata 指定了配置项的元数据，用于 WebUI 生成表单。如果不指定，WebUI 将会把配置项渲染为原始的键值对编辑框。
    """

    def decorator(cls):
@@ -49,6 +52,8 @@ def register_platform_adapter(
            logo_path=logo_path,
            support_streaming_message=support_streaming_message,
            module_path=module_path,
+            i18n_resources=i18n_resources,
+            config_metadata=config_metadata,
        )
        platform_registry.append(pm)
        platform_cls_map[adapter_name] = cls
@@ -21,7 +21,7 @@ from astrbot.api.platform import (
 )
 from astrbot.core import sp
 from astrbot.core.platform.astr_message_event import MessageSesion
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_file
 from astrbot.core.utils.media_utils import (
    convert_audio_format,
@@ -253,9 +253,9 @@ class DingtalkPlatformAdapter(Platform):
            "downloadCode": download_code,
            "robotCode": robot_code,
        }
-        temp_dir = Path(get_astrbot_data_path()) / "temp"
+        temp_dir = Path(get_astrbot_temp_path())
        temp_dir.mkdir(parents=True, exist_ok=True)
-        f_path = temp_dir / f"dingtalk_file_{uuid.uuid4()}.{ext}"
+        f_path = temp_dir / f"dingtalk_{uuid.uuid4()}.{ext}"
        async with (
            aiohttp.ClientSession() as session,
            session.post(
@@ -3,10 +3,13 @@ import base64
 import json
 import re
 import time
+from pathlib import Path
 from typing import Any, cast
+from uuid import uuid4

 import lark_oapi as lark
 from lark_oapi.api.im.v1 import (
+    GetMessageRequest,
    GetMessageResourceRequest,
 )
 from lark_oapi.api.im.v1.processor import P2ImMessageReceiveV1Processor
@@ -22,6 +25,7 @@ from astrbot.api.platform import (
    PlatformMetadata,
 )
 from astrbot.core.platform.astr_message_event import MessageSesion
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.webhook_utils import log_webhook_info

 from ...register import register_platform_adapter
@@ -91,6 +95,347 @@ class LarkPlatformAdapter(Platform):

        self.event_id_timestamps: dict[str, float] = {}

+    async def _download_message_resource(
+        self,
+        *,
+        message_id: str,
+        file_key: str,
+        resource_type: str,
+    ) -> bytes | None:
+        if self.lark_api.im is None:
+            logger.error("[Lark] API Client im 模块未初始化")
+            return None
+
+        request = (
+            GetMessageResourceRequest.builder()
+            .message_id(message_id)
+            .file_key(file_key)
+            .type(resource_type)
+            .build()
+        )
+        response = await self.lark_api.im.v1.message_resource.aget(request)
+        if not response.success():
+            logger.error(
+                f"[Lark] 下载消息资源失败 type={resource_type}, key={file_key}, "
+                f"code={response.code}, msg={response.msg}",
+            )
+            return None
+
+        if response.file is None:
+            logger.error(f"[Lark] 消息资源响应中不包含文件流: {file_key}")
+            return None
+
+        return response.file.read()
+
+    @staticmethod
+    def _build_message_str_from_components(
+        components: list[Comp.BaseMessageComponent],
+    ) -> str:
+        parts: list[str] = []
+        for comp in components:
+            if isinstance(comp, Comp.Plain):
+                text = comp.text.strip()
+                if text:
+                    parts.append(text)
+            elif isinstance(comp, Comp.At):
+                name = str(comp.name or comp.qq or "").strip()
+                if name:
+                    parts.append(f"@{name}")
+            elif isinstance(comp, Comp.Image):
+                parts.append("[image]")
+            elif isinstance(comp, Comp.File):
+                parts.append(str(comp.name or "[file]"))
+            elif isinstance(comp, Comp.Record):
+                parts.append("[audio]")
+            elif isinstance(comp, Comp.Video):
+                parts.append("[video]")
+
+        return " ".join(parts).strip()
+
+    @staticmethod
+    def _parse_post_content(content: dict[str, Any]) -> list[dict[str, Any]]:
+        result: list[dict[str, Any]] = []
+        for item in content.get("content", []):
+            if isinstance(item, list):
+                for comp in item:
+                    if isinstance(comp, dict):
+                        result.append(comp)
+            elif isinstance(item, dict):
+                result.append(item)
+        return result
+
+    @staticmethod
+    def _build_at_map(mentions: list[Any] | None) -> dict[str, Comp.At]:
+        at_map: dict[str, Comp.At] = {}
+        if not mentions:
+            return at_map
+
+        for mention in mentions:
+            key = getattr(mention, "key", None)
+            if not key:
+                continue
+
+            mention_id = getattr(mention, "id", None)
+            open_id = ""
+            if mention_id is not None:
+                if hasattr(mention_id, "open_id"):
+                    open_id = getattr(mention_id, "open_id", "") or ""
+                else:
+                    open_id = str(mention_id)
+
+            mention_name = str(getattr(mention, "name", "") or "")
+            at_map[key] = Comp.At(qq=open_id, name=mention_name)
+
+        return at_map
+
+    async def _parse_message_components(
+        self,
+        *,
+        message_id: str | None,
+        message_type: str,
+        content: dict[str, Any],
+        at_map: dict[str, Comp.At],
+    ) -> list[Comp.BaseMessageComponent]:
+        components: list[Comp.BaseMessageComponent] = []
+
+        if message_type == "text":
+            message_str_raw = str(content.get("text", ""))
+            at_pattern = r"(@_user_\d+)"
+            parts = re.split(at_pattern, message_str_raw)
+            for part in parts:
+                segment = part.strip()
+                if not segment:
+                    continue
+                if segment in at_map:
+                    components.append(at_map[segment])
+                else:
+                    components.append(Comp.Plain(segment))
+            return components
+
+        if message_type in ("post", "image"):
+            if message_type == "image":
+                comp_list = [
+                    {
+                        "tag": "img",
+                        "image_key": content.get("image_key"),
+                    },
+                ]
+            else:
+                comp_list = self._parse_post_content(content)
+
+            for comp in comp_list:
+                tag = comp.get("tag")
+                if tag == "at":
+                    user_key = str(comp.get("user_id", ""))
+                    if user_key in at_map:
+                        components.append(at_map[user_key])
+                elif tag == "text":
+                    text = str(comp.get("text", "")).strip()
+                    if text:
+                        components.append(Comp.Plain(text))
+                elif tag == "a":
+                    text = str(comp.get("text", "")).strip()
+                    href = str(comp.get("href", "")).strip()
+                    if text and href:
+                        components.append(Comp.Plain(f"{text}({href})"))
+                    elif text:
+                        components.append(Comp.Plain(text))
+                elif tag == "img":
+                    image_key = str(comp.get("image_key", "")).strip()
+                    if not image_key:
+                        continue
+                    if not message_id:
+                        logger.error("[Lark] 图片消息缺少 message_id")
+                        continue
+                    image_bytes = await self._download_message_resource(
+                        message_id=message_id,
+                        file_key=image_key,
+                        resource_type="image",
+                    )
+                    if image_bytes is None:
+                        continue
+                    image_base64 = base64.b64encode(image_bytes).decode()
+                    components.append(Comp.Image.fromBase64(image_base64))
+                elif tag == "media":
+                    file_key = str(comp.get("file_key", "")).strip()
+                    file_name = (
+                        str(comp.get("file_name", "")).strip() or "lark_media.mp4"
+                    )
+                    if not file_key:
+                        continue
+                    if not message_id:
+                        logger.error("[Lark] 富文本视频消息缺少 message_id")
+                        continue
+                    file_path = await self._download_file_resource_to_temp(
+                        message_id=message_id,
+                        file_key=file_key,
+                        message_type="post_media",
+                        file_name=file_name,
+                        default_suffix=".mp4",
+                    )
+                    if file_path:
+                        components.append(Comp.Video(file=file_path, path=file_path))
+
+            return components
+
+        if message_type == "file":
+            file_key = str(content.get("file_key", "")).strip()
+            file_name = str(content.get("file_name", "")).strip() or "lark_file"
+            if not message_id:
+                logger.error("[Lark] 文件消息缺少 message_id")
+                return components
+            if not file_key:
+                logger.error("[Lark] 文件消息缺少 file_key")
+                return components
+            file_path = await self._download_file_resource_to_temp(
+                message_id=message_id,
+                file_key=file_key,
+                message_type="file",
+                file_name=file_name,
+            )
+            if file_path:
+                components.append(Comp.File(name=file_name, file=file_path))
+            return components
+
+        if message_type == "audio":
+            file_key = str(content.get("file_key", "")).strip()
+            if not message_id:
+                logger.error("[Lark] 音频消息缺少 message_id")
+                return components
+            if not file_key:
+                logger.error("[Lark] 音频消息缺少 file_key")
+                return components
+            file_path = await self._download_file_resource_to_temp(
+                message_id=message_id,
+                file_key=file_key,
+                message_type="audio",
+                default_suffix=".opus",
+            )
+            if file_path:
+                components.append(Comp.Record(file=file_path, url=file_path))
+            return components
+
+        if message_type == "media":
+            file_key = str(content.get("file_key", "")).strip()
+            file_name = str(content.get("file_name", "")).strip() or "lark_media.mp4"
+            if not message_id:
+                logger.error("[Lark] 视频消息缺少 message_id")
+                return components
+            if not file_key:
+                logger.error("[Lark] 视频消息缺少 file_key")
+                return components
+            file_path = await self._download_file_resource_to_temp(
+                message_id=message_id,
+                file_key=file_key,
+                message_type="media",
+                file_name=file_name,
+                default_suffix=".mp4",
+            )
+            if file_path:
+                components.append(Comp.Video(file=file_path, path=file_path))
+            return components
+
+        return components
+
+    async def _build_reply_from_parent_id(
+        self,
+        parent_message_id: str,
+    ) -> Comp.Reply | None:
+        if self.lark_api.im is None:
+            logger.error("[Lark] API Client im 模块未初始化")
+            return None
+
+        request = GetMessageRequest.builder().message_id(parent_message_id).build()
+        response = await self.lark_api.im.v1.message.aget(request)
+        if not response.success():
+            logger.error(
+                f"[Lark] 获取引用消息失败 id={parent_message_id}, "
+                f"code={response.code}, msg={response.msg}",
+            )
+            return None
+
+        if response.data is None or not response.data.items:
+            logger.error(
+                f"[Lark] 引用消息响应为空 id={parent_message_id}",
+            )
+            return None
+
+        parent_message = response.data.items[0]
+        quoted_message_id = parent_message.message_id or parent_message_id
+        quoted_sender_id = (
+            parent_message.sender.id
+            if parent_message.sender and parent_message.sender.id
+            else "unknown"
+        )
+        quoted_time_raw = parent_message.create_time or 0
+        quoted_time = (
+            quoted_time_raw // 1000
+            if isinstance(quoted_time_raw, int) and quoted_time_raw > 10**11
+            else quoted_time_raw
+        )
+        quoted_content = (
+            parent_message.body.content if parent_message.body else ""
+        ) or ""
+        quoted_type = parent_message.msg_type or ""
+        quoted_content_json: dict[str, Any] = {}
+        if quoted_content:
+            try:
+                parsed = json.loads(quoted_content)
+                if isinstance(parsed, dict):
+                    quoted_content_json = parsed
+            except json.JSONDecodeError:
+                logger.warning(
+                    f"[Lark] 解析引用消息内容失败 id={quoted_message_id}",
+                )
+
+        quoted_at_map = self._build_at_map(parent_message.mentions)
+        quoted_chain = await self._parse_message_components(
+            message_id=quoted_message_id,
+            message_type=quoted_type,
+            content=quoted_content_json,
+            at_map=quoted_at_map,
+        )
+        quoted_text = self._build_message_str_from_components(quoted_chain)
+        sender_nickname = (
+            quoted_sender_id[:8] if quoted_sender_id != "unknown" else "unknown"
+        )
+
+        return Comp.Reply(
+            id=quoted_message_id,
+            chain=quoted_chain,
+            sender_id=quoted_sender_id,
+            sender_nickname=sender_nickname,
+            time=quoted_time,
+            message_str=quoted_text,
+            text=quoted_text,
+        )
+
+    async def _download_file_resource_to_temp(
+        self,
+        *,
+        message_id: str,
+        file_key: str,
+        message_type: str,
+        file_name: str = "",
+        default_suffix: str = ".bin",
+    ) -> str | None:
+        file_bytes = await self._download_message_resource(
+            message_id=message_id,
+            file_key=file_key,
+            resource_type="file",
+        )
+        if file_bytes is None:
+            return None
+
+        suffix = Path(file_name).suffix if file_name else default_suffix
+        temp_dir = Path(get_astrbot_temp_path())
+        temp_dir.mkdir(parents=True, exist_ok=True)
+        temp_path = (
+            temp_dir / f"lark_{message_type}_{file_name}_{uuid4().hex[:4]}{suffix}"
+        )
+        temp_path.write_bytes(file_bytes)
+        return str(temp_path.resolve())
+
    def _clean_expired_events(self) -> None:
        """清理超过 30 分钟的事件记录"""
        current_time = time.time()
@@ -176,6 +521,11 @@ class LarkPlatformAdapter(Platform):
        abm.message_str = ""

        at_list = {}
+        if message.parent_id:
+            reply_seg = await self._build_reply_from_parent_id(message.parent_id)
+            if reply_seg:
+                abm.message.append(reply_seg)
+
        if message.mentions:
            for m in message.mentions:
                if m.id is None:
@@ -198,80 +548,19 @@ class LarkPlatformAdapter(Platform):
            logger.error(f"[Lark] 解析消息内容失败: {message.content}")
            return

-        if message.message_type == "text":
-            message_str_raw = content_json_b.get("text", "")  # 带有 @ 的消息
-            at_pattern = r"(@_user_\d+)"  # 可以根据需求修改正则
-            # at_users = re.findall(at_pattern, message_str_raw)
-            # 拆分文本，去掉AT符号部分
-            parts = re.split(at_pattern, message_str_raw)
-            for i in range(len(parts)):
-                s = parts[i].strip()
-                if not s:
-                    continue
-                if s in at_list:
-                    abm.message.append(at_list[s])
-                else:
-                    abm.message.append(Comp.Plain(parts[i].strip()))
-        elif message.message_type == "post":
-            _ls = []
+        if not isinstance(content_json_b, dict):
+            logger.error(f"[Lark] 消息内容不是 JSON Object: {message.content}")
+            return

-            content_ls = content_json_b.get("content", [])
-            for comp in content_ls:
-                if isinstance(comp, list):
-                    _ls.extend(comp)
-                elif isinstance(comp, dict):
-                    _ls.append(comp)
-            content_json_b = _ls
-        elif message.message_type == "image":
-            content_json_b = [
-                {
-                    "tag": "img",
-                    "image_key": content_json_b.get("image_key"),
-                    "style": [],
-                },
-            ]
-
-        if message.message_type in ("post", "image"):
-            for comp in content_json_b:
-                if comp.get("tag") == "at":
-                    user_id = comp.get("user_id")
-                    if user_id in at_list:
-                        abm.message.append(at_list[user_id])
-                elif comp.get("tag") == "text" and comp.get("text", "").strip():
-                    abm.message.append(Comp.Plain(comp["text"].strip()))
-                elif comp.get("tag") == "img":
-                    image_key = comp.get("image_key")
-                    if not image_key:
-                        continue
-
-                    request = (
-                        GetMessageResourceRequest.builder()
-                        .message_id(cast(str, message.message_id))
-                        .file_key(image_key)
-                        .type("image")
-                        .build()
-                    )
-
-                    if self.lark_api.im is None:
-                        logger.error("[Lark] API Client im 模块未初始化")
-                        continue
-
-                    response = await self.lark_api.im.v1.message_resource.aget(request)
-                    if not response.success():
-                        logger.error(f"无法下载飞书图片: {image_key}")
-                        continue
-
-                    if response.file is None:
-                        logger.error(f"飞书图片响应中不包含文件流: {image_key}")
-                        continue
-
-                    image_bytes = response.file.read()
-                    image_base64 = base64.b64encode(image_bytes).decode()
-                    abm.message.append(Comp.Image.fromBase64(image_base64))
-
-        for comp in abm.message:
-            if isinstance(comp, Comp.Plain):
-                abm.message_str += comp.text
+        logger.debug(f"[Lark] 解析消息内容: {content_json_b}")
+        parsed_components = await self._parse_message_components(
+            message_id=message.message_id,
+            message_type=message.message_type or "unknown",
+            content=content_json_b,
+            at_map=at_list,
+        )
+        abm.message.extend(parsed_components)
+        abm.message_str = self._build_message_str_from_components(parsed_components)

        if message.message_id is None:
            logger.error("[Lark] 消息缺少 message_id")
@@ -296,7 +585,6 @@ class LarkPlatformAdapter(Platform):
        else:
            abm.session_id = abm.sender.user_id

-        logger.debug(abm)
        await self.handle_msg(abm)

    async def handle_msg(self, abm: AstrBotMessage) -> None:
@@ -21,7 +21,7 @@ from astrbot import logger
 from astrbot.api.event import AstrMessageEvent, MessageChain
 from astrbot.api.message_components import At, File, Plain, Record, Video
 from astrbot.api.message_components import Image as AstrBotImage
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_image_by_url
 from astrbot.core.utils.media_utils import (
    convert_audio_to_opus,
@@ -202,8 +202,11 @@ class LarkMessageEvent(AstrMessageEvent):
                    base64_str = comp.file.removeprefix("base64://")
                    image_data = base64.b64decode(base64_str)
                    # save as temp file
-                    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-                    file_path = os.path.join(temp_dir, f"{uuid.uuid4()}_test.jpg")
+                    temp_dir = get_astrbot_temp_path()
+                    file_path = os.path.join(
+                        temp_dir,
+                        f"lark_image_{uuid.uuid4().hex[:8]}.jpg",
+                    )
                    with open(file_path, "wb") as f:
                        f.write(BytesIO(image_data).getvalue())
                else:
@@ -0,0 +1,474 @@
+import asyncio
+import mimetypes
+import time
+import uuid
+from pathlib import Path
+from typing import Any, cast
+
+from astrbot.api import logger
+from astrbot.api.event import MessageChain
+from astrbot.api.message_components import At, File, Image, Plain, Record, Video
+from astrbot.api.platform import (
+    AstrBotMessage,
+    Group,
+    MessageMember,
+    MessageType,
+    Platform,
+    PlatformMetadata,
+)
+from astrbot.core.platform.astr_message_event import MessageSesion
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
+from astrbot.core.utils.webhook_utils import log_webhook_info
+
+from ...register import register_platform_adapter
+from .line_api import LineAPIClient
+from .line_event import LineMessageEvent
+
+LINE_CONFIG_METADATA = {
+    "channel_access_token": {
+        "description": "LINE Channel Access Token",
+        "type": "string",
+        "hint": "LINE Messaging API 的 channel access token。",
+    },
+    "channel_secret": {
+        "description": "LINE Channel Secret",
+        "type": "string",
+        "hint": "用于校验 LINE Webhook 签名。",
+    },
+}
+
+LINE_I18N_RESOURCES = {
+    "zh-CN": {
+        "channel_access_token": {
+            "description": "LINE Channel Access Token",
+            "hint": "LINE Messaging API 的 channel access token。",
+        },
+        "channel_secret": {
+            "description": "LINE Channel Secret",
+            "hint": "用于校验 LINE Webhook 签名。",
+        },
+    },
+    "en-US": {
+        "channel_access_token": {
+            "description": "LINE Channel Access Token",
+            "hint": "Channel access token for LINE Messaging API.",
+        },
+        "channel_secret": {
+            "description": "LINE Channel Secret",
+            "hint": "Used to verify LINE webhook signatures.",
+        },
+    },
+}
+
+
+@register_platform_adapter(
+    "line",
+    "LINE Messaging API 适配器",
+    support_streaming_message=False,
+    default_config_tmpl={
+        "id": "line",
+        "type": "line",
+        "enable": False,
+        "channel_access_token": "",
+        "channel_secret": "",
+        "unified_webhook_mode": True,
+        "webhook_uuid": "",
+    },
+    config_metadata=LINE_CONFIG_METADATA,
+    i18n_resources=LINE_I18N_RESOURCES,
+)
+class LinePlatformAdapter(Platform):
+    def __init__(
+        self,
+        platform_config: dict,
+        platform_settings: dict,
+        event_queue: asyncio.Queue,
+    ) -> None:
+        super().__init__(platform_config, event_queue)
+        self.config["unified_webhook_mode"] = True
+        self.destination = "unknown"
+        self.settings = platform_settings
+        self._event_id_timestamps: dict[str, float] = {}
+        self.shutdown_event = asyncio.Event()
+
+        channel_access_token = str(platform_config.get("channel_access_token", ""))
+        channel_secret = str(platform_config.get("channel_secret", ""))
+        if not channel_access_token or not channel_secret:
+            raise ValueError(
+                "LINE 适配器需要 channel_access_token 和 channel_secret。",
+            )
+
+        self.line_api = LineAPIClient(
+            channel_access_token=channel_access_token,
+            channel_secret=channel_secret,
+        )
+
+    async def send_by_session(
+        self,
+        session: MessageSesion,
+        message_chain: MessageChain,
+    ) -> None:
+        messages = await LineMessageEvent.build_line_messages(message_chain)
+        if messages:
+            await self.line_api.push_message(session.session_id, messages)
+        await super().send_by_session(session, message_chain)
+
+    def meta(self) -> PlatformMetadata:
+        return PlatformMetadata(
+            name="line",
+            description="LINE Messaging API 适配器",
+            id=cast(str, self.config.get("id", "line")),
+            support_streaming_message=False,
+        )
+
+    async def run(self) -> None:
+        webhook_uuid = self.config.get("webhook_uuid")
+        if webhook_uuid:
+            log_webhook_info(f"{self.meta().id}(LINE)", webhook_uuid)
+        else:
+            logger.warning("[LINE] webhook_uuid 为空，统一 Webhook 可能无法接收消息。")
+        await self.shutdown_event.wait()
+
+    async def terminate(self) -> None:
+        self.shutdown_event.set()
+        await self.line_api.close()
+
+    async def webhook_callback(self, request: Any) -> Any:
+        raw_body = await request.get_data()
+        signature = request.headers.get("x-line-signature")
+        if not self.line_api.verify_signature(raw_body, signature):
+            logger.warning("[LINE] invalid webhook signature")
+            return "invalid signature", 400
+
+        try:
+            payload = await request.get_json(force=True, silent=False)
+        except Exception as e:
+            logger.warning("[LINE] invalid webhook body: %s", e)
+            return "bad request", 400
+
+        if not isinstance(payload, dict):
+            return "bad request", 400
+
+        await self.handle_webhook_event(payload)
+        return "ok", 200
+
+    async def handle_webhook_event(self, payload: dict[str, Any]) -> None:
+        destination = str(payload.get("destination", "")).strip()
+        if destination:
+            self.destination = destination
+
+        events = payload.get("events")
+        if not isinstance(events, list):
+            return
+
+        for event in events:
+            if not isinstance(event, dict):
+                continue
+
+            event_id = str(event.get("webhookEventId", ""))
+            if event_id and self._is_duplicate_event(event_id):
+                logger.debug("[LINE] duplicate event skipped: %s", event_id)
+                continue
+
+            abm = await self.convert_message(event)
+            if abm is None:
+                continue
+            await self.handle_msg(abm)
+
+    async def convert_message(self, event: dict[str, Any]) -> AstrBotMessage | None:
+        if str(event.get("type", "")) != "message":
+            return None
+        if str(event.get("mode", "active")) == "standby":
+            return None
+
+        source = event.get("source", {})
+        if not isinstance(source, dict):
+            return None
+
+        message = event.get("message", {})
+        if not isinstance(message, dict):
+            return None
+
+        source_type = str(source.get("type", ""))
+        user_id = str(source.get("userId", "")).strip()
+        group_id = str(source.get("groupId", "")).strip()
+        room_id = str(source.get("roomId", "")).strip()
+
+        abm = AstrBotMessage()
+        abm.self_id = self.destination or self.meta().id
+        abm.message = []
+        abm.raw_message = event
+        abm.message_id = str(
+            message.get("id")
+            or event.get("webhookEventId")
+            or event.get("deliveryContext", {}).get("deliveryId", "")
+            or uuid.uuid4().hex
+        )
+
+        event_timestamp = event.get("timestamp")
+        if isinstance(event_timestamp, int):
+            abm.timestamp = (
+                event_timestamp // 1000
+                if event_timestamp > 1_000_000_000_000
+                else event_timestamp
+            )
+        else:
+            abm.timestamp = int(time.time())
+
+        if source_type in {"group", "room"}:
+            abm.type = MessageType.GROUP_MESSAGE
+            container_id = group_id or room_id
+            abm.group = Group(group_id=container_id, group_name=container_id)
+            abm.session_id = container_id
+            sender_id = user_id or container_id
+        elif source_type == "user":
+            abm.type = MessageType.FRIEND_MESSAGE
+            abm.session_id = user_id
+            sender_id = user_id
+        else:
+            abm.type = MessageType.OTHER_MESSAGE
+            abm.session_id = user_id or group_id or room_id or "unknown"
+            sender_id = abm.session_id
+
+        abm.sender = MessageMember(user_id=sender_id, nickname=sender_id[:8])
+
+        components = await self._parse_line_message_components(message)
+        if not components:
+            return None
+        abm.message = components
+        abm.message_str = self._build_message_str(components)
+        return abm
+
+    async def _parse_line_message_components(
+        self,
+        message: dict[str, Any],
+    ) -> list:
+        msg_type = str(message.get("type", ""))
+        message_id = str(message.get("id", "")).strip()
+
+        if msg_type == "text":
+            text = str(message.get("text", ""))
+            mention = message.get("mention")
+            if isinstance(mention, dict):
+                return self._parse_text_with_mentions(text, mention)
+            return [Plain(text=text)] if text else []
+
+        if msg_type == "image":
+            image_component = await self._build_image_component(message_id, message)
+            return [image_component] if image_component else [Plain(text="[image]")]
+
+        if msg_type == "video":
+            video_component = await self._build_video_component(message_id, message)
+            return [video_component] if video_component else [Plain(text="[video]")]
+
+        if msg_type == "audio":
+            audio_component = await self._build_audio_component(message_id, message)
+            return [audio_component] if audio_component else [Plain(text="[audio]")]
+
+        if msg_type == "file":
+            file_component = await self._build_file_component(message_id, message)
+            return [file_component] if file_component else [Plain(text="[file]")]
+
+        if msg_type == "sticker":
+            return [Plain(text="[sticker]")]
+
+        return [Plain(text=f"[{msg_type}]")]
+
+    def _parse_text_with_mentions(self, text: str, mention_obj: dict[str, Any]) -> list:
+        mentions = mention_obj.get("mentionees", [])
+        if not isinstance(mentions, list) or not mentions:
+            return [Plain(text=text)] if text else []
+
+        normalized = []
+        for item in mentions:
+            if not isinstance(item, dict):
+                continue
+            start = item.get("index")
+            length = item.get("length")
+            if not isinstance(start, int) or not isinstance(length, int):
+                continue
+            normalized.append((start, length, item))
+        normalized.sort(key=lambda x: x[0])
+
+        ret = []
+        cursor = 0
+        for start, length, item in normalized:
+            if start > cursor:
+                part = text[cursor:start]
+                if part:
+                    ret.append(Plain(text=part))
+
+            label = text[start : start + length] or "@user"
+            mention_type = str(item.get("type", ""))
+            if mention_type == "user":
+                target_id = str(item.get("userId", "")).strip()
+                ret.append(At(qq=target_id, name=label.lstrip("@")))
+            else:
+                ret.append(Plain(text=label))
+            cursor = max(cursor, start + length)
+
+        if cursor < len(text):
+            tail = text[cursor:]
+            if tail:
+                ret.append(Plain(text=tail))
+        return ret
+
+    async def _build_image_component(
+        self,
+        message_id: str,
+        message: dict[str, Any],
+    ) -> Image | None:
+        external_url = self._get_external_content_url(message)
+        if external_url:
+            return Image.fromURL(external_url)
+
+        content = await self.line_api.get_message_content(message_id)
+        if not content:
+            return None
+        content_bytes, _, _ = content
+        return Image.fromBytes(content_bytes)
+
+    async def _build_video_component(
+        self,
+        message_id: str,
+        message: dict[str, Any],
+    ) -> Video | None:
+        external_url = self._get_external_content_url(message)
+        if external_url:
+            return Video.fromURL(external_url)
+
+        content = await self.line_api.get_message_content(message_id)
+        if not content:
+            return None
+        content_bytes, content_type, _ = content
+        suffix = self._guess_suffix(content_type, ".mp4")
+        file_path = self._store_temp_content("video", message_id, content_bytes, suffix)
+        return Video(file=file_path, path=file_path)
+
+    async def _build_audio_component(
+        self,
+        message_id: str,
+        message: dict[str, Any],
+    ) -> Record | None:
+        external_url = self._get_external_content_url(message)
+        if external_url:
+            return Record.fromURL(external_url)
+
+        content = await self.line_api.get_message_content(message_id)
+        if not content:
+            return None
+        content_bytes, content_type, _ = content
+        suffix = self._guess_suffix(content_type, ".m4a")
+        file_path = self._store_temp_content("audio", message_id, content_bytes, suffix)
+        return Record(file=file_path, url=file_path)
+
+    async def _build_file_component(
+        self,
+        message_id: str,
+        message: dict[str, Any],
+    ) -> File | None:
+        content = await self.line_api.get_message_content(message_id)
+        if not content:
+            return None
+        content_bytes, content_type, filename = content
+        default_name = str(message.get("fileName", "")).strip() or f"{message_id}.bin"
+        suffix = Path(default_name).suffix or self._guess_suffix(content_type, ".bin")
+        final_name = filename or default_name
+        file_path = self._store_temp_content(
+            "file",
+            message_id,
+            content_bytes,
+            suffix,
+            original_name=final_name,
+        )
+        return File(name=final_name, file=file_path, url=file_path)
+
+    @staticmethod
+    def _get_external_content_url(message: dict[str, Any]) -> str:
+        provider = message.get("contentProvider")
+        if not isinstance(provider, dict):
+            return ""
+        if str(provider.get("type", "")) != "external":
+            return ""
+        return str(provider.get("originalContentUrl", "")).strip()
+
+    @staticmethod
+    def _guess_suffix(content_type: str | None, fallback: str) -> str:
+        if not content_type:
+            return fallback
+        base_type = content_type.split(";", 1)[0].strip().lower()
+        guessed = mimetypes.guess_extension(base_type)
+        if guessed:
+            return guessed
+        return fallback
+
+    @staticmethod
+    def _store_temp_content(
+        content_type: str,
+        message_id: str,
+        content: bytes,
+        suffix: str,
+        original_name: str = "",
+    ) -> str:
+        temp_dir = Path(get_astrbot_temp_path())
+        temp_dir.mkdir(parents=True, exist_ok=True)
+        name_prefix = f"line_{content_type}"
+        if original_name:
+            safe_stem = Path(original_name).stem.strip()
+            safe_stem = "".join(
+                ch if ch.isalnum() or ch in ("-", "_", ".") else "_" for ch in safe_stem
+            )
+            safe_stem = safe_stem.strip("._")
+            if safe_stem:
+                name_prefix = safe_stem[:64]
+        file_path = temp_dir / f"{name_prefix}_{message_id}_{uuid.uuid4().hex[:6]}"
+        file_path = file_path.with_suffix(suffix)
+        file_path.write_bytes(content)
+        return str(file_path.resolve())
+
+    @staticmethod
+    def _build_message_str(components: list) -> str:
+        parts: list[str] = []
+        for comp in components:
+            if isinstance(comp, Plain):
+                parts.append(comp.text)
+            elif isinstance(comp, At):
+                parts.append(f"@{comp.name or comp.qq}")
+            elif isinstance(comp, Image):
+                parts.append("[image]")
+            elif isinstance(comp, Video):
+                parts.append("[video]")
+            elif isinstance(comp, Record):
+                parts.append("[audio]")
+            elif isinstance(comp, File):
+                parts.append(str(comp.name or "[file]"))
+            else:
+                parts.append(f"[{comp.type}]")
+        return " ".join(i for i in parts if i).strip()
+
+    def _clean_expired_events(self) -> None:
+        current = time.time()
+        expired = [
+            event_id
+            for event_id, ts in self._event_id_timestamps.items()
+            if current - ts > 1800
+        ]
+        for event_id in expired:
+            del self._event_id_timestamps[event_id]
+
+    def _is_duplicate_event(self, event_id: str) -> bool:
+        self._clean_expired_events()
+        if event_id in self._event_id_timestamps:
+            return True
+        self._event_id_timestamps[event_id] = time.time()
+        return False
+
+    async def handle_msg(self, abm: AstrBotMessage) -> None:
+        event = LineMessageEvent(
+            message_str=abm.message_str,
+            message_obj=abm,
+            platform_meta=self.meta(),
+            session_id=abm.session_id,
+            line_api=self.line_api,
+        )
+        self._event_queue.put_nowait(event)
@@ -0,0 +1,203 @@
+import asyncio
+import base64
+import hmac
+import json
+from hashlib import sha256
+from typing import Any
+from urllib.parse import unquote
+
+import aiohttp
+
+from astrbot.api import logger
+
+
+class LineAPIClient:
+    def __init__(
+        self,
+        *,
+        channel_access_token: str,
+        channel_secret: str,
+        timeout_seconds: int = 30,
+    ) -> None:
+        self.channel_access_token = channel_access_token.strip()
+        self.channel_secret = channel_secret.strip()
+        self.timeout = aiohttp.ClientTimeout(total=timeout_seconds)
+        self._session: aiohttp.ClientSession | None = None
+
+    async def _get_session(self) -> aiohttp.ClientSession:
+        if self._session is None or self._session.closed:
+            self._session = aiohttp.ClientSession(timeout=self.timeout)
+        return self._session
+
+    async def close(self) -> None:
+        if self._session and not self._session.closed:
+            await self._session.close()
+
+    def verify_signature(self, raw_body: bytes, signature: str | None) -> bool:
+        if not signature:
+            return False
+        digest = hmac.new(
+            self.channel_secret.encode("utf-8"),
+            raw_body,
+            sha256,
+        ).digest()
+        expected = base64.b64encode(digest).decode("utf-8")
+        return hmac.compare_digest(expected, signature.strip())
+
+    @property
+    def _auth_headers(self) -> dict[str, str]:
+        return {"Authorization": f"Bearer {self.channel_access_token}"}
+
+    async def reply_message(
+        self,
+        reply_token: str,
+        messages: list[dict[str, Any]],
+        *,
+        notification_disabled: bool = False,
+    ) -> bool:
+        payload = {
+            "replyToken": reply_token,
+            "messages": messages[:5],
+            "notificationDisabled": notification_disabled,
+        }
+        return await self._post_json(
+            "https://api.line.me/v2/bot/message/reply",
+            payload=payload,
+            op_name="reply",
+        )
+
+    async def push_message(
+        self,
+        to: str,
+        messages: list[dict[str, Any]],
+        *,
+        notification_disabled: bool = False,
+    ) -> bool:
+        payload = {
+            "to": to,
+            "messages": messages[:5],
+            "notificationDisabled": notification_disabled,
+        }
+        return await self._post_json(
+            "https://api.line.me/v2/bot/message/push",
+            payload=payload,
+            op_name="push",
+        )
+
+    async def _post_json(
+        self,
+        url: str,
+        *,
+        payload: dict[str, Any],
+        op_name: str,
+    ) -> bool:
+        session = await self._get_session()
+        headers = {
+            **self._auth_headers,
+            "Content-Type": "application/json",
+        }
+        try:
+            async with session.post(url, json=payload, headers=headers) as resp:
+                if resp.status < 400:
+                    return True
+                body = await resp.text()
+                logger.error(
+                    "[LINE] %s message failed: status=%s body=%s",
+                    op_name,
+                    resp.status,
+                    body,
+                )
+                return False
+        except Exception as e:
+            logger.error("[LINE] %s message request failed: %s", op_name, e)
+            return False
+
+    async def get_message_content(
+        self,
+        message_id: str,
+    ) -> tuple[bytes, str | None, str | None] | None:
+        session = await self._get_session()
+        url = f"https://api-data.line.me/v2/bot/message/{message_id}/content"
+        headers = self._auth_headers
+
+        async with session.get(url, headers=headers) as resp:
+            if resp.status == 202:
+                if not await self._wait_for_transcoding(message_id):
+                    return None
+                async with session.get(url, headers=headers) as retry_resp:
+                    if retry_resp.status != 200:
+                        body = await retry_resp.text()
+                        logger.warning(
+                            "[LINE] get content retry failed: message_id=%s status=%s body=%s",
+                            message_id,
+                            retry_resp.status,
+                            body,
+                        )
+                        return None
+                    return await self._read_content_response(retry_resp)
+
+            if resp.status != 200:
+                body = await resp.text()
+                logger.warning(
+                    "[LINE] get content failed: message_id=%s status=%s body=%s",
+                    message_id,
+                    resp.status,
+                    body,
+                )
+                return None
+            return await self._read_content_response(resp)
+
+    async def _read_content_response(
+        self,
+        resp: aiohttp.ClientResponse,
+    ) -> tuple[bytes, str | None, str | None]:
+        content = await resp.read()
+        content_type = resp.headers.get("Content-Type")
+        disposition = resp.headers.get("Content-Disposition")
+        filename = self._extract_filename_from_disposition(disposition)
+        return content, content_type, filename
+
+    def _extract_filename_from_disposition(self, disposition: str | None) -> str | None:
+        if not disposition:
+            return None
+        for part in disposition.split(";"):
+            token = part.strip()
+            if token.startswith("filename*="):
+                val = token.split("=", 1)[1].strip().strip('"')
+                if val.lower().startswith("utf-8''"):
+                    val = val[7:]
+                return unquote(val)
+            if token.startswith("filename="):
+                return token.split("=", 1)[1].strip().strip('"')
+        return None
+
+    async def _wait_for_transcoding(
+        self,
+        message_id: str,
+        *,
+        max_attempts: int = 10,
+        interval_seconds: float = 1.0,
+    ) -> bool:
+        session = await self._get_session()
+        url = (
+            f"https://api-data.line.me/v2/bot/message/{message_id}/content/transcoding"
+        )
+        headers = self._auth_headers
+
+        for _ in range(max_attempts):
+            try:
+                async with session.get(url, headers=headers) as resp:
+                    if resp.status != 200:
+                        await asyncio.sleep(interval_seconds)
+                        continue
+                    body = await resp.text()
+                    data = json.loads(body)
+                    status = str(data.get("status", "")).lower()
+                    if status == "succeeded":
+                        return True
+                    if status == "failed":
+                        return False
+            except Exception:
+                pass
+            await asyncio.sleep(interval_seconds)
+        return False
@@ -0,0 +1,285 @@
+import asyncio
+import os
+import re
+import uuid
+from collections.abc import AsyncGenerator
+from pathlib import Path
+
+from astrbot.api import logger
+from astrbot.api.event import AstrMessageEvent, MessageChain
+from astrbot.api.message_components import (
+    At,
+    BaseMessageComponent,
+    File,
+    Image,
+    Plain,
+    Record,
+    Video,
+)
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
+from astrbot.core.utils.media_utils import get_media_duration
+
+from .line_api import LineAPIClient
+
+
+class LineMessageEvent(AstrMessageEvent):
+    def __init__(
+        self,
+        message_str,
+        message_obj,
+        platform_meta,
+        session_id,
+        line_api: LineAPIClient,
+    ) -> None:
+        super().__init__(message_str, message_obj, platform_meta, session_id)
+        self.line_api = line_api
+
+    @staticmethod
+    async def _component_to_message_object(
+        segment: BaseMessageComponent,
+    ) -> dict | None:
+        if isinstance(segment, Plain):
+            text = segment.text.strip()
+            if not text:
+                return None
+            return {"type": "text", "text": text[:5000]}
+
+        if isinstance(segment, At):
+            name = str(segment.name or segment.qq or "").strip()
+            if not name:
+                return None
+            return {"type": "text", "text": f"@{name}"[:5000]}
+
+        if isinstance(segment, Image):
+            image_url = await LineMessageEvent._resolve_image_url(segment)
+            if not image_url:
+                return None
+            return {
+                "type": "image",
+                "originalContentUrl": image_url,
+                "previewImageUrl": image_url,
+            }
+
+        if isinstance(segment, Record):
+            audio_url = await LineMessageEvent._resolve_record_url(segment)
+            if not audio_url:
+                return None
+            duration = await LineMessageEvent._resolve_record_duration(segment)
+            return {
+                "type": "audio",
+                "originalContentUrl": audio_url,
+                "duration": duration,
+            }
+
+        if isinstance(segment, Video):
+            video_url = await LineMessageEvent._resolve_video_url(segment)
+            if not video_url:
+                return None
+            preview_url = await LineMessageEvent._resolve_video_preview_url(segment)
+            if not preview_url:
+                return None
+            return {
+                "type": "video",
+                "originalContentUrl": video_url,
+                "previewImageUrl": preview_url,
+            }
+
+        if isinstance(segment, File):
+            file_url = await LineMessageEvent._resolve_file_url(segment)
+            if not file_url:
+                return None
+            file_name = str(segment.name or "").strip() or "file.bin"
+            file_size = await LineMessageEvent._resolve_file_size(segment)
+            if file_size <= 0:
+                return None
+            return {
+                "type": "file",
+                "fileName": file_name,
+                "fileSize": file_size,
+                "originalContentUrl": file_url,
+            }
+
+        return None
+
+    @staticmethod
+    async def _resolve_image_url(segment: Image) -> str:
+        candidate = (segment.url or segment.file or "").strip()
+        if candidate.startswith("http://") or candidate.startswith("https://"):
+            return candidate
+        try:
+            return await segment.register_to_file_service()
+        except Exception as e:
+            logger.debug("[LINE] resolve image url failed: %s", e)
+            return ""
+
+    @staticmethod
+    async def _resolve_record_url(segment: Record) -> str:
+        candidate = (segment.url or segment.file or "").strip()
+        if candidate.startswith("http://") or candidate.startswith("https://"):
+            return candidate
+        try:
+            return await segment.register_to_file_service()
+        except Exception as e:
+            logger.debug("[LINE] resolve record url failed: %s", e)
+            return ""
+
+    @staticmethod
+    async def _resolve_record_duration(segment: Record) -> int:
+        try:
+            file_path = await segment.convert_to_file_path()
+            duration_ms = await get_media_duration(file_path)
+            if isinstance(duration_ms, int) and duration_ms > 0:
+                return duration_ms
+        except Exception as e:
+            logger.debug("[LINE] resolve record duration failed: %s", e)
+        return 1000
+
+    @staticmethod
+    async def _resolve_video_url(segment: Video) -> str:
+        candidate = (segment.file or "").strip()
+        if candidate.startswith("http://") or candidate.startswith("https://"):
+            return candidate
+        try:
+            return await segment.register_to_file_service()
+        except Exception as e:
+            logger.debug("[LINE] resolve video url failed: %s", e)
+            return ""
+
+    @staticmethod
+    async def _resolve_video_preview_url(segment: Video) -> str:
+        cover_candidate = (segment.cover or "").strip()
+        if cover_candidate.startswith("http://") or cover_candidate.startswith(
+            "https://"
+        ):
+            return cover_candidate
+
+        if cover_candidate:
+            try:
+                cover_seg = Image(file=cover_candidate)
+                return await cover_seg.register_to_file_service()
+            except Exception as e:
+                logger.debug("[LINE] resolve video cover failed: %s", e)
+
+        try:
+            video_path = await segment.convert_to_file_path()
+            temp_dir = Path(get_astrbot_temp_path())
+            temp_dir.mkdir(parents=True, exist_ok=True)
+            thumb_path = temp_dir / f"line_video_preview_{uuid.uuid4().hex}.jpg"
+
+            process = await asyncio.create_subprocess_exec(
+                "ffmpeg",
+                "-y",
+                "-ss",
+                "00:00:01",
+                "-i",
+                video_path,
+                "-frames:v",
+                "1",
+                str(thumb_path),
+                stdout=asyncio.subprocess.PIPE,
+                stderr=asyncio.subprocess.PIPE,
+            )
+            await process.communicate()
+            if process.returncode != 0 or not thumb_path.exists():
+                return ""
+
+            cover_seg = Image.fromFileSystem(str(thumb_path))
+            return await cover_seg.register_to_file_service()
+        except Exception as e:
+            logger.debug("[LINE] generate video preview failed: %s", e)
+            return ""
+
+    @staticmethod
+    async def _resolve_file_url(segment: File) -> str:
+        if segment.url and segment.url.startswith(("http://", "https://")):
+            return segment.url
+        try:
+            return await segment.register_to_file_service()
+        except Exception as e:
+            logger.debug("[LINE] resolve file url failed: %s", e)
+            return ""
+
+    @staticmethod
+    async def _resolve_file_size(segment: File) -> int:
+        try:
+            file_path = await segment.get_file(allow_return_url=False)
+            if file_path and os.path.exists(file_path):
+                return int(os.path.getsize(file_path))
+        except Exception as e:
+            logger.debug("[LINE] resolve file size failed: %s", e)
+        return 0
+
+    @classmethod
+    async def build_line_messages(cls, message_chain: MessageChain) -> list[dict]:
+        messages: list[dict] = []
+        for segment in message_chain.chain:
+            obj = await cls._component_to_message_object(segment)
+            if obj:
+                messages.append(obj)
+
+        if not messages:
+            return []
+
+        if len(messages) > 5:
+            logger.warning(
+                "[LINE] message count exceeds 5, extra segments will be dropped."
+            )
+            messages = messages[:5]
+        return messages
+
+    async def send(self, message: MessageChain) -> None:
+        messages = await self.build_line_messages(message)
+        if not messages:
+            return
+
+        raw = self.message_obj.raw_message
+        reply_token = ""
+        if isinstance(raw, dict):
+            reply_token = str(raw.get("replyToken") or "")
+
+        sent = False
+        if reply_token:
+            sent = await self.line_api.reply_message(reply_token, messages)
+
+        if not sent:
+            target_id = self.get_group_id() or self.get_sender_id()
+            if target_id:
+                await self.line_api.push_message(target_id, messages)
+
+        await super().send(message)
+
+    async def send_streaming(
+        self,
+        generator: AsyncGenerator,
+        use_fallback: bool = False,
+    ):
+        if not use_fallback:
+            buffer = None
+            async for chain in generator:
+                if not buffer:
+                    buffer = chain
+                else:
+                    buffer.chain.extend(chain.chain)
+            if not buffer:
+                return None
+            buffer.squash_plain()
+            await self.send(buffer)
+            return await super().send_streaming(generator, use_fallback)
+
+        buffer = ""
+        pattern = re.compile(r"[^。？！~…]+[。？！~…]+")
+
+        async for chain in generator:
+            if isinstance(chain, MessageChain):
+                for comp in chain.chain:
+                    if isinstance(comp, Plain):
+                        buffer += comp.text
+                        if any(p in buffer for p in "。？！~…"):
+                            buffer = await self.process_buffer(buffer, pattern)
+                    else:
+                        await self.send(MessageChain(chain=[comp]))
+                        await asyncio.sleep(1.5)
+
+        if buffer.strip():
+            await self.send(MessageChain([Plain(buffer)]))
+        return await super().send_streaming(generator, use_fallback)
@@ -21,7 +21,7 @@ try:
 except Exception:
    magic = None

-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from .misskey_event import MisskeyPlatformEvent
 from .misskey_utils import (
@@ -498,7 +498,7 @@ class MisskeyPlatformAdapter(Platform):
                finally:
                    # 清理临时文件
                    if local_path and isinstance(local_path, str):
-                        data_temp = os.path.join(get_astrbot_data_path(), "temp")
+                        data_temp = get_astrbot_temp_path()
                        if local_path.startswith(data_temp) and os.path.exists(
                            local_path,
                        ):
@@ -19,7 +19,7 @@ from astrbot.api import logger
 from astrbot.api.event import AstrMessageEvent, MessageChain
 from astrbot.api.message_components import Image, Plain, Record
 from astrbot.api.platform import AstrBotMessage, PlatformMetadata
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_image_by_url, file_to_base64
 from astrbot.core.utils.tencent_record_helper import wav_to_tencent_silk

@@ -350,10 +350,10 @@ class QQOfficialMessageEvent(AstrMessageEvent):
            elif isinstance(i, Record):
                if i.file:
                    record_wav_path = await i.convert_to_file_path()  # wav 路径
-                    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+                    temp_dir = get_astrbot_temp_path()
                    record_tecent_silk_path = os.path.join(
                        temp_dir,
-                        f"{uuid.uuid4()}.silk",
+                        f"qqofficial_{uuid.uuid4()}.silk",
                    )
                    try:
                        duration = await wav_to_tencent_silk(
@@ -8,13 +8,11 @@ from typing import cast

 import botpy
 import botpy.message
-import botpy.types
-import botpy.types.message
 from botpy import Client

 from astrbot import logger
 from astrbot.api.event import MessageChain
-from astrbot.api.message_components import At, Image, Plain
+from astrbot.api.message_components import At, File, Image, Plain
 from astrbot.api.platform import (
    AstrBotMessage,
    MessageMember,
@@ -143,6 +141,41 @@ class QQOfficialPlatformAdapter(Platform):
            support_proactive_message=False,
        )

+    @staticmethod
+    def _normalize_attachment_url(url: str | None) -> str:
+        if not url:
+            return ""
+        if url.startswith("http://") or url.startswith("https://"):
+            return url
+        return f"https://{url}"
+
+    @staticmethod
+    def _append_attachments(
+        msg: list[BaseMessageComponent],
+        attachments: list | None,
+    ) -> None:
+        if not attachments:
+            return
+
+        for attachment in attachments:
+            content_type = cast(str, getattr(attachment, "content_type", "") or "")
+            url = QQOfficialPlatformAdapter._normalize_attachment_url(
+                cast(str | None, getattr(attachment, "url", None))
+            )
+            if not url:
+                continue
+
+            if content_type.startswith("image"):
+                msg.append(Image.fromURL(url))
+            else:
+                filename = cast(
+                    str,
+                    getattr(attachment, "filename", None)
+                    or getattr(attachment, "name", None)
+                    or "attachment",
+                )
+                msg.append(File(name=filename, file=url, url=url))
+
    @staticmethod
    def _parse_from_qqofficial(
        message: botpy.message.Message
@@ -172,14 +205,7 @@ class QQOfficialPlatformAdapter(Platform):
            abm.self_id = "unknown_selfid"
            msg.append(At(qq="qq_official"))
            msg.append(Plain(abm.message_str))
-            if message.attachments:
-                for i in message.attachments:
-                    if i.content_type.startswith("image"):
-                        url = i.url
-                        if not url.startswith("http"):
-                            url = "https://" + url
-                        img = Image.fromURL(url)
-                        msg.append(img)
+            QQOfficialPlatformAdapter._append_attachments(msg, message.attachments)
            abm.message = msg

        elif isinstance(message, botpy.message.Message) or isinstance(
@@ -196,14 +222,7 @@ class QQOfficialPlatformAdapter(Platform):
                "",
            ).strip()

-            if message.attachments:
-                for i in message.attachments:
-                    if i.content_type.startswith("image"):
-                        url = i.url
-                        if not url.startswith("http"):
-                            url = "https://" + url
-                        img = Image.fromURL(url)
-                        msg.append(img)
+            QQOfficialPlatformAdapter._append_attachments(msg, message.attachments)
            abm.message = msg
            abm.message_str = plain_content
            abm.sender = MessageMember(
@@ -1,11 +1,11 @@
 import asyncio
 import logging
+import random
+from types import SimpleNamespace
 from typing import Any, cast

 import botpy
 import botpy.message
-import botpy.types
-import botpy.types.message
 from botpy import Client

 from astrbot import logger
@@ -15,6 +15,7 @@ from astrbot.core.platform.astr_message_event import MessageSesion
 from astrbot.core.utils.webhook_utils import log_webhook_info

 from ...register import register_platform_adapter
+from ..qqofficial.qqofficial_message_event import QQOfficialMessageEvent
 from ..qqofficial.qqofficial_platform_adapter import QQOfficialPlatformAdapter
 from .qo_webhook_event import QQOfficialWebhookMessageEvent
 from .qo_webhook_server import QQOfficialWebhook
@@ -39,6 +40,7 @@ class botClient(Client):
        )
        abm.group_id = cast(str, message.group_openid)
        abm.session_id = abm.group_id
+        self.platform.remember_session_scene(abm.session_id, "group")
        self._commit(abm)

    # 收到频道消息
@@ -49,6 +51,7 @@ class botClient(Client):
        )
        abm.group_id = message.channel_id
        abm.session_id = abm.group_id
+        self.platform.remember_session_scene(abm.session_id, "channel")
        self._commit(abm)

    # 收到私聊消息
@@ -60,6 +63,7 @@ class botClient(Client):
            MessageType.FRIEND_MESSAGE,
        )
        abm.session_id = abm.sender.user_id
+        self.platform.remember_session_scene(abm.session_id, "friend")
        self._commit(abm)

    # 收到 C2C 消息
@@ -69,9 +73,11 @@ class botClient(Client):
            MessageType.FRIEND_MESSAGE,
        )
        abm.session_id = abm.sender.user_id
+        self.platform.remember_session_scene(abm.session_id, "friend")
        self._commit(abm)

    def _commit(self, abm: AstrBotMessage) -> None:
+        self.platform.remember_session_message_id(abm.session_id, abm.message_id)
        self.platform.commit_event(
            QQOfficialWebhookMessageEvent(
                abm.message_str,
@@ -109,20 +115,129 @@ class QQOfficialWebhookPlatformAdapter(Platform):
        )
        self.client.set_platform(self)
        self.webhook_helper = None
+        self._session_last_message_id: dict[str, str] = {}
+        self._session_scene: dict[str, str] = {}

    async def send_by_session(
        self,
        session: MessageSesion,
        message_chain: MessageChain,
    ) -> None:
-        raise NotImplementedError("QQ 机器人官方 API 适配器不支持 send_by_session")
+        (
+            plain_text,
+            image_base64,
+            image_path,
+            record_file_path,
+        ) = await QQOfficialMessageEvent._parse_to_qqofficial(message_chain)
+        if not plain_text and not image_path:
+            return
+
+        msg_id = self._session_last_message_id.get(session.session_id)
+        if not msg_id:
+            logger.warning(
+                "[QQOfficialWebhook] No cached msg_id for session: %s, skip send_by_session",
+                session.session_id,
+            )
+            return
+
+        payload: dict[str, Any] = {"content": plain_text, "msg_id": msg_id}
+        ret: Any = None
+        send_helper = SimpleNamespace(bot=self.client)
+        if session.message_type == MessageType.GROUP_MESSAGE:
+            scene = self._session_scene.get(session.session_id)
+            if scene == "group":
+                payload["msg_seq"] = random.randint(1, 10000)
+                if image_base64:
+                    media = await QQOfficialMessageEvent.upload_group_and_c2c_image(
+                        send_helper,  # type: ignore
+                        image_base64,
+                        1,
+                        group_openid=session.session_id,
+                    )
+                    payload["media"] = media
+                    payload["msg_type"] = 7
+                if record_file_path:
+                    media = await QQOfficialMessageEvent.upload_group_and_c2c_record(
+                        send_helper,  # type: ignore
+                        record_file_path,
+                        3,
+                        group_openid=session.session_id,
+                    )
+                    payload["media"] = media
+                    payload["msg_type"] = 7
+                ret = await self.client.api.post_group_message(
+                    group_openid=session.session_id,
+                    **payload,
+                )
+            else:
+                if image_path:
+                    payload["file_image"] = image_path
+                ret = await self.client.api.post_message(
+                    channel_id=session.session_id,
+                    **payload,
+                )
+        elif session.message_type == MessageType.FRIEND_MESSAGE:
+            payload["msg_seq"] = random.randint(1, 10000)
+            if image_base64:
+                media = await QQOfficialMessageEvent.upload_group_and_c2c_image(
+                    send_helper,  # type: ignore
+                    image_base64,
+                    1,
+                    openid=session.session_id,
+                )
+                payload["media"] = media
+                payload["msg_type"] = 7
+            if record_file_path:
+                media = await QQOfficialMessageEvent.upload_group_and_c2c_record(
+                    send_helper,  # type: ignore
+                    record_file_path,
+                    3,
+                    openid=session.session_id,
+                )
+                payload["media"] = media
+                payload["msg_type"] = 7
+            ret = await QQOfficialMessageEvent.post_c2c_message(
+                send_helper,  # type: ignore
+                openid=session.session_id,
+                **payload,
+            )
+        else:
+            logger.warning(
+                "[QQOfficialWebhook] Unsupported message type for send_by_session: %s",
+                session.message_type,
+            )
+            return
+
+        sent_message_id = self._extract_message_id(ret)
+        if sent_message_id:
+            self.remember_session_message_id(session.session_id, sent_message_id)
+        await super().send_by_session(session, message_chain)
+
+    def remember_session_message_id(self, session_id: str, message_id: str) -> None:
+        if not session_id or not message_id:
+            return
+        self._session_last_message_id[session_id] = message_id
+
+    def remember_session_scene(self, session_id: str, scene: str) -> None:
+        if not session_id or not scene:
+            return
+        self._session_scene[session_id] = scene
+
+    def _extract_message_id(self, ret: Any) -> str | None:
+        if isinstance(ret, dict):
+            message_id = ret.get("id")
+            return str(message_id) if message_id else None
+        message_id = getattr(ret, "id", None)
+        if message_id:
+            return str(message_id)
+        return None

    def meta(self) -> PlatformMetadata:
        return PlatformMetadata(
            name="qq_official_webhook",
            description="QQ 机器人官方 API 适配器",
            id=cast(str, self.config.get("id")),
-            support_proactive_message=False,
+            support_proactive_message=True,
        )

    async def run(self) -> None:
@@ -5,6 +5,7 @@ from typing import Any, cast

 import telegramify_markdown
 from telegram import ReactionTypeCustomEmoji, ReactionTypeEmoji
+from telegram.constants import ChatAction
 from telegram.ext import ExtBot

 from astrbot import logger
@@ -31,6 +32,14 @@ class TelegramPlatformEvent(AstrMessageEvent):
        "word": re.compile(r"\s"),
    }

+    # 消息类型到 chat action 的映射，用于优先级判断
+    ACTION_BY_TYPE: dict[type, str] = {
+        Record: ChatAction.UPLOAD_VOICE,
+        File: ChatAction.UPLOAD_DOCUMENT,
+        Image: ChatAction.UPLOAD_PHOTO,
+        Plain: ChatAction.TYPING,
+    }
+
    def __init__(
        self,
        message_str: str,
@@ -67,6 +76,71 @@ class TelegramPlatformEvent(AstrMessageEvent):

        return chunks

+    @classmethod
+    async def _send_chat_action(
+        cls,
+        client: ExtBot,
+        chat_id: str,
+        action: ChatAction | str,
+        message_thread_id: str | None = None,
+    ) -> None:
+        """发送聊天状态动作"""
+        try:
+            payload: dict[str, Any] = {"chat_id": chat_id, "action": action}
+            if message_thread_id:
+                payload["message_thread_id"] = message_thread_id
+            await client.send_chat_action(**payload)
+        except Exception as e:
+            logger.warning(f"[Telegram] 发送 chat action 失败: {e}")
+
+    @classmethod
+    def _get_chat_action_for_chain(cls, chain: list[Any]) -> ChatAction | str:
+        """根据消息链中的组件类型确定合适的 chat action（按优先级）"""
+        for seg_type, action in cls.ACTION_BY_TYPE.items():
+            if any(isinstance(seg, seg_type) for seg in chain):
+                return action
+        return ChatAction.TYPING
+
+    @classmethod
+    async def _send_media_with_action(
+        cls,
+        client: ExtBot,
+        upload_action: ChatAction | str,
+        send_coro,
+        *,
+        user_name: str,
+        message_thread_id: str | None = None,
+        **payload: Any,
+    ) -> None:
+        """发送媒体时显示 upload action，发送完成后恢复 typing"""
+        await cls._send_chat_action(client, user_name, upload_action, message_thread_id)
+        await send_coro(**payload)
+        await cls._send_chat_action(
+            client, user_name, ChatAction.TYPING, message_thread_id
+        )
+
+    async def _ensure_typing(
+        self,
+        user_name: str,
+        message_thread_id: str | None = None,
+    ) -> None:
+        """确保显示 typing 状态"""
+        await self._send_chat_action(
+            self.client, user_name, ChatAction.TYPING, message_thread_id
+        )
+
+    async def send_typing(self) -> None:
+        message_thread_id = None
+        if self.get_message_type() == MessageType.GROUP_MESSAGE:
+            user_name = self.message_obj.group_id
+        else:
+            user_name = self.get_sender_id()
+
+        if "#" in user_name:
+            user_name, message_thread_id = user_name.split("#")
+
+        await self._ensure_typing(user_name, message_thread_id)
+
    @classmethod
    async def send_with_client(
        cls,
@@ -91,6 +165,11 @@ class TelegramPlatformEvent(AstrMessageEvent):
        if "#" in user_name:
            # it's a supergroup chat with message_thread_id
            user_name, message_thread_id = user_name.split("#")
+
+        # 根据消息链确定合适的 chat action 并发送
+        action = cls._get_chat_action_for_chain(message.chain)
+        await cls._send_chat_action(client, user_name, action, message_thread_id)
+
        for i in message.chain:
            payload = {
                "chat_id": user_name,
@@ -195,6 +274,12 @@ class TelegramPlatformEvent(AstrMessageEvent):
        message_id = None
        last_edit_time = 0  # 上次编辑消息的时间
        throttle_interval = 0.6  # 编辑消息的间隔时间 (秒)
+        last_chat_action_time = 0  # 上次发送 chat action 的时间
+        chat_action_interval = 0.5  # chat action 的节流间隔 (秒)
+
+        # 发送初始 typing 状态
+        await self._ensure_typing(user_name, message_thread_id)
+        last_chat_action_time = asyncio.get_event_loop().time()

        async for chain in generator:
            if isinstance(chain, MessageChain):
@@ -219,15 +304,25 @@ class TelegramPlatformEvent(AstrMessageEvent):
                        delta += i.text
                    elif isinstance(i, Image):
                        image_path = await i.convert_to_file_path()
-                        await self.client.send_photo(
-                            photo=image_path, **cast(Any, payload)
+                        await self._send_media_with_action(
+                            self.client,
+                            ChatAction.UPLOAD_PHOTO,
+                            self.client.send_photo,
+                            user_name=user_name,
+                            message_thread_id=message_thread_id,
+                            photo=image_path,
+                            **cast(Any, payload),
                        )
                        continue
                    elif isinstance(i, File):
                        path = await i.get_file()
                        name = i.name or os.path.basename(path)
-
-                        await self.client.send_document(
+                        await self._send_media_with_action(
+                            self.client,
+                            ChatAction.UPLOAD_DOCUMENT,
+                            self.client.send_document,
+                            user_name=user_name,
+                            message_thread_id=message_thread_id,
                            document=path,
                            filename=name,
                            **cast(Any, payload),
@@ -235,7 +330,15 @@ class TelegramPlatformEvent(AstrMessageEvent):
                        continue
                    elif isinstance(i, Record):
                        path = await i.convert_to_file_path()
-                        await self.client.send_voice(voice=path, **cast(Any, payload))
+                        await self._send_media_with_action(
+                            self.client,
+                            ChatAction.UPLOAD_VOICE,
+                            self.client.send_voice,
+                            user_name=user_name,
+                            message_thread_id=message_thread_id,
+                            voice=path,
+                            **cast(Any, payload),
+                        )
                        continue
                    else:
                        logger.warning(f"不支持的消息类型: {type(i)}")
@@ -248,6 +351,11 @@ class TelegramPlatformEvent(AstrMessageEvent):

                    # 如果距离上次编辑的时间 >= 设定的间隔，等待一段时间
                    if time_since_last_edit >= throttle_interval:
+                        # 发送 typing 状态（带节流）
+                        current_time = asyncio.get_event_loop().time()
+                        if current_time - last_chat_action_time >= chat_action_interval:
+                            await self._ensure_typing(user_name, message_thread_id)
+                            last_chat_action_time = current_time
                        # 编辑消息
                        try:
                            await self.client.edit_message_text(
@@ -263,6 +371,11 @@ class TelegramPlatformEvent(AstrMessageEvent):
                        )  # 更新上次编辑的时间
                else:
                    # delta 长度一般不会大于 4096，因此这里直接发送
+                    # 发送 typing 状态（带节流）
+                    current_time = asyncio.get_event_loop().time()
+                    if current_time - last_chat_action_time >= chat_action_interval:
+                        await self._ensure_typing(user_name, message_thread_id)
+                        last_chat_action_time = current_time
                    try:
                        msg = await self.client.send_message(
                            text=delta, **cast(Any, payload)
@@ -26,14 +26,23 @@ from .webchat_queue_mgr import WebChatQueueMgr, webchat_queue_mgr


 class QueueListener:
-    def __init__(self, webchat_queue_mgr: WebChatQueueMgr, callback: Callable) -> None:
+    def __init__(
+        self,
+        webchat_queue_mgr: WebChatQueueMgr,
+        callback: Callable,
+        stop_event: asyncio.Event,
+    ) -> None:
        self.webchat_queue_mgr = webchat_queue_mgr
        self.callback = callback
+        self.stop_event = stop_event

    async def run(self) -> None:
        """Register callback and keep adapter task alive."""
        self.webchat_queue_mgr.set_listener(self.callback)
-        await asyncio.Event().wait()
+        try:
+            await self.stop_event.wait()
+        finally:
+            await self.webchat_queue_mgr.clear_listener()


@register_platform_adapter("webchat", "webchat")
@@ -56,6 +65,8 @@ class WebChatAdapter(Platform):
            id="webchat",
            support_proactive_message=False,
        )
+        self._shutdown_event = asyncio.Event()
+        self._webchat_queue_mgr = webchat_queue_mgr

    async def send_by_session(
        self,
@@ -184,7 +195,7 @@ class WebChatAdapter(Platform):
            abm = await self.convert_message(data)
            await self.handle_msg(abm)

-        bot = QueueListener(webchat_queue_mgr, callback)
+        bot = QueueListener(self._webchat_queue_mgr, callback, self._shutdown_event)
        return bot.run()

    def meta(self) -> PlatformMetadata:
@@ -209,5 +220,4 @@ class WebChatAdapter(Platform):
        self.commit_event(message_event)

    async def terminate(self) -> None:
-        # Do nothing
-        pass
+        self._shutdown_event.set()
@@ -87,6 +87,19 @@ class WebChatQueueMgr:
        for conversation_id in list(self.queues.keys()):
            self._start_listener_if_needed(conversation_id)

+    async def clear_listener(self) -> None:
+        self._listener_callback = None
+        for close_event in list(self._queue_close_events.values()):
+            close_event.set()
+        self._queue_close_events.clear()
+
+        listener_tasks = list(self._listener_tasks.values())
+        for task in listener_tasks:
+            task.cancel()
+        if listener_tasks:
+            await asyncio.gather(*listener_tasks, return_exceptions=True)
+        self._listener_tasks.clear()
+
    def _start_listener_if_needed(self, conversation_id: str):
        if self._listener_callback is None:
            return
@@ -25,7 +25,7 @@ from astrbot.api.platform import (
 )
 from astrbot.core import logger
 from astrbot.core.platform.astr_message_event import MessageSesion
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.media_utils import convert_audio_to_wav
 from astrbot.core.utils.webhook_utils import log_webhook_info

@@ -344,7 +344,7 @@ class WecomPlatformAdapter(Platform):
                self.client.media.download,
                msg.media_id,
            )
-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+            temp_dir = get_astrbot_temp_path()
            path = os.path.join(temp_dir, f"wecom_{msg.media_id}.amr")
            with open(path, "wb") as f:
                f.write(resp.content)
@@ -400,7 +400,8 @@ class WecomPlatformAdapter(Platform):
                self.client.media.download,
                media_id,
            )
-            path = f"data/temp/wechat_kf_{media_id}.jpg"
+            temp_dir = get_astrbot_temp_path()
+            path = os.path.join(temp_dir, f"weixinkefu_{media_id}.jpg")
            with open(path, "wb") as f:
                f.write(resp.content)
            abm.message = [Image(file=path, url=path)]
@@ -412,7 +413,7 @@ class WecomPlatformAdapter(Platform):
                media_id,
            )

-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+            temp_dir = get_astrbot_temp_path()
            path = os.path.join(temp_dir, f"weixinkefu_{media_id}.amr")
            with open(path, "wb") as f:
                f.write(resp.content)
@@ -1,4 +1,5 @@
 import asyncio
+import os
 import sys
 import uuid
 from collections.abc import Awaitable, Callable
@@ -24,6 +25,7 @@ from astrbot.api.platform import (
 )
 from astrbot.core import logger
 from astrbot.core.platform.astr_message_event import MessageSesion
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.media_utils import convert_audio_to_wav
 from astrbot.core.utils.webhook_utils import log_webhook_info

@@ -290,12 +292,16 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
                self.client.media.download,
                msg.media_id,
            )
-            path = f"data/temp/wecom_{msg.media_id}.amr"
+            temp_dir = get_astrbot_temp_path()
+            path = os.path.join(temp_dir, f"weixin_offacc_{msg.media_id}.amr")
            with open(path, "wb") as f:
                f.write(resp.content)

            try:
-                path_wav = f"data/temp/wecom_{msg.media_id}.wav"
+                path_wav = os.path.join(
+                    temp_dir,
+                    f"weixin_offacc_{msg.media_id}.wav",
+                )
                path_wav = await convert_audio_to_wav(path, path_wav)
            except Exception as e:
                logger.error(
@@ -22,6 +22,7 @@ from astrbot.core.utils.network_utils import (
 )

 from ..register import register_provider_adapter
+from .default import with_model_request_retry


@register_provider_adapter(
@@ -204,6 +205,7 @@ class ProviderAnthropic(Provider):
        if usage.output_tokens is not None:
            token_usage.output = usage.output_tokens

+    @with_model_request_retry()
    async def _query(self, payloads: dict, tools: ToolSet | None) -> LLMResponse:
        if tools:
            if tool_list := tools.get_func_desc_anthropic_style():
@@ -265,6 +267,10 @@ class ProviderAnthropic(Provider):

        return llm_response

+    @with_model_request_retry()
+    async def _create_message_stream(self, payloads: dict, extra_body: dict):
+        return self.client.messages.stream(**payloads, extra_body=extra_body)
+
    async def _query_stream(
        self,
        payloads: dict,
@@ -293,9 +299,8 @@ class ProviderAnthropic(Provider):
                "type": "enabled",
            }

-        async with self.client.messages.stream(
-            **payloads, extra_body=extra_body
-        ) as stream:
+        stream_ctx = await self._create_message_stream(payloads, extra_body)
+        async with stream_ctx as stream:
            assert isinstance(stream, anthropic.AsyncMessageStream)
            async for event in stream:
                if event.type == "message_start":
@@ -12,12 +12,13 @@ from httpx import AsyncClient, Timeout

 from astrbot import logger
 from astrbot.core.config.default import VERSION
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
 from ..register import register_provider_adapter

-TEMP_DIR = Path("data/temp/azure_tts")
+TEMP_DIR = Path(get_astrbot_temp_path()) / "azure_tts"
 TEMP_DIR.mkdir(parents=True, exist_ok=True)


@@ -15,7 +15,7 @@ except (
 ):  # pragma: no cover - older dashscope versions without Qwen TTS support
    MultiModalConversation = None

-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -45,7 +45,7 @@ class ProviderDashscopeTTSAPI(TTSProvider):
        if not model:
            raise RuntimeError("Dashscope TTS model is not configured.")

-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        os.makedirs(temp_dir, exist_ok=True)

        if self._is_qwen_tts_model(model):
@@ -0,0 +1,38 @@
+from tenacity import (
+    AsyncRetrying,
+    retry,
+    retry_if_exception_type,
+    stop_after_attempt,
+    wait_exponential,
+)
+
+MODEL_REQUEST_RETRY_ATTEMPTS = 5
+MODEL_REQUEST_RETRY_WAIT_MAX_SECONDS = 15
+MODEL_REQUEST_RETRY_WAIT_MIN_SECONDS = 1
+MODEL_REQUEST_RETRY_WAIT_MULTIPLIER = 1
+
+
+def with_model_request_retry():
+    return retry(
+        retry=retry_if_exception_type(Exception),
+        stop=stop_after_attempt(MODEL_REQUEST_RETRY_ATTEMPTS),
+        wait=wait_exponential(
+            multiplier=MODEL_REQUEST_RETRY_WAIT_MULTIPLIER,
+            min=MODEL_REQUEST_RETRY_WAIT_MIN_SECONDS,
+            max=MODEL_REQUEST_RETRY_WAIT_MAX_SECONDS,
+        ),
+        reraise=True,
+    )
+
+
+def get_model_request_async_retrying() -> AsyncRetrying:
+    return AsyncRetrying(
+        retry=retry_if_exception_type(Exception),
+        stop=stop_after_attempt(MODEL_REQUEST_RETRY_ATTEMPTS),
+        wait=wait_exponential(
+            multiplier=MODEL_REQUEST_RETRY_WAIT_MULTIPLIER,
+            min=MODEL_REQUEST_RETRY_WAIT_MIN_SECONDS,
+            max=MODEL_REQUEST_RETRY_WAIT_MAX_SECONDS,
+        ),
+        reraise=True,
+    )
@@ -6,7 +6,7 @@ import uuid
 import edge_tts

 from astrbot.core import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -46,7 +46,7 @@ class ProviderEdgeTTS(TTSProvider):
        self.set_model("edge_tts")

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        mp3_path = os.path.join(temp_dir, f"edge_tts_temp_{uuid.uuid4()}.mp3")
        wav_path = os.path.join(temp_dir, f"edge_tts_{uuid.uuid4()}.wav")

@@ -8,7 +8,7 @@ from httpx import AsyncClient
 from pydantic import BaseModel, conint

 from astrbot import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -142,7 +142,7 @@ class ProviderFishAudioTTSAPI(TTSProvider):
        )

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        path = os.path.join(temp_dir, f"fishaudio_tts_api_{uuid.uuid4()}.wav")
        self.headers["content-type"] = "application/msgpack"
        request = await self._generate_request(text)
@@ -21,6 +21,7 @@ from astrbot.core.utils.io import download_image_by_url
 from astrbot.core.utils.network_utils import is_connection_error, log_connection_failure

 from ..register import register_provider_adapter
+from .default import get_model_request_async_retrying, with_model_request_retry


 class SuppressNonTextPartsWarning(logging.Filter):
@@ -513,6 +514,7 @@ class ProviderGoogleGenAI(Provider):
                llm_response.reasoning_signature = base64.b64encode(ts).decode("utf-8")
        return MessageChain(chain=chain)

+    @with_model_request_retry()
    async def _query(self, payloads: dict, tools: ToolSet | None) -> LLMResponse:
        """非流式请求 Gemini API"""
        system_instruction = next(
@@ -601,6 +603,17 @@ class ProviderGoogleGenAI(Provider):
        self,
        payloads: dict,
        tools: ToolSet | None,
+    ) -> AsyncGenerator[LLMResponse, None]:
+        async for attempt in get_model_request_async_retrying():
+            with attempt:
+                async for response in self._query_stream_once(payloads, tools):
+                    yield response
+                return
+
+    async def _query_stream_once(
+        self,
+        payloads: dict,
+        tools: ToolSet | None,
    ) -> AsyncGenerator[LLMResponse, None]:
        """流式请求 Gemini API"""
        system_instruction = next(
@@ -759,18 +772,7 @@ class ProviderGoogleGenAI(Provider):

        payloads = {"messages": context_query, "model": model}

-        retry = 10
-        keys = self.api_keys.copy()
-
-        for _ in range(retry):
-            try:
-                return await self._query(payloads, func_tool)
-            except APIError as e:
-                if await self._handle_api_error(e, keys):
-                    continue
-                break
-
-        raise Exception("请求失败。")
+        return await self._query(payloads, func_tool)

    async def text_chat_stream(
        self,
@@ -814,18 +816,8 @@ class ProviderGoogleGenAI(Provider):

        payloads = {"messages": context_query, "model": model}

-        retry = 10
-        keys = self.api_keys.copy()
-
-        for _ in range(retry):
-            try:
-                async for response in self._query_stream(payloads, func_tool):
-                    yield response
-                break
-            except APIError as e:
-                if await self._handle_api_error(e, keys):
-                    continue
-                break
+        async for response in self._query_stream(payloads, func_tool):
+            yield response

    async def get_models(self):
        try:
@@ -6,7 +6,7 @@ from google import genai
 from google.genai import types

 from astrbot import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -49,7 +49,7 @@ class ProviderGeminiTTSAPI(TTSProvider):
        self.voice_name: str = provider_config.get("gemini_tts_voice_name", "Leda")

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        path = os.path.join(temp_dir, f"gemini_tts_{uuid.uuid4()}.wav")
        prompt = f"{self.prefix}: {text}" if self.prefix else text
        response = await self.client.models.generate_content(
@@ -6,7 +6,7 @@ from astrbot.core import logger
 from astrbot.core.provider.entities import ProviderType
 from astrbot.core.provider.provider import TTSProvider
 from astrbot.core.provider.register import register_provider_adapter
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 try:
    import genie_tts as genie  # type: ignore
@@ -54,7 +54,7 @@ class GenieTTSProvider(TTSProvider):
        return True

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        os.makedirs(temp_dir, exist_ok=True)
        filename = f"genie_tts_{uuid.uuid4()}.wav"
        path = os.path.join(temp_dir, filename)
@@ -94,7 +94,7 @@ class GenieTTSProvider(TTSProvider):
                break

            try:
-                temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+                temp_dir = get_astrbot_temp_path()
                os.makedirs(temp_dir, exist_ok=True)
                filename = f"genie_tts_{uuid.uuid4()}.wav"
                path = os.path.join(temp_dir, filename)
@@ -5,7 +5,7 @@ import uuid
 import aiohttp

 from astrbot import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -121,7 +121,7 @@ class ProviderGSVTTS(TTSProvider):

        params = self.build_synthesis_params(text)

-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        os.makedirs(temp_dir, exist_ok=True)
        path = os.path.join(temp_dir, f"gsv_tts_{uuid.uuid4().hex}.wav")

@@ -4,7 +4,7 @@ import uuid

 import aiohttp

-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -29,7 +29,7 @@ class ProviderGSVITTS(TTSProvider):
        self.emotion = provider_config.get("emotion")

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        path = os.path.join(temp_dir, f"gsvi_tts_{uuid.uuid4()}.wav")
        params = {"text": text}

@@ -6,7 +6,7 @@ from collections.abc import AsyncIterator
 import aiohttp

 from astrbot.api import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -145,7 +145,7 @@ class ProviderMiniMaxTTSAPI(TTSProvider):
        return b"".join(chunks)

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        os.makedirs(temp_dir, exist_ok=True)
        path = os.path.join(temp_dir, f"minimax_tts_api_{uuid.uuid4()}.mp3")

@@ -5,6 +5,7 @@ import json
 import random
 import re
 from collections.abc import AsyncGenerator
+from typing import Any

 import httpx
 from openai import AsyncAzureOpenAI, AsyncOpenAI
@@ -27,8 +28,10 @@ from astrbot.core.utils.network_utils import (
    is_connection_error,
    log_connection_failure,
 )
+from astrbot.core.utils.string_utils import normalize_and_dedupe_strings

 from ..register import register_provider_adapter
+from .default import get_model_request_async_retrying, with_model_request_retry


@register_provider_adapter(
@@ -36,6 +39,128 @@ from ..register import register_provider_adapter
    "OpenAI API Chat Completion 提供商适配器",
 )
 class ProviderOpenAIOfficial(Provider):
+    _ERROR_TEXT_CANDIDATE_MAX_CHARS = 4096
+
+    @classmethod
+    def _truncate_error_text_candidate(cls, text: str) -> str:
+        if len(text) <= cls._ERROR_TEXT_CANDIDATE_MAX_CHARS:
+            return text
+        return text[: cls._ERROR_TEXT_CANDIDATE_MAX_CHARS]
+
+    @staticmethod
+    def _safe_json_dump(value: Any) -> str | None:
+        try:
+            return json.dumps(value, ensure_ascii=False, default=str)
+        except Exception:
+            return None
+
+    def _get_image_moderation_error_patterns(self) -> list[str]:
+        """Return configured moderation patterns (case-insensitive substring match, not regex)."""
+        configured = self.provider_config.get("image_moderation_error_patterns", [])
+        patterns: list[str] = []
+        if isinstance(configured, str):
+            configured = [configured]
+        if isinstance(configured, list):
+            for pattern in configured:
+                if not isinstance(pattern, str):
+                    continue
+                pattern = pattern.strip()
+                if pattern:
+                    patterns.append(pattern)
+        return patterns
+
+    @staticmethod
+    def _extract_error_text_candidates(error: Exception) -> list[str]:
+        candidates: list[str] = []
+
+        def _append_candidate(candidate: Any):
+            if candidate is None:
+                return
+            text = str(candidate).strip()
+            if not text:
+                return
+            candidates.append(
+                ProviderOpenAIOfficial._truncate_error_text_candidate(text)
+            )
+
+        _append_candidate(str(error))
+
+        body = getattr(error, "body", None)
+        if isinstance(body, dict):
+            err_obj = body.get("error")
+            body_text = ProviderOpenAIOfficial._safe_json_dump(
+                {"error": err_obj} if isinstance(err_obj, dict) else body
+            )
+            _append_candidate(body_text)
+            if isinstance(err_obj, dict):
+                for field in ("message", "type", "code", "param"):
+                    value = err_obj.get(field)
+                    if value is not None:
+                        _append_candidate(value)
+        elif isinstance(body, str):
+            _append_candidate(body)
+
+        response = getattr(error, "response", None)
+        if response is not None:
+            response_text = getattr(response, "text", None)
+            if isinstance(response_text, str):
+                _append_candidate(response_text)
+
+        return normalize_and_dedupe_strings(candidates)
+
+    def _is_content_moderated_upload_error(self, error: Exception) -> bool:
+        patterns = [
+            pattern.lower() for pattern in self._get_image_moderation_error_patterns()
+        ]
+        if not patterns:
+            return False
+        candidates = [
+            candidate.lower()
+            for candidate in self._extract_error_text_candidates(error)
+        ]
+        for pattern in patterns:
+            if any(pattern in candidate for candidate in candidates):
+                return True
+        return False
+
+    @staticmethod
+    def _context_contains_image(contexts: list[dict]) -> bool:
+        for context in contexts:
+            content = context.get("content")
+            if not isinstance(content, list):
+                continue
+            for item in content:
+                if isinstance(item, dict) and item.get("type") == "image_url":
+                    return True
+        return False
+
+    async def _fallback_to_text_only_and_retry(
+        self,
+        payloads: dict,
+        context_query: list,
+        chosen_key: str,
+        available_api_keys: list[str],
+        func_tool: ToolSet | None,
+        reason: str,
+        *,
+        image_fallback_used: bool = False,
+    ) -> tuple:
+        logger.warning(
+            "检测到图片请求失败（%s），已移除图片并重试（保留文本内容）。",
+            reason,
+        )
+        new_contexts = await self._remove_image_from_context(context_query)
+        payloads["messages"] = new_contexts
+        return (
+            False,
+            chosen_key,
+            available_api_keys,
+            payloads,
+            new_contexts,
+            func_tool,
+            image_fallback_used,
+        )
+
    def _create_http_client(self, provider_config: dict) -> httpx.AsyncClient | None:
        """创建带代理的 HTTP 客户端"""
        proxy = provider_config.get("proxy", "")
@@ -97,6 +222,7 @@ class ProviderOpenAIOfficial(Provider):
        except NotFoundError as e:
            raise Exception(f"获取模型列表失败：{e}")

+    @with_model_request_retry()
    async def _query(self, payloads: dict, tools: ToolSet | None) -> LLMResponse:
        if tools:
            model = payloads.get("model", "").lower()
@@ -122,8 +248,6 @@ class ProviderOpenAIOfficial(Provider):
        if isinstance(custom_extra_body, dict):
            extra_body.update(custom_extra_body)

-        model = payloads.get("model", "").lower()
-
        completion = await self.client.chat.completions.create(
            **payloads,
            stream=False,
@@ -145,6 +269,17 @@ class ProviderOpenAIOfficial(Provider):
        self,
        payloads: dict,
        tools: ToolSet | None,
+    ) -> AsyncGenerator[LLMResponse, None]:
+        async for attempt in get_model_request_async_retrying():
+            with attempt:
+                async for response in self._query_stream_once(payloads, tools):
+                    yield response
+                return
+
+    async def _query_stream_once(
+        self,
+        payloads: dict,
+        tools: ToolSet | None,
    ) -> AsyncGenerator[LLMResponse, None]:
        """流式查询API，逐步返回结果"""
        if tools:
@@ -199,7 +334,8 @@ class ProviderOpenAIOfficial(Provider):
                llm_response.reasoning_content = reasoning
                _y = True
            if delta.content:
-                completion_text = delta.content
+                # Don't strip streaming chunks to preserve spaces between words
+                completion_text = self._normalize_content(delta.content, strip=False)
                llm_response.result_chain = MessageChain(
                    chain=[Comp.Plain(completion_text)],
                )
@@ -247,6 +383,86 @@ class ProviderOpenAIOfficial(Provider):
            output=completion_tokens,
        )

+    @staticmethod
+    def _normalize_content(raw_content: Any, strip: bool = True) -> str:
+        """Normalize content from various formats to plain string.
+
+        Some LLM providers return content as list[dict] format
+        like [{'type': 'text', 'text': '...'}] instead of
+        plain string. This method handles both formats.
+
+        Args:
+            raw_content: The raw content from LLM response, can be str, list, or other.
+            strip: Whether to strip whitespace from the result. Set to False for
+                   streaming chunks to preserve spaces between words.
+
+        Returns:
+            Normalized plain text string.
+        """
+        if isinstance(raw_content, list):
+            # Check if this looks like OpenAI content-part format
+            # Only process if at least one item has {'type': 'text', 'text': ...} structure
+            has_content_part = any(
+                isinstance(part, dict) and part.get("type") == "text"
+                for part in raw_content
+            )
+            if has_content_part:
+                text_parts = []
+                for part in raw_content:
+                    if isinstance(part, dict) and part.get("type") == "text":
+                        text_val = part.get("text", "")
+                        # Coerce to str in case text is null or non-string
+                        text_parts.append(str(text_val) if text_val is not None else "")
+                return "".join(text_parts)
+            # Not content-part format, return string representation
+            return str(raw_content)
+
+        if isinstance(raw_content, str):
+            content = raw_content.strip() if strip else raw_content
+            # Check if the string is a JSON-encoded list (e.g., "[{'type': 'text', ...}]")
+            # This can happen when streaming concatenates content that was originally list format
+            # Only check if it looks like a complete JSON array (requires strip for check)
+            check_content = raw_content.strip()
+            if (
+                check_content.startswith("[")
+                and check_content.endswith("]")
+                and len(check_content) < 8192
+            ):
+                try:
+                    # First try standard JSON parsing
+                    parsed = json.loads(check_content)
+                except json.JSONDecodeError:
+                    # If that fails, try parsing as Python literal (handles single quotes)
+                    # This is safer than blind replace("'", '"') which corrupts apostrophes
+                    try:
+                        import ast
+
+                        parsed = ast.literal_eval(check_content)
+                    except (ValueError, SyntaxError):
+                        parsed = None
+
+                if isinstance(parsed, list):
+                    # Only convert if it matches OpenAI content-part schema
+                    # i.e., at least one item has {'type': 'text', 'text': ...}
+                    has_content_part = any(
+                        isinstance(part, dict) and part.get("type") == "text"
+                        for part in parsed
+                    )
+                    if has_content_part:
+                        text_parts = []
+                        for part in parsed:
+                            if isinstance(part, dict) and part.get("type") == "text":
+                                text_val = part.get("text", "")
+                                # Coerce to str in case text is null or non-string
+                                text_parts.append(
+                                    str(text_val) if text_val is not None else ""
+                                )
+                        if text_parts:
+                            return "".join(text_parts)
+            return content
+
+        return str(raw_content)
+
    async def _parse_openai_completion(
        self, completion: ChatCompletion, tools: ToolSet | None
    ) -> LLMResponse:
@@ -259,8 +475,7 @@ class ProviderOpenAIOfficial(Provider):

        # parse the text completion
        if choice.message.content is not None:
-            # text completion
-            completion_text = str(choice.message.content).strip()
+            completion_text = self._normalize_content(choice.message.content)
            # specially, some providers may set <think> tags around reasoning content in the completion text,
            # we use regex to remove them, and store then in reasoning_content field
            reasoning_pattern = re.compile(r"<think>(.*?)</think>", re.DOTALL)
@@ -270,6 +485,8 @@ class ProviderOpenAIOfficial(Provider):
                    [match.strip() for match in matches],
                )
                completion_text = reasoning_pattern.sub("", completion_text).strip()
+            # Also clean up orphan </think> tags that may leak from some models
+            completion_text = re.sub(r"</think>\s*$", "", completion_text).strip()
            llm_response.result_chain = MessageChain().message(completion_text)

        # parse the reasoning content if any
@@ -403,6 +620,7 @@ class ProviderOpenAIOfficial(Provider):
        available_api_keys: list[str],
        retry_cnt: int,
        max_retries: int,
+        image_fallback_used: bool = False,
    ) -> tuple:
        """处理API错误并尝试恢复"""
        if "429" in str(e):
@@ -422,6 +640,7 @@ class ProviderOpenAIOfficial(Provider):
                    payloads,
                    context_query,
                    func_tool,
+                    image_fallback_used,
                )
            raise e
        if "maximum context length" in str(e):
@@ -437,20 +656,34 @@ class ProviderOpenAIOfficial(Provider):
                payloads,
                context_query,
                func_tool,
+                image_fallback_used,
            )
        if "The model is not a VLM" in str(e):  # siliconcloud
+            if image_fallback_used or not self._context_contains_image(context_query):
+                raise e
            # 尝试删除所有 image
-            new_contexts = await self._remove_image_from_context(context_query)
-            payloads["messages"] = new_contexts
-            context_query = new_contexts
-            return (
-                False,
-                chosen_key,
-                available_api_keys,
+            return await self._fallback_to_text_only_and_retry(
                payloads,
                context_query,
+                chosen_key,
+                available_api_keys,
                func_tool,
+                "model_not_vlm",
+                image_fallback_used=True,
            )
+        if self._is_content_moderated_upload_error(e):
+            if image_fallback_used or not self._context_contains_image(context_query):
+                raise e
+            return await self._fallback_to_text_only_and_retry(
+                payloads,
+                context_query,
+                chosen_key,
+                available_api_keys,
+                func_tool,
+                "image_content_moderated",
+                image_fallback_used=True,
+            )
+
        if (
            "Function calling is not enabled" in str(e)
            or ("tool" in str(e).lower() and "support" in str(e).lower())
@@ -461,7 +694,15 @@ class ProviderOpenAIOfficial(Provider):
                f"{self.get_model()} 不支持函数工具调用，已自动去除，不影响使用。",
            )
            payloads.pop("tools", None)
-            return False, chosen_key, available_api_keys, payloads, context_query, None
+            return (
+                False,
+                chosen_key,
+                available_api_keys,
+                payloads,
+                context_query,
+                None,
+                image_fallback_used,
+            )
        # logger.error(f"发生了错误。Provider 配置如下: {self.provider_config}")

        if "tool" in str(e).lower() and "support" in str(e).lower():
@@ -486,7 +727,7 @@ class ProviderOpenAIOfficial(Provider):
        extra_user_content_parts=None,
        **kwargs,
    ) -> LLMResponse:
-        payloads, context_query = await self._prepare_chat_payload(
+        payloads, _ = await self._prepare_chat_payload(
            prompt,
            image_urls,
            contexts,
@@ -498,44 +739,9 @@ class ProviderOpenAIOfficial(Provider):
        )

        llm_response = None
-        max_retries = 10
-        available_api_keys = self.api_keys.copy()
-        chosen_key = random.choice(available_api_keys)
-
-        last_exception = None
-        retry_cnt = 0
-        for retry_cnt in range(max_retries):
-            try:
-                self.client.api_key = chosen_key
-                llm_response = await self._query(payloads, func_tool)
-                break
-            except Exception as e:
-                last_exception = e
-                (
-                    success,
-                    chosen_key,
-                    available_api_keys,
-                    payloads,
-                    context_query,
-                    func_tool,
-                ) = await self._handle_api_error(
-                    e,
-                    payloads,
-                    context_query,
-                    func_tool,
-                    chosen_key,
-                    available_api_keys,
-                    retry_cnt,
-                    max_retries,
-                )
-                if success:
-                    break
-
-        if retry_cnt == max_retries - 1 or llm_response is None:
-            logger.error(f"API 调用失败，重试 {max_retries} 次仍然失败。")
-            if last_exception is None:
-                raise Exception("未知错误")
-            raise last_exception
+        if self.api_keys:
+            self.client.api_key = random.choice(self.api_keys)
+        llm_response = await self._query(payloads, func_tool)
        return llm_response

    async def text_chat_stream(
@@ -551,7 +757,7 @@ class ProviderOpenAIOfficial(Provider):
        **kwargs,
    ) -> AsyncGenerator[LLMResponse, None]:
        """流式对话，与服务商交互并逐步返回结果"""
-        payloads, context_query = await self._prepare_chat_payload(
+        payloads, _ = await self._prepare_chat_payload(
            prompt,
            image_urls,
            contexts,
@@ -561,45 +767,10 @@ class ProviderOpenAIOfficial(Provider):
            **kwargs,
        )

-        max_retries = 10
-        available_api_keys = self.api_keys.copy()
-        chosen_key = random.choice(available_api_keys)
-
-        last_exception = None
-        retry_cnt = 0
-        for retry_cnt in range(max_retries):
-            try:
-                self.client.api_key = chosen_key
-                async for response in self._query_stream(payloads, func_tool):
-                    yield response
-                break
-            except Exception as e:
-                last_exception = e
-                (
-                    success,
-                    chosen_key,
-                    available_api_keys,
-                    payloads,
-                    context_query,
-                    func_tool,
-                ) = await self._handle_api_error(
-                    e,
-                    payloads,
-                    context_query,
-                    func_tool,
-                    chosen_key,
-                    available_api_keys,
-                    retry_cnt,
-                    max_retries,
-                )
-                if success:
-                    break
-
-        if retry_cnt == max_retries - 1:
-            logger.error(f"API 调用失败，重试 {max_retries} 次仍然失败。")
-            if last_exception is None:
-                raise Exception("未知错误")
-            raise last_exception
+        if self.api_keys:
+            self.client.api_key = random.choice(self.api_keys)
+        async for response in self._query_stream(payloads, func_tool):
+            yield response

    async def _remove_image_from_context(self, contexts: list):
        """从上下文中删除所有带有 image 的记录"""
@@ -5,7 +5,7 @@ import httpx
 from openai import NOT_GIVEN, AsyncOpenAI

 from astrbot import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -46,7 +46,7 @@ class ProviderOpenAITTSAPI(TTSProvider):
        self.set_model(provider_config.get("model", ""))

    async def get_audio(self, text: str) -> str:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        path = os.path.join(temp_dir, f"openai_tts_api_{uuid.uuid4()}.wav")
        async with self.client.audio.speech.with_streaming_response.create(
            model=self.model_name,
@@ -8,6 +8,7 @@ import uuid
 import aiohttp

 from astrbot import logger
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..entities import ProviderType
 from ..provider import TTSProvider
@@ -92,9 +93,12 @@ class ProviderVolcengineTTS(TTSProvider):
                    if "data" in resp_data:
                        audio_data = base64.b64decode(resp_data["data"])

-                        os.makedirs("data/temp", exist_ok=True)
-
-                        file_path = f"data/temp/volcengine_tts_{uuid.uuid4()}.mp3"
+                        temp_dir = get_astrbot_temp_path()
+                        os.makedirs(temp_dir, exist_ok=True)
+                        file_path = os.path.join(
+                            temp_dir,
+                            f"volcengine_tts_{uuid.uuid4()}.mp3",
+                        )

                        loop = asyncio.get_running_loop()
                        await loop.run_in_executor(
@@ -4,7 +4,7 @@ import uuid
 from openai import NOT_GIVEN, AsyncOpenAI

 from astrbot.core import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_file
 from astrbot.core.utils.tencent_record_helper import (
    convert_to_pcm_wav,
@@ -65,9 +65,11 @@ class ProviderOpenAIWhisperAPI(STTProvider):
            if "multimedia.nt.qq.com.cn" in audio_url:
                is_tencent = True

-            name = str(uuid.uuid4())
-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-            path = os.path.join(temp_dir, name)
+            temp_dir = get_astrbot_temp_path()
+            path = os.path.join(
+                temp_dir,
+                f"whisper_api_{uuid.uuid4().hex[:8]}.input",
+            )
            await download_file(audio_url, path)
            audio_url = path

@@ -79,8 +81,11 @@ class ProviderOpenAIWhisperAPI(STTProvider):

            # 判断是否需要转换
            if file_format in ["silk", "amr"]:
-                temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-                output_path = os.path.join(temp_dir, str(uuid.uuid4()) + ".wav")
+                temp_dir = get_astrbot_temp_path()
+                output_path = os.path.join(
+                    temp_dir,
+                    f"whisper_api_{uuid.uuid4().hex[:8]}.wav",
+                )

                if file_format == "silk":
                    logger.info(
@@ -6,7 +6,7 @@ from typing import cast
 import whisper

 from astrbot.core import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.io import download_file
 from astrbot.core.utils.tencent_record_helper import tencent_silk_to_wav

@@ -58,9 +58,11 @@ class ProviderOpenAIWhisperSelfHost(STTProvider):
            if "multimedia.nt.qq.com.cn" in audio_url:
                is_tencent = True

-            name = str(uuid.uuid4())
-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-            path = os.path.join(temp_dir, name)
+            temp_dir = get_astrbot_temp_path()
+            path = os.path.join(
+                temp_dir,
+                f"whisper_selfhost_{uuid.uuid4().hex[:8]}.input",
+            )
            await download_file(audio_url, path)
            audio_url = path

@@ -71,8 +73,11 @@ class ProviderOpenAIWhisperSelfHost(STTProvider):
            is_silk = await self._is_silk_file(audio_url)
            if is_silk:
                logger.info("Converting silk file to wav ...")
-                temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-                output_path = os.path.join(temp_dir, str(uuid.uuid4()) + ".wav")
+                temp_dir = get_astrbot_temp_path()
+                output_path = os.path.join(
+                    temp_dir,
+                    f"whisper_selfhost_{uuid.uuid4().hex[:8]}.wav",
+                )
                await tencent_silk_to_wav(audio_url, output_path)
                audio_url = output_path

@@ -7,7 +7,7 @@ from xinference_client.client.restful.async_restful_client import (
 )

 from astrbot.core import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
 from astrbot.core.utils.tencent_record_helper import (
    convert_to_pcm_wav,
    tencent_silk_to_wav,
@@ -130,11 +130,17 @@ class ProviderXinferenceSTT(STTProvider):
                logger.info(
                    f"Audio requires conversion ({conversion_type}), using temporary files..."
                )
-                temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+                temp_dir = get_astrbot_temp_path()
                os.makedirs(temp_dir, exist_ok=True)

-                input_path = os.path.join(temp_dir, str(uuid.uuid4()))
-                output_path = os.path.join(temp_dir, str(uuid.uuid4()) + ".wav")
+                input_path = os.path.join(
+                    temp_dir,
+                    f"xinference_stt_{uuid.uuid4().hex[:8]}.input",
+                )
+                output_path = os.path.join(
+                    temp_dir,
+                    f"xinference_stt_{uuid.uuid4().hex[:8]}.wav",
+                )
                temp_files.extend([input_path, output_path])

                with open(input_path, "wb") as f:
@@ -93,7 +93,6 @@ class SkillManager:
        self.skills_root = skills_root or get_astrbot_skills_path()
        self.config_path = os.path.join(get_astrbot_data_path(), SKILLS_CONFIG_FILENAME)
        os.makedirs(self.skills_root, exist_ok=True)
-        os.makedirs(get_astrbot_temp_path(), exist_ok=True)

    def _load_config(self) -> dict:
        if not os.path.exists(self.config_path):
@@ -20,6 +20,7 @@ class PlatformAdapterType(enum.Flag):
    WEIXIN_OFFICIAL_ACCOUNT = enum.auto()
    SATORI = enum.auto()
    MISSKEY = enum.auto()
+    LINE = enum.auto()
    ALL = (
        AIOCQHTTP
        | QQOFFICIAL
@@ -34,6 +35,7 @@ class PlatformAdapterType(enum.Flag):
        | WEIXIN_OFFICIAL_ACCOUNT
        | SATORI
        | MISSKEY
+        | LINE
    )


@@ -51,6 +53,7 @@ ADAPTER_NAME_2_TYPE = {
    "weixin_official_account": PlatformAdapterType.WEIXIN_OFFICIAL_ACCOUNT,
    "satori": PlatformAdapterType.SATORI,
    "misskey": PlatformAdapterType.MISSKEY,
+    "line": PlatformAdapterType.LINE,
 }


@@ -62,6 +62,9 @@ class PluginManager:
        self._pm_lock = asyncio.Lock()
        """StarManager操作互斥锁"""

+        self.failed_plugin_dict = {}
+        """加载失败插件的信息，用于后续可能的热重载"""
+
        self.failed_plugin_info = ""
        if os.getenv("ASTRBOT_RELOAD", "0") == "1":
            asyncio.create_task(self._watch_plugins_changes())
@@ -191,6 +194,38 @@ class PluginManager:
                    await pip_installer.install(requirements_path=pth)
                except Exception as e:
                    logger.error(f"更新插件 {p} 的依赖失败。Code: {e!s}")
+        return True
+
+    async def _import_plugin_with_dependency_recovery(
+        self,
+        path: str,
+        module_str: str,
+        root_dir_name: str,
+        requirements_path: str,
+    ) -> ModuleType:
+        try:
+            return __import__(path, fromlist=[module_str])
+        except (ModuleNotFoundError, ImportError) as import_exc:
+            if os.path.exists(requirements_path):
+                try:
+                    logger.info(
+                        f"插件 {root_dir_name} 导入失败，尝试从已安装依赖恢复: {import_exc!s}"
+                    )
+                    pip_installer.prefer_installed_dependencies(
+                        requirements_path=requirements_path
+                    )
+                    module = __import__(path, fromlist=[module_str])
+                    logger.info(
+                        f"插件 {root_dir_name} 已从 site-packages 恢复依赖，跳过重新安装。"
+                    )
+                    return module
+                except Exception as recover_exc:
+                    logger.info(
+                        f"插件 {root_dir_name} 已安装依赖恢复失败，将重新安装依赖: {recover_exc!s}"
+                    )
+
+            await self._check_plugin_dept_update(target_plugin=root_dir_name)
+            return __import__(path, fromlist=[module_str])

    @staticmethod
    def _load_plugin_metadata(plugin_path: str, plugin_obj=None) -> StarMetadata | None:
@@ -295,6 +330,28 @@ class PluginManager:
                except KeyError:
                    logger.warning(f"模块 {module_name} 未载入")

+    async def reload_failed_plugin(self, dir_name):
+        """
+        重新加载未注册（加载失败）的插件
+        Args:
+            dir_name (str): 要重载的特定插件名称。
+        Returns:
+            tuple: 返回 load() 方法的结果，包含 (success, error_message)
+                - success (bool): 重载是否成功
+                - error_message (str|None): 错误信息，成功时为 None
+        """
+        async with self._pm_lock:
+            if dir_name in self.failed_plugin_dict:
+                success, error = await self.load(specified_dir_name=dir_name)
+                if success:
+                    self.failed_plugin_dict.pop(dir_name, None)
+                    if not self.failed_plugin_dict:
+                        self.failed_plugin_info = ""
+                    return success, None
+                else:
+                    return False, error
+            return False, "插件不存在于失败列表中"
+
    async def reload(self, specified_plugin_name=None):
        """重新加载插件

@@ -385,6 +442,12 @@ class PluginManager:
                    "reserved",
                    False,
                )  # 是否是保留插件。目前在 astrbot/builtin_stars 目录下的都是保留插件。保留插件不可以卸载。
+                plugin_dir_path = (
+                    os.path.join(self.plugin_store_path, root_dir_name)
+                    if not reserved
+                    else os.path.join(self.reserved_plugin_path, root_dir_name)
+                )
+                requirements_path = os.path.join(plugin_dir_path, "requirements.txt")

                path = "data.plugins." if not reserved else "astrbot.builtin_stars."
                path += root_dir_name + "." + module_str
@@ -399,11 +462,12 @@ class PluginManager:

                # 尝试导入模块
                try:
-                    module = __import__(path, fromlist=[module_str])
-                except (ModuleNotFoundError, ImportError):
-                    # 尝试安装依赖
-                    await self._check_plugin_dept_update(target_plugin=root_dir_name)
-                    module = __import__(path, fromlist=[module_str])
+                    module = await self._import_plugin_with_dependency_recovery(
+                        path=path,
+                        module_str=module_str,
+                        root_dir_name=root_dir_name,
+                        requirements_path=requirements_path,
+                    )
                except Exception as e:
                    logger.error(traceback.format_exc())
                    logger.error(f"插件 {root_dir_name} 导入失败。原因：{e!s}")
@@ -411,11 +475,6 @@ class PluginManager:

                # 检查 _conf_schema.json
                plugin_config = None
-                plugin_dir_path = (
-                    os.path.join(self.plugin_store_path, root_dir_name)
-                    if not reserved
-                    else os.path.join(self.reserved_plugin_path, root_dir_name)
-                )
                plugin_schema_path = os.path.join(
                    plugin_dir_path,
                    self.conf_schema_fname,
@@ -629,6 +688,11 @@ class PluginManager:
                    logger.error(f"| {line}")
                logger.error("----------------------------------")
                fail_rec += f"加载 {root_dir_name} 插件时出现问题，原因 {e!s}。\n"
+                self.failed_plugin_dict[root_dir_name] = {
+                    "error": str(e),
+                    "traceback": errors,
+                }
+                # 记录注册失败的插件名称，以便后续重载插件

        # 清除 pip.main 导致的多余的 logging handlers
        for handler in logging.root.handlers[:]:
@@ -644,6 +708,49 @@ class PluginManager:
        self.failed_plugin_info = fail_rec
        return False, fail_rec

+    async def _cleanup_failed_plugin_install(
+        self,
+        dir_name: str,
+        plugin_path: str,
+    ) -> None:
+        plugin = None
+        for star in self.context.get_all_stars():
+            if star.root_dir_name == dir_name:
+                plugin = star
+                break
+
+        if plugin and plugin.name and plugin.module_path:
+            try:
+                await self._terminate_plugin(plugin)
+            except Exception:
+                logger.warning(traceback.format_exc())
+            try:
+                await self._unbind_plugin(plugin.name, plugin.module_path)
+            except Exception:
+                logger.warning(traceback.format_exc())
+
+        if os.path.exists(plugin_path):
+            try:
+                remove_dir(plugin_path)
+                logger.warning(f"已清理安装失败的插件目录: {plugin_path}")
+            except Exception as e:
+                logger.warning(
+                    f"清理安装失败插件目录失败: {plugin_path}，原因: {e!s}",
+                )
+
+        plugin_config_path = os.path.join(
+            self.plugin_config_path,
+            f"{dir_name}_config.json",
+        )
+        if os.path.exists(plugin_config_path):
+            try:
+                os.remove(plugin_config_path)
+                logger.warning(f"已清理安装失败插件配置: {plugin_config_path}")
+            except Exception as e:
+                logger.warning(
+                    f"清理安装失败插件配置失败: {plugin_config_path}，原因: {e!s}",
+                )
+
    async def install_plugin(self, repo_url: str, proxy=""):
        """从仓库 URL 安装插件

@@ -669,44 +776,62 @@ class PluginManager:
        )

        async with self._pm_lock:
-            plugin_path = await self.updator.install(repo_url, proxy)
-            # reload the plugin
-            dir_name = os.path.basename(plugin_path)
-            await self.load(specified_dir_name=dir_name)
+            plugin_path = ""
+            dir_name = ""
+            cleanup_required = False
+            try:
+                plugin_path = await self.updator.install(repo_url, proxy)
+                cleanup_required = True

-            # Get the plugin metadata to return repo info
-            plugin = self.context.get_registered_star(dir_name)
-            if not plugin:
-                # Try to find by other name if directory name doesn't match plugin name
-                for star in self.context.get_all_stars():
-                    if star.root_dir_name == dir_name:
-                        plugin = star
-                        break
-
-            # Extract README.md content if exists
-            readme_content = None
-            readme_path = os.path.join(plugin_path, "README.md")
-            if not os.path.exists(readme_path):
-                readme_path = os.path.join(plugin_path, "readme.md")
-
-            if os.path.exists(readme_path):
-                try:
-                    with open(readme_path, encoding="utf-8") as f:
-                        readme_content = f.read()
-                except Exception as e:
-                    logger.warning(
-                        f"读取插件 {dir_name} 的 README.md 文件失败: {e!s}",
+                # reload the plugin
+                dir_name = os.path.basename(plugin_path)
+                success, error_message = await self.load(specified_dir_name=dir_name)
+                if not success:
+                    raise Exception(
+                        error_message
+                        or f"安装插件 {dir_name} 失败，请检查插件依赖或兼容性。"
                    )

-            plugin_info = None
-            if plugin:
-                plugin_info = {
-                    "repo": plugin.repo,
-                    "readme": readme_content,
-                    "name": plugin.name,
-                }
+                # Get the plugin metadata to return repo info
+                plugin = self.context.get_registered_star(dir_name)
+                if not plugin:
+                    # Try to find by other name if directory name doesn't match plugin name
+                    for star in self.context.get_all_stars():
+                        if star.root_dir_name == dir_name:
+                            plugin = star
+                            break

-            return plugin_info
+                # Extract README.md content if exists
+                readme_content = None
+                readme_path = os.path.join(plugin_path, "README.md")
+                if not os.path.exists(readme_path):
+                    readme_path = os.path.join(plugin_path, "readme.md")
+
+                if os.path.exists(readme_path):
+                    try:
+                        with open(readme_path, encoding="utf-8") as f:
+                            readme_content = f.read()
+                    except Exception as e:
+                        logger.warning(
+                            f"读取插件 {dir_name} 的 README.md 文件失败: {e!s}",
+                        )
+
+                plugin_info = None
+                if plugin:
+                    plugin_info = {
+                        "repo": plugin.repo,
+                        "readme": readme_content,
+                        "name": plugin.name,
+                    }
+
+                return plugin_info
+            except Exception:
+                if cleanup_required and dir_name and plugin_path:
+                    await self._cleanup_failed_plugin_install(
+                        dir_name=dir_name,
+                        plugin_path=plugin_path,
+                    )
+                raise

    async def uninstall_plugin(
        self,
@@ -968,6 +1093,7 @@ class PluginManager:
        dir_name = os.path.basename(zip_file_path).replace(".zip", "")
        dir_name = dir_name.removesuffix("-master").removesuffix("-main").lower()
        desti_dir = os.path.join(self.plugin_store_path, dir_name)
+        cleanup_required = False

        # 第一步：检查是否已安装同目录名的插件，先终止旧插件
        existing_plugin = None
@@ -987,74 +1113,88 @@ class PluginManager:
                    existing_plugin.name, existing_plugin.module_path
                )

-        self.updator.unzip_file(zip_file_path, desti_dir)
-
-        # 第二步：解压后，读取新插件的 metadata.yaml，检查是否存在同名但不同目录的插件
        try:
-            new_metadata = self._load_plugin_metadata(desti_dir)
-            if new_metadata and new_metadata.name:
-                for star in self.context.get_all_stars():
-                    if (
-                        star.name == new_metadata.name
-                        and star.root_dir_name != dir_name
-                    ):
-                        logger.warning(
-                            f"检测到同名插件 {star.name} 存在于不同目录 {star.root_dir_name}，正在终止..."
-                        )
-                        try:
-                            await self._terminate_plugin(star)
-                        except Exception:
-                            logger.warning(traceback.format_exc())
-                        if star.name and star.module_path:
-                            await self._unbind_plugin(star.name, star.module_path)
-                        break  # 只处理第一个匹配的
-        except Exception as e:
-            logger.debug(f"读取新插件 metadata.yaml 失败，跳过同名检查: {e!s}")
+            self.updator.unzip_file(zip_file_path, desti_dir)
+            cleanup_required = True

-        # remove the zip
-        try:
-            os.remove(zip_file_path)
-        except BaseException as e:
-            logger.warning(f"删除插件压缩包失败: {e!s}")
-        # await self.reload()
-        await self.load(specified_dir_name=dir_name)
-
-        # Get the plugin metadata to return repo info
-        plugin = self.context.get_registered_star(dir_name)
-        if not plugin:
-            # Try to find by other name if directory name doesn't match plugin name
-            for star in self.context.get_all_stars():
-                if star.root_dir_name == dir_name:
-                    plugin = star
-                    break
-
-        # Extract README.md content if exists
-        readme_content = None
-        readme_path = os.path.join(desti_dir, "README.md")
-        if not os.path.exists(readme_path):
-            readme_path = os.path.join(desti_dir, "readme.md")
-
-        if os.path.exists(readme_path):
+            # 第二步：解压后，读取新插件的 metadata.yaml，检查是否存在同名但不同目录的插件
            try:
-                with open(readme_path, encoding="utf-8") as f:
-                    readme_content = f.read()
+                new_metadata = self._load_plugin_metadata(desti_dir)
+                if new_metadata and new_metadata.name:
+                    for star in self.context.get_all_stars():
+                        if (
+                            star.name == new_metadata.name
+                            and star.root_dir_name != dir_name
+                        ):
+                            logger.warning(
+                                f"检测到同名插件 {star.name} 存在于不同目录 {star.root_dir_name}，正在终止..."
+                            )
+                            try:
+                                await self._terminate_plugin(star)
+                            except Exception:
+                                logger.warning(traceback.format_exc())
+                            if star.name and star.module_path:
+                                await self._unbind_plugin(star.name, star.module_path)
+                            break  # 只处理第一个匹配的
            except Exception as e:
-                logger.warning(f"读取插件 {dir_name} 的 README.md 文件失败: {e!s}")
+                logger.debug(f"读取新插件 metadata.yaml 失败，跳过同名检查: {e!s}")

-        plugin_info = None
-        if plugin:
-            plugin_info = {
-                "repo": plugin.repo,
-                "readme": readme_content,
-                "name": plugin.name,
-            }
-
-            if plugin.repo:
-                asyncio.create_task(
-                    Metric.upload(
-                        et="install_star_f",  # install star
-                        repo=plugin.repo,
-                    ),
+            # remove the zip
+            try:
+                os.remove(zip_file_path)
+            except BaseException as e:
+                logger.warning(f"删除插件压缩包失败: {e!s}")
+            # await self.reload()
+            success, error_message = await self.load(specified_dir_name=dir_name)
+            if not success:
+                raise Exception(
+                    error_message
+                    or f"安装插件 {dir_name} 失败，请检查插件依赖或兼容性。"
                )

-        return plugin_info
+            # Get the plugin metadata to return repo info
+            plugin = self.context.get_registered_star(dir_name)
+            if not plugin:
+                # Try to find by other name if directory name doesn't match plugin name
+                for star in self.context.get_all_stars():
+                    if star.root_dir_name == dir_name:
+                        plugin = star
+                        break
+
+            # Extract README.md content if exists
+            readme_content = None
+            readme_path = os.path.join(desti_dir, "README.md")
+            if not os.path.exists(readme_path):
+                readme_path = os.path.join(desti_dir, "readme.md")
+
+            if os.path.exists(readme_path):
+                try:
+                    with open(readme_path, encoding="utf-8") as f:
+                        readme_content = f.read()
+                except Exception as e:
+                    logger.warning(f"读取插件 {dir_name} 的 README.md 文件失败: {e!s}")
+
+            plugin_info = None
+            if plugin:
+                plugin_info = {
+                    "repo": plugin.repo,
+                    "readme": readme_content,
+                    "name": plugin.name,
+                }
+
+                if plugin.repo:
+                    asyncio.create_task(
+                        Metric.upload(
+                            et="install_star_f",  # install star
+                            repo=plugin.repo,
+                        ),
+                    )
+
+            return plugin_info
+        except Exception:
+            if cleanup_required:
+                await self._cleanup_failed_plugin_install(
+                    dir_name=dir_name,
+                    plugin_path=desti_dir,
+                )
+            raise
@@ -14,7 +14,7 @@ import certifi
 import psutil
 from PIL import Image

-from .astrbot_path import get_astrbot_data_path
+from .astrbot_path import get_astrbot_data_path, get_astrbot_temp_path

 logger = logging.getLogger("astrbot")

@@ -50,21 +50,10 @@ def port_checker(port: int, host: str = "localhost") -> bool:


 def save_temp_img(img: Image.Image | bytes) -> str:
-    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
-    # 获得文件创建时间，清除超过 12 小时的
-    try:
-        for f in os.listdir(temp_dir):
-            path = os.path.join(temp_dir, f)
-            if os.path.isfile(path):
-                ctime = os.path.getctime(path)
-                if time.time() - ctime > 3600 * 12:
-                    os.remove(path)
-    except Exception as e:
-        print(f"清除临时文件失败: {e}")
-
+    temp_dir = get_astrbot_temp_path()
    # 获得时间戳
    timestamp = f"{int(time.time())}_{uuid.uuid4().hex[:8]}"
-    p = os.path.join(temp_dir, f"{timestamp}.jpg")
+    p = os.path.join(temp_dir, f"io_temp_img_{timestamp}.jpg")

    if isinstance(img, Image.Image):
        img.save(p)
@@ -10,7 +10,7 @@ import uuid
 from pathlib import Path

 from astrbot import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path


 async def get_media_duration(file_path: str) -> int | None:
@@ -77,9 +77,9 @@ async def convert_audio_to_opus(audio_path: str, output_path: str | None = None)

    # 生成输出文件路径
    if output_path is None:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        os.makedirs(temp_dir, exist_ok=True)
-        output_path = os.path.join(temp_dir, f"{uuid.uuid4()}.opus")
+        output_path = os.path.join(temp_dir, f"media_audio_{uuid.uuid4().hex}.opus")

    try:
        # 使用ffmpeg转换为opus格式
@@ -156,9 +156,12 @@ async def convert_video_format(

    # 生成输出文件路径
    if output_path is None:
-        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        temp_dir = get_astrbot_temp_path()
        os.makedirs(temp_dir, exist_ok=True)
-        output_path = os.path.join(temp_dir, f"{uuid.uuid4()}.{output_format}")
+        output_path = os.path.join(
+            temp_dir,
+            f"media_video_{uuid.uuid4().hex}.{output_format}",
+        )

    try:
        # 使用ffmpeg转换视频格式
@@ -227,9 +230,9 @@ async def convert_audio_format(
        return audio_path

    if output_path is None:
-        temp_dir = Path(get_astrbot_data_path()) / "temp"
+        temp_dir = Path(get_astrbot_temp_path())
        temp_dir.mkdir(parents=True, exist_ok=True)
-        output_path = str(temp_dir / f"{uuid.uuid4()}.{output_format}")
+        output_path = str(temp_dir / f"media_audio_{uuid.uuid4().hex}.{output_format}")

    args = ["ffmpeg", "-y", "-i", audio_path]
    if output_format == "amr":
@@ -283,9 +286,9 @@ async def extract_video_cover(
 ) -> str:
    """从视频中提取封面图（JPG）。"""
    if output_path is None:
-        temp_dir = Path(get_astrbot_data_path()) / "temp"
+        temp_dir = Path(get_astrbot_temp_path())
        temp_dir.mkdir(parents=True, exist_ok=True)
-        output_path = str(temp_dir / f"{uuid.uuid4()}.jpg")
+        output_path = str(temp_dir / f"media_cover_{uuid.uuid4().hex}.jpg")

    try:
        process = await asyncio.create_subprocess_exec(
@@ -77,9 +77,7 @@ def log_connection_failure(
            f"代理地址: {effective_proxy}，错误: {error}"
        )
    else:
-        logger.error(
-            f"[{provider_label}] 网络连接失败 ({error_type})，未配置代理。错误: {error}"
-        )
+        logger.error(f"[{provider_label}] 网络连接失败 ({error_type})。错误: {error}")


 def create_proxy_client(
@@ -1,10 +1,15 @@
 import asyncio
 import contextlib
 import importlib
+import importlib.metadata as importlib_metadata
+import importlib.util
 import io
 import logging
 import os
+import re
 import sys
+import threading
+from collections import deque

 from astrbot.core.utils.astrbot_path import get_astrbot_site_packages_path
 from astrbot.core.utils.runtime_env import is_packaged_electron_runtime
@@ -12,6 +17,11 @@ from astrbot.core.utils.runtime_env import is_packaged_electron_runtime
 logger = logging.getLogger("astrbot")

 _DISTLIB_FINDER_PATCH_ATTEMPTED = False
+_SITE_PACKAGES_IMPORT_LOCK = threading.RLock()
+
+
+def _canonicalize_distribution_name(name: str) -> str:
+    return re.sub(r"[-_.]+", "-", name).strip("-").lower()


 def _get_pip_main():
@@ -49,6 +59,373 @@ def _cleanup_added_root_handlers(original_handlers: list[logging.Handler]) -> No
                handler.close()


+def _prepend_sys_path(path: str) -> None:
+    normalized_target = os.path.realpath(path)
+    sys.path[:] = [
+        item for item in sys.path if os.path.realpath(item) != normalized_target
+    ]
+    sys.path.insert(0, normalized_target)
+
+
+def _module_exists_in_site_packages(module_name: str, site_packages_path: str) -> bool:
+    base_path = os.path.join(site_packages_path, *module_name.split("."))
+    package_init = os.path.join(base_path, "__init__.py")
+    module_file = f"{base_path}.py"
+    return os.path.isfile(package_init) or os.path.isfile(module_file)
+
+
+def _is_module_loaded_from_site_packages(
+    module_name: str,
+    site_packages_path: str,
+) -> bool:
+    module = sys.modules.get(module_name)
+    if module is None:
+        try:
+            module = importlib.import_module(module_name)
+        except Exception:
+            return False
+
+    module_file = getattr(module, "__file__", None)
+    if not module_file:
+        return False
+
+    module_path = os.path.realpath(module_file)
+    site_packages_real = os.path.realpath(site_packages_path)
+    try:
+        return (
+            os.path.commonpath([module_path, site_packages_real]) == site_packages_real
+        )
+    except ValueError:
+        return False
+
+
+def _extract_requirement_name(raw_requirement: str) -> str | None:
+    line = raw_requirement.split("#", 1)[0].strip()
+    if not line:
+        return None
+    if line.startswith(("-r", "--requirement", "-c", "--constraint")):
+        return None
+    if line.startswith("-"):
+        return None
+
+    egg_match = re.search(r"#egg=([A-Za-z0-9_.-]+)", raw_requirement)
+    if egg_match:
+        return _canonicalize_distribution_name(egg_match.group(1))
+
+    candidate = re.split(r"[<>=!~;\s\[]", line, maxsplit=1)[0].strip()
+    if not candidate:
+        return None
+    return _canonicalize_distribution_name(candidate)
+
+
+def _extract_requirement_names(requirements_path: str) -> set[str]:
+    names: set[str] = set()
+    try:
+        with open(requirements_path, encoding="utf-8") as requirements_file:
+            for line in requirements_file:
+                requirement_name = _extract_requirement_name(line)
+                if requirement_name:
+                    names.add(requirement_name)
+    except Exception as exc:
+        logger.warning("读取依赖文件失败，跳过冲突检测: %s", exc)
+    return names
+
+
+def _extract_top_level_modules(
+    distribution: importlib_metadata.Distribution,
+) -> set[str]:
+    try:
+        text = distribution.read_text("top_level.txt") or ""
+    except Exception:
+        return set()
+
+    modules: set[str] = set()
+    for line in text.splitlines():
+        candidate = line.strip()
+        if not candidate or candidate.startswith("#"):
+            continue
+        modules.add(candidate)
+    return modules
+
+
+def _collect_candidate_modules(
+    requirement_names: set[str],
+    site_packages_path: str,
+) -> set[str]:
+    by_name: dict[str, list[importlib_metadata.Distribution]] = {}
+    try:
+        for distribution in importlib_metadata.distributions(path=[site_packages_path]):
+            distribution_name = distribution.metadata.get("Name")
+            if not distribution_name:
+                continue
+            canonical_name = _canonicalize_distribution_name(distribution_name)
+            by_name.setdefault(canonical_name, []).append(distribution)
+    except Exception as exc:
+        logger.warning("读取 site-packages 元数据失败，使用回退模块名: %s", exc)
+
+    expanded_requirement_names: set[str] = set()
+    pending = deque(requirement_names)
+    while pending:
+        requirement_name = pending.popleft()
+        if requirement_name in expanded_requirement_names:
+            continue
+        expanded_requirement_names.add(requirement_name)
+
+        for distribution in by_name.get(requirement_name, []):
+            for dependency_line in distribution.requires or []:
+                dependency_name = _extract_requirement_name(dependency_line)
+                if not dependency_name:
+                    continue
+                if dependency_name in expanded_requirement_names:
+                    continue
+                pending.append(dependency_name)
+
+    candidates: set[str] = set()
+    for requirement_name in expanded_requirement_names:
+        matched_distributions = by_name.get(requirement_name, [])
+        modules_for_requirement: set[str] = set()
+        for distribution in matched_distributions:
+            modules_for_requirement.update(_extract_top_level_modules(distribution))
+
+        if modules_for_requirement:
+            candidates.update(modules_for_requirement)
+            continue
+
+        fallback_module_name = requirement_name.replace("-", "_")
+        if fallback_module_name:
+            candidates.add(fallback_module_name)
+
+    return candidates
+
+
+def _ensure_preferred_modules(
+    module_names: set[str],
+    site_packages_path: str,
+) -> None:
+    unresolved_prefer_reasons = _prefer_modules_from_site_packages(
+        module_names, site_packages_path
+    )
+
+    unresolved_modules: list[str] = []
+    for module_name in sorted(module_names):
+        if not _module_exists_in_site_packages(module_name, site_packages_path):
+            continue
+        if _is_module_loaded_from_site_packages(module_name, site_packages_path):
+            continue
+
+        failure_reason = unresolved_prefer_reasons.get(module_name)
+        if failure_reason:
+            unresolved_modules.append(f"{module_name} -> {failure_reason}")
+            continue
+
+        loaded_module = sys.modules.get(module_name)
+        loaded_from = getattr(loaded_module, "__file__", "unknown")
+        unresolved_modules.append(f"{module_name} -> {loaded_from}")
+
+    if unresolved_modules:
+        conflict_message = (
+            "检测到插件依赖与当前运行时发生冲突，无法安全加载该插件。"
+            f"冲突模块: {', '.join(unresolved_modules)}"
+        )
+        raise RuntimeError(conflict_message)
+
+
+def _prefer_module_from_site_packages(
+    module_name: str, site_packages_path: str
+) -> bool:
+    with _SITE_PACKAGES_IMPORT_LOCK:
+        base_path = os.path.join(site_packages_path, *module_name.split("."))
+        package_init = os.path.join(base_path, "__init__.py")
+        module_file = f"{base_path}.py"
+
+        module_location = None
+        submodule_search_locations = None
+
+        if os.path.isfile(package_init):
+            module_location = package_init
+            submodule_search_locations = [os.path.dirname(package_init)]
+        elif os.path.isfile(module_file):
+            module_location = module_file
+        else:
+            return False
+
+        spec = importlib.util.spec_from_file_location(
+            module_name,
+            module_location,
+            submodule_search_locations=submodule_search_locations,
+        )
+        if spec is None or spec.loader is None:
+            return False
+
+        matched_keys = [
+            key
+            for key in list(sys.modules.keys())
+            if key == module_name or key.startswith(f"{module_name}.")
+        ]
+        original_modules = {key: sys.modules[key] for key in matched_keys}
+
+        try:
+            for key in matched_keys:
+                sys.modules.pop(key, None)
+
+            module = importlib.util.module_from_spec(spec)
+            sys.modules[module_name] = module
+            spec.loader.exec_module(module)
+
+            if "." in module_name:
+                parent_name, child_name = module_name.rsplit(".", 1)
+                parent_module = sys.modules.get(parent_name)
+                if parent_module is not None:
+                    setattr(parent_module, child_name, module)
+
+            logger.info(
+                "Loaded %s from plugin site-packages: %s",
+                module_name,
+                module_location,
+            )
+            return True
+        except Exception:
+            failed_keys = [
+                key
+                for key in list(sys.modules.keys())
+                if key == module_name or key.startswith(f"{module_name}.")
+            ]
+            for key in failed_keys:
+                sys.modules.pop(key, None)
+            sys.modules.update(original_modules)
+            raise
+
+
+def _extract_conflicting_module_name(exc: Exception) -> str | None:
+    if isinstance(exc, ModuleNotFoundError):
+        missing_name = getattr(exc, "name", None)
+        if missing_name:
+            return missing_name.split(".", 1)[0]
+
+    message = str(exc)
+    from_match = re.search(r"from '([A-Za-z0-9_.]+)'", message)
+    if from_match:
+        return from_match.group(1).split(".", 1)[0]
+
+    no_module_match = re.search(r"No module named '([A-Za-z0-9_.]+)'", message)
+    if no_module_match:
+        return no_module_match.group(1).split(".", 1)[0]
+
+    return None
+
+
+def _prefer_module_with_dependency_recovery(
+    module_name: str,
+    site_packages_path: str,
+    max_attempts: int = 3,
+) -> bool:
+    recovered_dependencies: set[str] = set()
+
+    for _ in range(max_attempts):
+        try:
+            return _prefer_module_from_site_packages(module_name, site_packages_path)
+        except Exception as exc:
+            dependency_name = _extract_conflicting_module_name(exc)
+            if (
+                not dependency_name
+                or dependency_name == module_name
+                or dependency_name in recovered_dependencies
+            ):
+                raise
+
+            recovered_dependencies.add(dependency_name)
+            recovered = _prefer_module_from_site_packages(
+                dependency_name,
+                site_packages_path,
+            )
+            if not recovered:
+                raise
+            logger.info(
+                "Recovered dependency %s while preferring %s from plugin site-packages.",
+                dependency_name,
+                module_name,
+            )
+
+    return False
+
+
+def _prefer_modules_from_site_packages(
+    module_names: set[str],
+    site_packages_path: str,
+) -> dict[str, str]:
+    pending_modules = sorted(module_names)
+    unresolved_reasons: dict[str, str] = {}
+    max_rounds = max(2, min(6, len(pending_modules) + 1))
+
+    for _ in range(max_rounds):
+        if not pending_modules:
+            break
+
+        next_round_pending: list[str] = []
+        round_progress = False
+
+        for module_name in pending_modules:
+            try:
+                loaded = _prefer_module_with_dependency_recovery(
+                    module_name,
+                    site_packages_path,
+                )
+            except Exception as exc:
+                unresolved_reasons[module_name] = str(exc)
+                next_round_pending.append(module_name)
+                continue
+
+            unresolved_reasons.pop(module_name, None)
+            if loaded:
+                round_progress = True
+            else:
+                logger.debug(
+                    "Module %s not found in plugin site-packages: %s",
+                    module_name,
+                    site_packages_path,
+                )
+
+        if not next_round_pending:
+            pending_modules = []
+            break
+
+        if not round_progress and len(next_round_pending) == len(pending_modules):
+            pending_modules = next_round_pending
+            break
+
+        pending_modules = next_round_pending
+
+    final_unresolved = {
+        module_name: unresolved_reasons.get(module_name, "unknown import error")
+        for module_name in pending_modules
+    }
+    for module_name, reason in final_unresolved.items():
+        logger.warning(
+            "Failed to prefer module %s from plugin site-packages: %s",
+            module_name,
+            reason,
+        )
+
+    return final_unresolved
+
+
+def _ensure_plugin_dependencies_preferred(
+    target_site_packages: str,
+    requested_requirements: set[str],
+) -> None:
+    if not requested_requirements:
+        return
+
+    candidate_modules = _collect_candidate_modules(
+        requested_requirements,
+        target_site_packages,
+    )
+    if not candidate_modules:
+        return
+
+    _ensure_preferred_modules(candidate_modules, target_site_packages)
+
+
 def _get_loader_for_package(package: object) -> object | None:
    loader = getattr(package, "__loader__", None)
    if loader is not None:
@@ -73,7 +450,7 @@ def _try_register_distlib_finder(
        return False

    try:
-        register_finder(loader_type, resource_finder)
+        register_finder(loader, resource_finder)
    except Exception as exc:
        logger.warning(
            "Failed to patch pip distlib finder for loader %s (%s): %s",
@@ -165,10 +542,15 @@ class PipInstaller:
        mirror: str | None = None,
    ) -> None:
        args = ["install"]
+        requested_requirements: set[str] = set()
        if package_name:
            args.append(package_name)
+            requirement_name = _extract_requirement_name(package_name)
+            if requirement_name:
+                requested_requirements.add(requirement_name)
        elif requirements_path:
            args.extend(["-r", requirements_path])
+            requested_requirements = _extract_requirement_names(requirements_path)

        index_url = mirror or self.pypi_index_url or "https://pypi.org/simple"
        args.extend(["--trusted-host", "mirrors.aliyun.com", "-i", index_url])
@@ -177,7 +559,9 @@ class PipInstaller:
        if is_packaged_electron_runtime():
            target_site_packages = get_astrbot_site_packages_path()
            os.makedirs(target_site_packages, exist_ok=True)
+            _prepend_sys_path(target_site_packages)
            args.extend(["--target", target_site_packages])
+            args.extend(["--upgrade", "--force-reinstall"])

        if self.pip_install_arg:
            args.extend(self.pip_install_arg.split())
@@ -188,8 +572,32 @@ class PipInstaller:
        if result_code != 0:
            raise Exception(f"安装失败，错误码：{result_code}")

-        if target_site_packages and target_site_packages not in sys.path:
-            sys.path.insert(0, target_site_packages)
+        if target_site_packages:
+            _prepend_sys_path(target_site_packages)
+            _ensure_plugin_dependencies_preferred(
+                target_site_packages,
+                requested_requirements,
+            )
+        importlib.invalidate_caches()
+
+    def prefer_installed_dependencies(self, requirements_path: str) -> None:
+        """优先使用已安装在插件 site-packages 中的依赖，不执行安装。"""
+        if not is_packaged_electron_runtime():
+            return
+
+        target_site_packages = get_astrbot_site_packages_path()
+        if not os.path.isdir(target_site_packages):
+            return
+
+        requested_requirements = _extract_requirement_names(requirements_path)
+        if not requested_requirements:
+            return
+
+        _prepend_sys_path(target_site_packages)
+        _ensure_plugin_dependencies_preferred(
+            target_site_packages,
+            requested_requirements,
+        )
        importlib.invalidate_caches()

    async def _run_pip_in_process(self, args: list[str]) -> int:
@@ -0,0 +1,8 @@
+from __future__ import annotations
+
+from .extractor import extract_quoted_message_images, extract_quoted_message_text
+
+__all__ = [
+    "extract_quoted_message_text",
+    "extract_quoted_message_images",
+]
@@ -0,0 +1,505 @@
+from __future__ import annotations
+
+import json
+import re
+from typing import Any, TypedDict
+
+from astrbot.core.message.components import (
+    At,
+    AtAll,
+    File,
+    Forward,
+    Image,
+    Node,
+    Nodes,
+    Plain,
+    Reply,
+    Video,
+)
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
+from astrbot.core.utils.string_utils import normalize_and_dedupe_strings
+
+from .image_refs import looks_like_image_file_name, normalize_file_like_url
+from .settings import SETTINGS, QuotedMessageParserSettings
+
+_FORWARD_PLACEHOLDER_PATTERN = re.compile(
+    r"^(?:[\(\[]?[^\]:\)]*[\)\]]?\s*:\s*)?\[(?:forward message|转发消息|合并转发)\]$",
+    flags=re.IGNORECASE,
+)
+
+
+class ParsedOneBotPayload(TypedDict):
+    text: str | None
+    forward_ids: list[str]
+    image_refs: list[str]
+
+
+def _build_parsed_payload(
+    text: str | None,
+    forward_ids: list[str] | None = None,
+    image_refs: list[str] | None = None,
+) -> ParsedOneBotPayload:
+    return {
+        "text": text,
+        "forward_ids": forward_ids or [],
+        "image_refs": image_refs or [],
+    }
+
+
+def _join_text_parts(parts: list[str]) -> str | None:
+    text = "".join(parts).strip()
+    return text or None
+
+
+def _find_first_reply_component(event: AstrMessageEvent) -> Reply | None:
+    for comp in event.message_obj.message:
+        if isinstance(comp, Reply):
+            return comp
+    return None
+
+
+def _is_forward_placeholder_only_text(text: str | None) -> bool:
+    if not isinstance(text, str):
+        return False
+    lines = [line.strip() for line in text.splitlines() if line.strip()]
+    if not lines:
+        return False
+    return all(_FORWARD_PLACEHOLDER_PATTERN.match(line) for line in lines)
+
+
+def _extract_image_refs_from_component_chain(
+    chain: list[Any] | None,
+    *,
+    depth: int = 0,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> list[str]:
+    if not isinstance(chain, list) or depth > settings.max_component_chain_depth:
+        return []
+
+    image_refs: list[str] = []
+    for seg in chain:
+        if isinstance(seg, Image):
+            for candidate in (seg.url, seg.file, seg.path):
+                if isinstance(candidate, str) and candidate.strip():
+                    image_refs.append(candidate.strip())
+                    break
+        elif isinstance(seg, Reply):
+            image_refs.extend(
+                _extract_image_refs_from_reply_component(
+                    seg,
+                    depth=depth + 1,
+                    settings=settings,
+                )
+            )
+        elif isinstance(seg, Node):
+            image_refs.extend(
+                _extract_image_refs_from_component_chain(
+                    seg.content,
+                    depth=depth + 1,
+                    settings=settings,
+                )
+            )
+        elif isinstance(seg, Nodes):
+            for node in seg.nodes:
+                image_refs.extend(
+                    _extract_image_refs_from_component_chain(
+                        node.content,
+                        depth=depth + 1,
+                        settings=settings,
+                    )
+                )
+
+    return normalize_and_dedupe_strings(image_refs)
+
+
+def _extract_text_from_component_chain(
+    chain: list[Any] | None,
+    *,
+    depth: int = 0,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> str | None:
+    if not isinstance(chain, list) or depth > settings.max_component_chain_depth:
+        return None
+
+    parts: list[str] = []
+    for seg in chain:
+        if isinstance(seg, Plain):
+            if seg.text:
+                parts.append(seg.text)
+        elif isinstance(seg, At):
+            if seg.name:
+                parts.append(f"@{seg.name}")
+            elif seg.qq:
+                parts.append(f"@{seg.qq}")
+        elif isinstance(seg, AtAll):
+            parts.append("@all")
+        elif isinstance(seg, Image):
+            parts.append("[Image]")
+        elif isinstance(seg, Video):
+            parts.append("[Video]")
+        elif isinstance(seg, File):
+            file_name = seg.name or "file"
+            parts.append(f"[File:{file_name}]")
+        elif isinstance(seg, Forward):
+            parts.append("[Forward Message]")
+        elif isinstance(seg, Reply):
+            nested = _extract_text_from_reply_component(
+                seg,
+                depth=depth + 1,
+                settings=settings,
+            )
+            if nested:
+                parts.append(nested)
+        elif isinstance(seg, Node):
+            node_sender = seg.name or seg.uin or "Unknown User"
+            node_text = _extract_text_from_component_chain(
+                seg.content,
+                depth=depth + 1,
+                settings=settings,
+            )
+            if node_text:
+                parts.append(f"{node_sender}: {node_text}")
+        elif isinstance(seg, Nodes):
+            for node in seg.nodes:
+                node_sender = node.name or node.uin or "Unknown User"
+                node_text = _extract_text_from_component_chain(
+                    node.content,
+                    depth=depth + 1,
+                    settings=settings,
+                )
+                if node_text:
+                    parts.append(f"{node_sender}: {node_text}")
+
+    return _join_text_parts(parts)
+
+
+def _extract_image_refs_from_reply_component(
+    reply: Reply,
+    *,
+    depth: int = 0,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> list[str]:
+    for attr in ("chain", "message", "origin", "content"):
+        payload = getattr(reply, attr, None)
+        image_refs = _extract_image_refs_from_component_chain(
+            payload,
+            depth=depth,
+            settings=settings,
+        )
+        if image_refs:
+            return image_refs
+    return []
+
+
+def _extract_text_from_reply_component(
+    reply: Reply,
+    *,
+    depth: int = 0,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> str | None:
+    for attr in ("chain", "message", "origin", "content"):
+        payload = getattr(reply, attr, None)
+        text = _extract_text_from_component_chain(
+            payload,
+            depth=depth,
+            settings=settings,
+        )
+        if text:
+            return text
+
+    if reply.message_str and reply.message_str.strip():
+        return reply.message_str.strip()
+    return None
+
+
+def _unwrap_onebot_data(payload: Any) -> dict[str, Any]:
+    if not isinstance(payload, dict):
+        return {}
+    data = payload.get("data")
+    if isinstance(data, dict):
+        return data
+    return payload
+
+
+def _extract_text_from_multimsg_json(raw_json: str) -> str | None:
+    try:
+        parsed = json.loads(raw_json)
+    except Exception:
+        return None
+
+    if not isinstance(parsed, dict):
+        return None
+    if parsed.get("app") != "com.tencent.multimsg":
+        return None
+    config = parsed.get("config")
+    if not isinstance(config, dict):
+        return None
+    if config.get("forward") != 1:
+        return None
+
+    meta = parsed.get("meta")
+    if not isinstance(meta, dict):
+        return None
+    detail = meta.get("detail")
+    if not isinstance(detail, dict):
+        return None
+    news_items = detail.get("news")
+    if not isinstance(news_items, list):
+        return None
+
+    texts: list[str] = []
+    for item in news_items:
+        if not isinstance(item, dict):
+            continue
+        text_content = item.get("text")
+        if not isinstance(text_content, str):
+            continue
+        cleaned = text_content.strip().replace("[图片]", "").strip()
+        if cleaned:
+            texts.append(cleaned)
+
+    return "\n".join(texts).strip() or None
+
+
+def _parse_onebot_segments(
+    segments: list[Any],
+    *,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> ParsedOneBotPayload:
+    text_parts: list[str] = []
+    forward_ids: list[str] = []
+    image_refs: list[str] = []
+
+    for seg in segments:
+        if not isinstance(seg, dict):
+            continue
+
+        seg_type = seg.get("type")
+        seg_data = seg.get("data", {}) if isinstance(seg.get("data"), dict) else {}
+
+        if seg_type in ("text", "plain"):
+            text = seg_data.get("text")
+            if isinstance(text, str) and text:
+                text_parts.append(text)
+        elif seg_type == "image":
+            text_parts.append("[Image]")
+            candidate = seg_data.get("url") or seg_data.get("file")
+            if isinstance(candidate, str) and candidate.strip():
+                image_refs.append(candidate.strip())
+        elif seg_type == "video":
+            text_parts.append("[Video]")
+        elif seg_type == "file":
+            file_name = (
+                seg_data.get("name")
+                or seg_data.get("file_name")
+                or seg_data.get("file")
+                or "file"
+            )
+            text_parts.append(f"[File:{file_name}]")
+            candidate_url = seg_data.get("url")
+            if (
+                isinstance(candidate_url, str)
+                and candidate_url.strip()
+                and looks_like_image_file_name(normalize_file_like_url(candidate_url))
+            ):
+                image_refs.append(candidate_url.strip())
+            candidate_file = seg_data.get("file")
+            if (
+                isinstance(candidate_file, str)
+                and candidate_file.strip()
+                and looks_like_image_file_name(
+                    normalize_file_like_url(
+                        seg_data.get("name")
+                        or seg_data.get("file_name")
+                        or candidate_file
+                    )
+                )
+            ):
+                image_refs.append(candidate_file.strip())
+        elif seg_type in ("forward", "forward_msg", "nodes"):
+            fid = seg_data.get("id") or seg_data.get("message_id")
+            if isinstance(fid, (str, int)) and str(fid):
+                forward_ids.append(str(fid))
+            else:
+                nested_nodes = seg_data.get("content")
+                nested_text, nested_forward_ids, nested_images = (
+                    _extract_text_forward_ids_and_images_from_forward_nodes(
+                        nested_nodes if isinstance(nested_nodes, list) else [],
+                        depth=1,
+                        settings=settings,
+                    )
+                )
+                if nested_text:
+                    text_parts.append(nested_text)
+                if nested_forward_ids:
+                    forward_ids.extend(nested_forward_ids)
+                if nested_images:
+                    image_refs.extend(nested_images)
+        elif seg_type == "json":
+            raw_json = seg_data.get("data")
+            if isinstance(raw_json, str) and raw_json.strip():
+                raw_json = raw_json.replace("&#44;", ",")
+                multimsg_text = _extract_text_from_multimsg_json(raw_json)
+                if multimsg_text:
+                    text_parts.append(multimsg_text)
+
+    return _build_parsed_payload(
+        _join_text_parts(text_parts),
+        forward_ids,
+        normalize_and_dedupe_strings(image_refs),
+    )
+
+
+def _extract_text_forward_ids_and_images_from_forward_nodes(
+    nodes: list[Any],
+    *,
+    depth: int = 0,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> tuple[str | None, list[str], list[str]]:
+    if not isinstance(nodes, list) or depth > settings.max_forward_node_depth:
+        return None, [], []
+
+    texts: list[str] = []
+    forward_ids: list[str] = []
+    image_refs: list[str] = []
+    indent = "  " * depth
+
+    for node in nodes:
+        if not isinstance(node, dict):
+            continue
+
+        sender = node.get("sender") if isinstance(node.get("sender"), dict) else {}
+        sender_name = (
+            sender.get("nickname")
+            or sender.get("card")
+            or sender.get("user_id")
+            or "Unknown User"
+        )
+
+        raw_content = node.get("message") or node.get("content") or []
+        chain: list[Any] = []
+        if isinstance(raw_content, list):
+            chain = raw_content
+        elif isinstance(raw_content, str):
+            raw_content = raw_content.strip()
+            if raw_content:
+                try:
+                    parsed = json.loads(raw_content)
+                except Exception:
+                    parsed = None
+                if isinstance(parsed, list):
+                    chain = parsed
+                else:
+                    chain = [{"type": "text", "data": {"text": raw_content}}]
+
+        parsed_segments = _parse_onebot_segments(chain, settings=settings)
+        node_text = parsed_segments["text"]
+        node_forward_ids = parsed_segments["forward_ids"]
+        node_images = parsed_segments["image_refs"]
+        if node_text:
+            texts.append(f"{indent}{sender_name}: {node_text}")
+        if node_forward_ids:
+            forward_ids.extend(node_forward_ids)
+        if node_images:
+            image_refs.extend(node_images)
+
+    return (
+        "\n".join(texts).strip() or None,
+        normalize_and_dedupe_strings(forward_ids),
+        normalize_and_dedupe_strings(image_refs),
+    )
+
+
+def _parse_onebot_get_msg_payload(
+    payload: dict[str, Any],
+    *,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> ParsedOneBotPayload:
+    data = _unwrap_onebot_data(payload)
+    segments = data.get("message") or data.get("messages")
+    if isinstance(segments, list):
+        return _parse_onebot_segments(segments, settings=settings)
+
+    text: str | None = None
+    if isinstance(segments, str) and segments.strip():
+        text = segments.strip()
+    else:
+        raw = data.get("raw_message")
+        if isinstance(raw, str) and raw.strip():
+            text = raw.strip()
+    return _build_parsed_payload(text)
+
+
+def _parse_onebot_get_forward_payload(
+    payload: dict[str, Any],
+    *,
+    settings: QuotedMessageParserSettings = SETTINGS,
+) -> ParsedOneBotPayload:
+    data = _unwrap_onebot_data(payload)
+    nodes = (
+        data.get("messages")
+        or data.get("message")
+        or data.get("nodes")
+        or data.get("nodeList")
+    )
+    if not isinstance(nodes, list):
+        return _build_parsed_payload(None)
+
+    text, forward_ids, image_refs = (
+        _extract_text_forward_ids_and_images_from_forward_nodes(
+            nodes,
+            settings=settings,
+        )
+    )
+    return _build_parsed_payload(text, forward_ids, image_refs)
+
+
+class ReplyChainParser:
+    def __init__(self, settings: QuotedMessageParserSettings = SETTINGS):
+        self._settings = settings
+
+    @staticmethod
+    def find_first_reply_component(event: AstrMessageEvent) -> Reply | None:
+        return _find_first_reply_component(event)
+
+    @staticmethod
+    def is_forward_placeholder_only_text(text: str | None) -> bool:
+        return _is_forward_placeholder_only_text(text)
+
+    def extract_text_from_reply_component(
+        self,
+        reply: Reply,
+        *,
+        depth: int = 0,
+    ) -> str | None:
+        return _extract_text_from_reply_component(
+            reply,
+            depth=depth,
+            settings=self._settings,
+        )
+
+    def extract_image_refs_from_reply_component(
+        self,
+        reply: Reply,
+        *,
+        depth: int = 0,
+    ) -> list[str]:
+        return _extract_image_refs_from_reply_component(
+            reply,
+            depth=depth,
+            settings=self._settings,
+        )
+
+
+class OneBotPayloadParser:
+    def __init__(self, settings: QuotedMessageParserSettings = SETTINGS):
+        self._settings = settings
+
+    def parse_get_msg_payload(self, payload: dict[str, Any]) -> ParsedOneBotPayload:
+        return _parse_onebot_get_msg_payload(payload, settings=self._settings)
+
+    def parse_get_forward_payload(
+        self,
+        payload: dict[str, Any],
+    ) -> ParsedOneBotPayload:
+        return _parse_onebot_get_forward_payload(payload, settings=self._settings)
@@ -0,0 +1,211 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+from astrbot import logger
+from astrbot.core.message.components import Reply
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
+from astrbot.core.utils.string_utils import normalize_and_dedupe_strings
+
+from .chain_parser import OneBotPayloadParser, ReplyChainParser
+from .image_resolver import ImageResolver
+from .onebot_client import OneBotClient
+from .settings import SETTINGS, QuotedMessageParserSettings
+
+
+async def _collect_text_and_images_from_forward_ids(
+    onebot_client: OneBotClient,
+    payload_parser: OneBotPayloadParser,
+    forward_ids: list[str],
+    *,
+    max_fetch: int,
+) -> tuple[list[str], list[str]]:
+    texts: list[str] = []
+    image_refs: list[str] = []
+    pending: list[str] = []
+    seen: set[str] = set()
+
+    for fid in forward_ids:
+        if not isinstance(fid, str):
+            continue
+        cleaned = fid.strip()
+        if cleaned:
+            pending.append(cleaned)
+
+    fetch_count = 0
+    while pending and fetch_count < max_fetch:
+        current_id = pending.pop(0)
+        if current_id in seen:
+            continue
+        seen.add(current_id)
+        fetch_count += 1
+
+        forward_payload = await onebot_client.get_forward_msg(current_id)
+        if not forward_payload:
+            continue
+
+        parsed = payload_parser.parse_get_forward_payload(forward_payload)
+        if parsed["text"]:
+            texts.append(parsed["text"])
+        if parsed["image_refs"]:
+            image_refs.extend(parsed["image_refs"])
+        for nested_id in parsed["forward_ids"]:
+            if nested_id not in seen:
+                pending.append(nested_id)
+
+    if pending:
+        logger.warning(
+            "quoted_message_parser: stop fetching nested forward messages after %d hops",
+            max_fetch,
+        )
+
+    return texts, normalize_and_dedupe_strings(image_refs)
+
+
+@dataclass(slots=True)
+class QuotedMessageContent:
+    embedded_text: str | None
+    embedded_image_refs: list[str]
+    reply_id: str
+    direct_text: str | None
+    direct_image_refs: list[str]
+    forward_texts: list[str]
+    forward_image_refs: list[str]
+
+
+class QuotedMessageExtractor:
+    def __init__(
+        self,
+        event: AstrMessageEvent,
+        settings: QuotedMessageParserSettings = SETTINGS,
+    ):
+        self._event = event
+        self._settings = settings
+        self._reply_parser = ReplyChainParser(settings=settings)
+        self._payload_parser = OneBotPayloadParser(settings=settings)
+        self._client = OneBotClient(event, settings=settings)
+        self._image_resolver = ImageResolver(event, self._client)
+
+    async def _fetch_quoted_content(
+        self,
+        reply_component: Reply | None = None,
+        *,
+        fetch_remote: bool,
+    ) -> QuotedMessageContent | None:
+        reply = reply_component or self._reply_parser.find_first_reply_component(
+            self._event
+        )
+        if not reply:
+            return None
+
+        embedded_text = self._reply_parser.extract_text_from_reply_component(reply)
+        embedded_image_refs = list(
+            self._reply_parser.extract_image_refs_from_reply_component(reply)
+        )
+
+        reply_id = getattr(reply, "id", None)
+        reply_id_str = str(reply_id).strip() if reply_id is not None else ""
+        if not fetch_remote or not reply_id_str:
+            return QuotedMessageContent(
+                embedded_text=embedded_text,
+                embedded_image_refs=embedded_image_refs,
+                reply_id=reply_id_str,
+                direct_text=None,
+                direct_image_refs=[],
+                forward_texts=[],
+                forward_image_refs=[],
+            )
+
+        msg_payload = await self._client.get_msg(reply_id_str)
+        if not msg_payload:
+            return QuotedMessageContent(
+                embedded_text=embedded_text,
+                embedded_image_refs=embedded_image_refs,
+                reply_id=reply_id_str,
+                direct_text=None,
+                direct_image_refs=[],
+                forward_texts=[],
+                forward_image_refs=[],
+            )
+
+        parsed = self._payload_parser.parse_get_msg_payload(msg_payload)
+        forward_texts, forward_images = await _collect_text_and_images_from_forward_ids(
+            self._client,
+            self._payload_parser,
+            parsed["forward_ids"],
+            max_fetch=self._settings.max_forward_fetch,
+        )
+        return QuotedMessageContent(
+            embedded_text=embedded_text,
+            embedded_image_refs=embedded_image_refs,
+            reply_id=reply_id_str,
+            direct_text=parsed["text"],
+            direct_image_refs=list(parsed["image_refs"]),
+            forward_texts=forward_texts,
+            forward_image_refs=forward_images,
+        )
+
+    async def text(self, reply_component: Reply | None = None) -> str | None:
+        embedded_content = await self._fetch_quoted_content(
+            reply_component,
+            fetch_remote=False,
+        )
+        if not embedded_content:
+            return None
+
+        if (
+            embedded_content.embedded_text
+            and not self._reply_parser.is_forward_placeholder_only_text(
+                embedded_content.embedded_text
+            )
+        ):
+            return embedded_content.embedded_text
+
+        if not embedded_content.reply_id:
+            return embedded_content.embedded_text
+
+        fetched_content = await self._fetch_quoted_content(
+            reply_component,
+            fetch_remote=True,
+        )
+        if not fetched_content:
+            return embedded_content.embedded_text
+
+        text_parts: list[str] = []
+        if fetched_content.direct_text:
+            text_parts.append(fetched_content.direct_text)
+        text_parts.extend(fetched_content.forward_texts)
+
+        return "\n".join(text_parts).strip() or embedded_content.embedded_text
+
+    async def images(self, reply_component: Reply | None = None) -> list[str]:
+        content = await self._fetch_quoted_content(reply_component, fetch_remote=True)
+        if not content:
+            return []
+
+        image_refs: list[str] = []
+        image_refs.extend(content.embedded_image_refs)
+        image_refs.extend(content.direct_image_refs)
+        image_refs.extend(content.forward_image_refs)
+
+        return await self._image_resolver.resolve_for_llm(image_refs)
+
+
+async def extract_quoted_message_text(
+    event: AstrMessageEvent,
+    reply_component: Reply | None = None,
+    settings: QuotedMessageParserSettings | None = None,
+) -> str | None:
+    return await QuotedMessageExtractor(event, settings=settings or SETTINGS).text(
+        reply_component
+    )
+
+
+async def extract_quoted_message_images(
+    event: AstrMessageEvent,
+    reply_component: Reply | None = None,
+    settings: QuotedMessageParserSettings | None = None,
+) -> list[str]:
+    return await QuotedMessageExtractor(event, settings=settings or SETTINGS).images(
+        reply_component
+    )
@@ -0,0 +1,94 @@
+from __future__ import annotations
+
+import os
+from urllib.parse import urlsplit
+
+IMAGE_EXTENSIONS = {
+    ".jpg",
+    ".jpeg",
+    ".png",
+    ".webp",
+    ".bmp",
+    ".tif",
+    ".tiff",
+    ".gif",
+}
+
+
+def normalize_file_like_url(path: str | None) -> str | None:
+    if path is None:
+        return None
+    if not isinstance(path, str):
+        return None
+    if "?" not in path and "#" not in path:
+        return path
+    try:
+        split = urlsplit(path)
+    except Exception:
+        return path
+    return split.path or path
+
+
+def looks_like_image_file_name(name: str) -> bool:
+    normalized_name = normalize_file_like_url(name)
+    if not isinstance(normalized_name, str) or not normalized_name.strip():
+        return False
+    _, ext = os.path.splitext(normalized_name.strip().lower())
+    return ext in IMAGE_EXTENSIONS
+
+
+def convert_data_image_to_base64_ref(image_ref: str) -> str | None:
+    if not isinstance(image_ref, str):
+        return None
+    value = image_ref.strip()
+    if not value:
+        return None
+    lower_value = value.lower()
+    if not lower_value.startswith("data:image/"):
+        return None
+
+    comma_index = value.find(",")
+    if comma_index <= 0:
+        return None
+    header = value[:comma_index].lower()
+    payload = value[comma_index + 1 :].strip()
+    if ";base64" not in header or not payload:
+        return None
+    return f"base64://{payload}"
+
+
+def get_existing_local_path(value: str) -> str | None:
+    lower_value = value.lower()
+    if lower_value.startswith("file://"):
+        file_path = value[7:]
+        if file_path.startswith("/") and len(file_path) > 3 and file_path[2] == ":":
+            file_path = file_path[1:]
+        if file_path and os.path.exists(file_path):
+            return os.path.abspath(file_path)
+        return None
+    if os.path.exists(value):
+        return os.path.abspath(value)
+    return None
+
+
+def normalize_image_ref(image_ref: str) -> str | None:
+    if not isinstance(image_ref, str):
+        return None
+    value = image_ref.strip()
+    if not value:
+        return None
+    lower_value = value.lower()
+
+    if lower_value.startswith(("http://", "https://")):
+        return value
+    if lower_value.startswith("base64://"):
+        return value
+
+    data_image_ref = convert_data_image_to_base64_ref(value)
+    if data_image_ref:
+        return data_image_ref
+
+    local_path = get_existing_local_path(value)
+    if local_path and looks_like_image_file_name(local_path):
+        return local_path
+    return None
@@ -0,0 +1,130 @@
+from __future__ import annotations
+
+import os
+from typing import Any
+
+from astrbot import logger
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
+from astrbot.core.utils.string_utils import normalize_and_dedupe_strings
+
+from .image_refs import IMAGE_EXTENSIONS, get_existing_local_path, normalize_image_ref
+from .onebot_client import OneBotClient
+
+
+def _build_image_id_candidates(image_ref: str) -> list[str]:
+    candidates: list[str] = [image_ref]
+    base_name, ext = os.path.splitext(image_ref)
+    if ext and base_name and base_name not in candidates:
+        if ext.lower() in IMAGE_EXTENSIONS:
+            candidates.append(base_name)
+    return candidates
+
+
+def _build_image_resolve_actions(
+    event: AstrMessageEvent,
+    image_ref: str,
+) -> list[tuple[str, dict[str, Any]]]:
+    actions: list[tuple[str, dict[str, Any]]] = []
+    candidates = _build_image_id_candidates(image_ref)
+
+    for candidate in candidates:
+        actions.extend(
+            [
+                ("get_image", {"file": candidate}),
+                ("get_image", {"file_id": candidate}),
+                ("get_image", {"id": candidate}),
+                ("get_image", {"image": candidate}),
+                ("get_file", {"file_id": candidate}),
+                ("get_file", {"file": candidate}),
+            ]
+        )
+
+    try:
+        group_id = event.get_group_id()
+    except Exception:
+        group_id = None
+    group_id_value = group_id
+    if isinstance(group_id, str) and group_id.isdigit():
+        group_id_value = int(group_id)
+
+    if group_id_value:
+        for candidate in candidates:
+            actions.append(
+                (
+                    "get_group_file_url",
+                    {"group_id": group_id_value, "file_id": candidate},
+                )
+            )
+    for candidate in candidates:
+        actions.append(("get_private_file_url", {"file_id": candidate}))
+
+    return actions
+
+
+class ImageResolver:
+    def __init__(
+        self,
+        event: AstrMessageEvent,
+        onebot_client: OneBotClient | None = None,
+    ):
+        self._event = event
+        self._client = onebot_client or OneBotClient(event)
+
+    async def resolve_for_llm(self, image_refs: list[str]) -> list[str]:
+        resolved: list[str] = []
+        unresolved: list[str] = []
+
+        for image_ref in normalize_and_dedupe_strings(image_refs):
+            normalized = normalize_image_ref(image_ref)
+            if normalized:
+                resolved.append(normalized)
+            elif get_existing_local_path(image_ref):
+                # Drop non-image local paths instead of treating them as remote IDs.
+                logger.debug(
+                    "quoted_message_parser: skip non-image local path ref=%s",
+                    image_ref[:128],
+                )
+            else:
+                unresolved.append(image_ref)
+
+        for image_ref in unresolved:
+            resolved_ref = await self._resolve_one(image_ref)
+            if resolved_ref:
+                resolved.append(resolved_ref)
+
+        return normalize_and_dedupe_strings(resolved)
+
+    async def _resolve_one(self, image_ref: str) -> str | None:
+        resolved = normalize_image_ref(image_ref)
+        if resolved:
+            return resolved
+
+        actions = _build_image_resolve_actions(self._event, image_ref)
+        for action, params in actions:
+            data = await self._client.call(
+                action,
+                params,
+                warn_on_all_failed=False,
+                unwrap_data=True,
+            )
+            if not isinstance(data, dict):
+                continue
+
+            url = data.get("url")
+            if isinstance(url, str):
+                normalized = normalize_image_ref(url)
+                if normalized:
+                    return normalized
+
+            file_value = data.get("file")
+            if isinstance(file_value, str):
+                normalized = normalize_image_ref(file_value)
+                if normalized:
+                    return normalized
+
+        logger.warning(
+            "quoted_message_parser: failed to resolve quoted image ref=%s after %d actions",
+            image_ref[:128],
+            len(actions),
+        )
+        return None
@@ -0,0 +1,119 @@
+from __future__ import annotations
+
+from typing import Any
+
+from astrbot import logger
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
+
+from .settings import SETTINGS, QuotedMessageParserSettings
+
+
+def _unwrap_action_response(ret: dict[str, Any] | None) -> dict[str, Any]:
+    if not isinstance(ret, dict):
+        return {}
+    data = ret.get("data")
+    if isinstance(data, dict):
+        return data
+    return ret
+
+
+class OneBotClient:
+    def __init__(
+        self,
+        event: AstrMessageEvent,
+        settings: QuotedMessageParserSettings = SETTINGS,
+    ):
+        self._call_action = self._resolve_call_action(event)
+        self._settings = settings
+
+    @staticmethod
+    def _resolve_call_action(event: AstrMessageEvent):
+        bot = getattr(event, "bot", None)
+        api = getattr(bot, "api", None)
+        call_action = getattr(api, "call_action", None)
+        if not callable(call_action):
+            return None
+        return call_action
+
+    async def _call_action_try_params(
+        self,
+        action: str,
+        params_list: list[dict[str, Any]],
+        *,
+        warn_on_all_failed: bool | None = None,
+    ) -> dict[str, Any] | None:
+        if self._call_action is None:
+            return None
+        if warn_on_all_failed is None:
+            warn_on_all_failed = self._settings.warn_on_action_failure
+
+        last_error: Exception | None = None
+        last_params: dict[str, Any] | None = None
+        for params in params_list:
+            try:
+                result = await self._call_action(action, **params)
+                if isinstance(result, dict):
+                    return result
+            except Exception as exc:
+                last_error = exc
+                last_params = params
+                logger.debug(
+                    "quoted_message_parser: action %s failed with params %s: %s",
+                    action,
+                    {k: str(v)[:64] for k, v in params.items()},
+                    exc,
+                )
+        if warn_on_all_failed and last_error is not None:
+            logger.warning(
+                "quoted_message_parser: all attempts failed for action %s, "
+                "last_params=%s, error=%s",
+                action,
+                (
+                    {k: str(v)[:64] for k, v in last_params.items()}
+                    if isinstance(last_params, dict)
+                    else None
+                ),
+                last_error,
+            )
+        return None
+
+    async def call(
+        self,
+        action: str,
+        params: dict[str, Any],
+        *,
+        warn_on_all_failed: bool = False,
+        unwrap_data: bool = True,
+    ) -> dict[str, Any] | None:
+        ret = await self._call_action_try_params(
+            action,
+            [params],
+            warn_on_all_failed=warn_on_all_failed,
+        )
+        if not unwrap_data:
+            return ret
+        return _unwrap_action_response(ret)
+
+    async def _call_action_compat(
+        self,
+        action: str,
+        message_id: str | int,
+    ) -> dict[str, Any] | None:
+        message_id_str = str(message_id).strip()
+        if not message_id_str:
+            return None
+
+        params_list: list[dict[str, Any]] = [
+            {"message_id": message_id_str},
+            {"id": message_id_str},
+        ]
+        if message_id_str.isdigit():
+            int_id = int(message_id_str)
+            params_list.extend([{"message_id": int_id}, {"id": int_id}])
+        return await self._call_action_try_params(action, params_list)
+
+    async def get_msg(self, message_id: str | int) -> dict[str, Any] | None:
+        return await self._call_action_compat("get_msg", message_id)
+
+    async def get_forward_msg(self, forward_id: str | int) -> dict[str, Any] | None:
+        return await self._call_action_compat("get_forward_msg", forward_id)
@@ -0,0 +1,85 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+from dataclasses import dataclass
+from typing import Any
+
+_DEFAULT_MAX_COMPONENT_CHAIN_DEPTH = 4
+_DEFAULT_MAX_FORWARD_NODE_DEPTH = 6
+_DEFAULT_MAX_FORWARD_FETCH = 32
+
+
+def _read_int_mapping(
+    mapping: Mapping[str, Any],
+    key: str,
+    default: int,
+) -> int:
+    raw = mapping.get(key)
+    if raw is None:
+        return default
+    try:
+        value = int(raw)
+    except (TypeError, ValueError):
+        return default
+    if value <= 0:
+        return default
+    return value
+
+
+def _read_bool_mapping(
+    mapping: Mapping[str, Any],
+    key: str,
+    default: bool,
+) -> bool:
+    raw = mapping.get(key)
+    if raw is None:
+        return default
+    if isinstance(raw, bool):
+        return raw
+    if isinstance(raw, str):
+        lowered = raw.strip().lower()
+        if lowered in {"1", "true", "yes", "on"}:
+            return True
+        if lowered in {"0", "false", "no", "off"}:
+            return False
+    return default
+
+
+@dataclass(frozen=True)
+class QuotedMessageParserSettings:
+    max_component_chain_depth: int = _DEFAULT_MAX_COMPONENT_CHAIN_DEPTH
+    max_forward_node_depth: int = _DEFAULT_MAX_FORWARD_NODE_DEPTH
+    max_forward_fetch: int = _DEFAULT_MAX_FORWARD_FETCH
+    warn_on_action_failure: bool = False
+
+    def with_overrides(
+        self,
+        overrides: Mapping[str, Any] | None = None,
+    ) -> QuotedMessageParserSettings:
+        if not overrides:
+            return self
+        return QuotedMessageParserSettings(
+            max_component_chain_depth=_read_int_mapping(
+                overrides,
+                "max_component_chain_depth",
+                self.max_component_chain_depth,
+            ),
+            max_forward_node_depth=_read_int_mapping(
+                overrides,
+                "max_forward_node_depth",
+                self.max_forward_node_depth,
+            ),
+            max_forward_fetch=_read_int_mapping(
+                overrides,
+                "max_forward_fetch",
+                self.max_forward_fetch,
+            ),
+            warn_on_action_failure=_read_bool_mapping(
+                overrides,
+                "warn_on_action_failure",
+                self.warn_on_action_failure,
+            ),
+        )
+
+
+SETTINGS = QuotedMessageParserSettings()
@@ -0,0 +1,11 @@
+from __future__ import annotations
+
+from astrbot.core.utils.quoted_message.extractor import (
+    extract_quoted_message_images,
+    extract_quoted_message_text,
+)
+
+__all__ = [
+    "extract_quoted_message_text",
+    "extract_quoted_message_images",
+]
@@ -0,0 +1,21 @@
+from __future__ import annotations
+
+from collections.abc import Iterable
+from typing import Any
+
+
+def normalize_and_dedupe_strings(items: Iterable[Any] | None) -> list[str]:
+    if items is None:
+        return []
+
+    normalized: list[str] = []
+    seen: set[str] = set()
+    for item in items:
+        if not isinstance(item, str):
+            continue
+        cleaned = item.strip()
+        if not cleaned or cleaned in seen:
+            continue
+        seen.add(cleaned)
+        normalized.append(cleaned)
+    return normalized
@@ -0,0 +1,150 @@
+import asyncio
+from collections.abc import Callable
+from dataclasses import dataclass
+from pathlib import Path
+
+from astrbot import logger
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path
+
+
+def parse_size_to_bytes(value: str | int | float | None) -> int:
+    """Parse size in MB to bytes."""
+    if value is None:
+        return 0
+
+    try:
+        size_mb = float(str(value).strip())
+    except (TypeError, ValueError):
+        return 0
+
+    if size_mb <= 0:
+        return 0
+
+    return int(size_mb * 1024**2)
+
+
+@dataclass
+class TempFileInfo:
+    path: Path
+    size: int
+    mtime: float
+
+
+class TempDirCleaner:
+    CONFIG_KEY = "temp_dir_max_size"
+    DEFAULT_MAX_SIZE = 1024
+    CHECK_INTERVAL_SECONDS = 10 * 60
+    CLEANUP_RATIO = 0.30
+
+    def __init__(
+        self,
+        max_size_getter: Callable[[], str | int | float | None],
+        temp_dir: Path | None = None,
+    ) -> None:
+        self._max_size_getter = max_size_getter
+        self._temp_dir = temp_dir or Path(get_astrbot_temp_path())
+        self._stop_event = asyncio.Event()
+
+    def _limit_bytes(self) -> int:
+        configured = self._max_size_getter()
+        parsed = parse_size_to_bytes(configured)
+        if parsed <= 0:
+            fallback = parse_size_to_bytes(self.DEFAULT_MAX_SIZE)
+            logger.warning(
+                f"Invalid {self.CONFIG_KEY}={configured!r}, fallback to {self.DEFAULT_MAX_SIZE}MB.",
+            )
+            return fallback
+        return parsed
+
+    def _scan_temp_files(self) -> tuple[int, list[TempFileInfo]]:
+        if not self._temp_dir.exists():
+            return 0, []
+
+        total_size = 0
+        files: list[TempFileInfo] = []
+        for path in self._temp_dir.rglob("*"):
+            if not path.is_file():
+                continue
+            try:
+                stat = path.stat()
+            except OSError as e:
+                logger.debug(f"Skip temp file {path} due to stat error: {e}")
+                continue
+            total_size += stat.st_size
+            files.append(
+                TempFileInfo(path=path, size=stat.st_size, mtime=stat.st_mtime)
+            )
+
+        return total_size, files
+
+    def _cleanup_empty_dirs(self) -> None:
+        if not self._temp_dir.exists():
+            return
+        for path in sorted(
+            self._temp_dir.rglob("*"), key=lambda p: len(p.parts), reverse=True
+        ):
+            if not path.is_dir():
+                continue
+            try:
+                path.rmdir()
+            except OSError:
+                continue
+
+    def cleanup_once(self) -> None:
+        limit = self._limit_bytes()
+        if limit <= 0:
+            return
+
+        total_size, files = self._scan_temp_files()
+        if total_size <= limit:
+            return
+
+        target_release = max(int(total_size * self.CLEANUP_RATIO), 1)
+        released = 0
+        removed_files = 0
+
+        for file_info in sorted(files, key=lambda item: item.mtime):
+            try:
+                file_info.path.unlink()
+            except OSError as e:
+                logger.warning(f"Failed to delete temp file {file_info.path}: {e}")
+                continue
+
+            released += file_info.size
+            removed_files += 1
+            if released >= target_release:
+                break
+
+        self._cleanup_empty_dirs()
+
+        logger.warning(
+            f"Temp dir exceeded limit ({total_size} > {limit}). "
+            f"Removed {removed_files} files, released {released} bytes "
+            f"(target {target_release} bytes).",
+        )
+
+    async def run(self) -> None:
+        logger.info(
+            f"TempDirCleaner started. interval={self.CHECK_INTERVAL_SECONDS}s "
+            f"cleanup_ratio={self.CLEANUP_RATIO}",
+        )
+        while not self._stop_event.is_set():
+            try:
+                # File-system traversal and deletion are blocking operations.
+                # Run cleanup in a worker thread to avoid blocking the event loop.
+                await asyncio.to_thread(self.cleanup_once)
+            except Exception as e:
+                logger.error(f"TempDirCleaner run failed: {e}", exc_info=True)
+
+            try:
+                await asyncio.wait_for(
+                    self._stop_event.wait(),
+                    timeout=self.CHECK_INTERVAL_SECONDS,
+                )
+            except asyncio.TimeoutError:
+                continue
+
+        logger.info("TempDirCleaner stopped.")
+
+    async def stop(self) -> None:
+        self._stop_event.set()
@@ -7,7 +7,7 @@ import wave
 from io import BytesIO

 from astrbot.core import logger
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path


 async def tencent_silk_to_wav(silk_path: str, output_path: str) -> str:
@@ -117,12 +117,13 @@ async def audio_to_tencent_silk_base64(audio_path: str) -> tuple[str, float]:
    except ImportError as e:
        raise Exception("未安装 pilk: pip install pilk") from e

-    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+    temp_dir = get_astrbot_temp_path()
    os.makedirs(temp_dir, exist_ok=True)

    # 是否需要转换为 WAV
    ext = os.path.splitext(audio_path)[1].lower()
    temp_wav = tempfile.NamedTemporaryFile(
+        prefix="tencent_record_",
        suffix=".wav",
        delete=False,
        dir=temp_dir,
@@ -140,6 +141,7 @@ async def audio_to_tencent_silk_base64(audio_path: str) -> tuple[str, float]:
        rate = wav_file.getframerate()

    silk_path = tempfile.NamedTemporaryFile(
+        prefix="tencent_record_",
        suffix=".silk",
        delete=False,
        dir=temp_dir,
@@ -1,3 +1,4 @@
+import os
 import uuid

 from astrbot.core import astrbot_config, logger
@@ -20,6 +21,20 @@ def _get_dashboard_port() -> int:
        return 6185


+def _is_dashboard_ssl_enabled() -> bool:
+    env_ssl = os.environ.get("DASHBOARD_SSL_ENABLE") or os.environ.get(
+        "ASTRBOT_DASHBOARD_SSL_ENABLE"
+    )
+    if env_ssl is not None:
+        return env_ssl.strip().lower() in {"1", "true", "yes", "on"}
+
+    try:
+        return bool(astrbot_config.get("dashboard", {}).get("ssl", {}).get("enable"))
+    except Exception as e:
+        logger.error(f"获取 dashboard SSL 配置失败: {e!s}")
+        return False
+
+
 def log_webhook_info(platform_name: str, webhook_uuid: str) -> None:
    """打印美观的 webhook 信息日志

@@ -38,12 +53,13 @@ def log_webhook_info(platform_name: str, webhook_uuid: str) -> None:

    callback_base = callback_base.rstrip("/")
    webhook_url = f"{callback_base}/api/platform/webhook/{webhook_uuid}"
+    scheme = "https" if _is_dashboard_ssl_enabled() else "http"

    display_log = (
        "\n====================\n"
        f"🔗 机器人平台 {platform_name} 已启用统一 Webhook 模式\n"
        f"📍 Webhook 回调地址: \n"
-        f"   ➜  http://<your-ip>:{_get_dashboard_port()}/api/platform/webhook/{webhook_uuid}\n"
+        f"   ➜  {scheme}://<your-ip>:{_get_dashboard_port()}/api/platform/webhook/{webhook_uuid}\n"
        f"   ➜  {webhook_url}\n"
        "====================\n"
    )
@@ -1290,6 +1290,30 @@ class ConfigRoute(Route):
                f"Unexpected error registering logo for platform {platform.name}: {e}",
            )

+    def _inject_platform_metadata_with_i18n(
+        self, platform, metadata, platform_i18n_translations: dict
+    ):
+        """将配置元数据注入到 metadata 中并处理国际化键转换。"""
+        metadata["platform_group"]["metadata"]["platform"].setdefault("items", {})
+        platform_items_to_inject = copy.deepcopy(platform.config_metadata)
+
+        if platform.i18n_resources:
+            i18n_prefix = f"platform_group.platform.{platform.name}"
+
+            for lang, lang_data in platform.i18n_resources.items():
+                platform_i18n_translations.setdefault(lang, {}).setdefault(
+                    "platform_group", {}
+                ).setdefault("platform", {})[platform.name] = lang_data
+
+            for field_key, field_value in platform_items_to_inject.items():
+                for key in ("description", "hint", "labels"):
+                    if key in field_value:
+                        field_value[key] = f"{i18n_prefix}.{field_key}.{key}"
+
+        metadata["platform_group"]["metadata"]["platform"]["items"].update(
+            platform_items_to_inject
+        )
+
    async def _get_astrbot_config(self):
        config = self.config
        metadata = copy.deepcopy(CONFIG_METADATA_2)
@@ -1311,11 +1335,23 @@ class ConfigRoute(Route):
            "config_template"
        ]

+        # 收集平台的 i18n 翻译数据
+        platform_i18n_translations = {}
+
        # 收集需要注册logo的平台
        logo_registration_tasks = []
        for platform in platform_registry:
            if platform.default_config_tmpl:
-                platform_default_tmpl[platform.name] = platform.default_config_tmpl
+                platform_default_tmpl[platform.name] = copy.deepcopy(
+                    platform.default_config_tmpl
+                )
+
+                # 注入配置元数据（在 convert_to_i18n_keys 之后，使用国际化键）
+                if platform.config_metadata:
+                    self._inject_platform_metadata_with_i18n(
+                        platform, metadata, platform_i18n_translations
+                    )
+
                # 收集logo注册任务
                if platform.logo_path:
                    logo_registration_tasks.append(
@@ -1334,7 +1370,11 @@ class ConfigRoute(Route):
            if provider.default_config_tmpl:
                provider_default_tmpl[provider.type] = provider.default_config_tmpl

-        return {"metadata": metadata, "config": config}
+        return {
+            "metadata": metadata,
+            "config": config,
+            "platform_i18n_translations": platform_i18n_translations,
+        }

    async def _get_plugin_config(self, plugin_name: str):
        ret: dict = {"metadata": None, "config": None}
@@ -12,6 +12,7 @@ from quart import request
 from astrbot.core import logger
 from astrbot.core.core_lifecycle import AstrBotCoreLifecycle
 from astrbot.core.provider.provider import EmbeddingProvider, RerankProvider
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from ..utils import generate_tsne_visualization
 from .route import Response, Route, RouteContext
@@ -703,7 +704,10 @@ class KnowledgeBaseRoute(Route):
                file_name = file.filename

                # 保存到临时文件
-                temp_file_path = f"data/temp/{uuid.uuid4()}_{file_name}"
+                temp_file_path = os.path.join(
+                    get_astrbot_temp_path(),
+                    f"kb_upload_{uuid.uuid4()}_{file_name}",
+                )
                await file.save(temp_file_path)

                try:
@@ -12,7 +12,7 @@ from quart import websocket
 from astrbot import logger
 from astrbot.core.core_lifecycle import AstrBotCoreLifecycle
 from astrbot.core.platform.sources.webchat.webchat_queue_mgr import webchat_queue_mgr
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from .route import Route, RouteContext

@@ -60,7 +60,7 @@ class LiveChatSession:

        # 组装 WAV 文件
        try:
-            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+            temp_dir = get_astrbot_temp_path()
            os.makedirs(temp_dir, exist_ok=True)
            audio_path = os.path.join(temp_dir, f"live_audio_{uuid.uuid4()}.wav")

@@ -20,6 +20,7 @@ from astrbot.core.star.filter.permission import PermissionTypeFilter
 from astrbot.core.star.filter.regex import RegexFilter
 from astrbot.core.star.star_handler import EventType, star_handlers_registry
 from astrbot.core.star.star_manager import PluginManager
+from astrbot.core.utils.astrbot_path import get_astrbot_temp_path

 from .route import Response, Route, RouteContext

@@ -53,11 +54,13 @@ class PluginRoute(Route):
            "/plugin/market_list": ("GET", self.get_online_plugins),
            "/plugin/off": ("POST", self.off_plugin),
            "/plugin/on": ("POST", self.on_plugin),
+            "/plugin/reload-failed": ("POST", self.reload_failed_plugins),
            "/plugin/reload": ("POST", self.reload_plugins),
            "/plugin/readme": ("GET", self.get_plugin_readme),
            "/plugin/changelog": ("GET", self.get_plugin_changelog),
            "/plugin/source/get": ("GET", self.get_custom_source),
            "/plugin/source/save": ("POST", self.save_custom_source),
+            "/plugin/source/get-failed-plugins": ("GET", self.get_failed_plugins),
        }
        self.core_lifecycle = core_lifecycle
        self.plugin_manager = plugin_manager
@@ -74,6 +77,33 @@ class PluginRoute(Route):

        self._logo_cache = {}

+    async def reload_failed_plugins(self):
+        if DEMO_MODE:
+            return (
+                Response()
+                .error("You are not permitted to do this operation in demo mode")
+                .__dict__
+            )
+        try:
+            data = await request.get_json()
+            dir_name = data.get("dir_name")  # 这里拿的是目录名，不是插件名
+
+            if not dir_name:
+                return Response().error("缺少插件目录名").__dict__
+
+            # 调用 star_manager.py 中的函数
+            # 注意：传入的是目录名
+            success, err = await self.plugin_manager.reload_failed_plugin(dir_name)
+
+            if success:
+                return Response().ok(None, f"插件 {dir_name} 重载成功。").__dict__
+            else:
+                return Response().error(f"重载失败: {err}").__dict__
+
+        except Exception as e:
+            logger.error(f"/api/plugin/reload-failed: {traceback.format_exc()}")
+            return Response().error(str(e)).__dict__
+
    async def reload_plugins(self):
        if DEMO_MODE:
            return (
@@ -333,6 +363,10 @@ class PluginRoute(Route):
            .__dict__
        )

+    async def get_failed_plugins(self):
+        """专门获取加载失败的插件列表(字典格式)"""
+        return Response().ok(self.plugin_manager.failed_plugin_dict).__dict__
+
    async def get_plugin_handlers_info(self, handler_full_names: list[str]):
        """解析插件行为"""
        handlers = []
@@ -431,7 +465,10 @@ class PluginRoute(Route):
            file = await request.files
            file = file["file"]
            logger.info(f"正在安装用户上传的插件 {file.filename}")
-            file_path = f"data/temp/{file.filename}"
+            file_path = os.path.join(
+                get_astrbot_temp_path(),
+                f"plugin_upload_{file.filename}",
+            )
            await file.save(file_path)
            plugin_info = await self.plugin_manager.install_plugin_from_file(file_path)
            # self.core_lifecycle.restart()
@@ -4,6 +4,7 @@ import threading
 import time
 import traceback
 from functools import cmp_to_key
+from pathlib import Path

 import aiohttp
 import psutil
@@ -37,6 +38,7 @@ class StatRoute(Route):
            "/stat/test-ghproxy-connection": ("POST", self.test_ghproxy_connection),
            "/stat/changelog": ("GET", self.get_changelog),
            "/stat/changelog/list": ("GET", self.list_changelog_versions),
+            "/stat/first-notice": ("GET", self.get_first_notice),
        }
        self.db_helper = db_helper
        self.register_routes()
@@ -279,3 +281,40 @@ class StatRoute(Route):
        except Exception as e:
            logger.error(traceback.format_exc())
            return Response().error(f"Error: {e!s}").__dict__
+
+    async def get_first_notice(self):
+        """读取项目根目录 FIRST_NOTICE.md 内容。"""
+        try:
+            locale = (request.args.get("locale") or "").strip()
+            if not re.match(r"^[A-Za-z0-9_-]*$", locale):
+                locale = ""
+
+            base_path = Path(get_astrbot_path())
+            candidates: list[Path] = []
+
+            if locale:
+                candidates.append(base_path / f"FIRST_NOTICE.{locale}.md")
+                if locale.lower().startswith("zh"):
+                    candidates.append(base_path / "FIRST_NOTICE.md")
+                    candidates.append(base_path / "FIRST_NOTICE.zh-CN.md")
+                elif locale.lower().startswith("en"):
+                    candidates.append(base_path / "FIRST_NOTICE.en-US.md")
+
+            candidates.extend(
+                [
+                    base_path / "FIRST_NOTICE.md",
+                    base_path / "FIRST_NOTICE.en-US.md",
+                ],
+            )
+
+            for notice_path in candidates:
+                if not notice_path.is_file():
+                    continue
+                content = notice_path.read_text(encoding="utf-8")
+                if content.strip():
+                    return Response().ok({"content": content}).__dict__
+
+            return Response().ok({"content": None}).__dict__
+        except Exception as e:
+            logger.error(traceback.format_exc())
+            return Response().error(f"Error: {e!s}").__dict__
@@ -2,6 +2,7 @@ import asyncio
 import logging
 import os
 import socket
+from pathlib import Path
 from typing import Protocol, cast

 import jwt
@@ -36,6 +37,12 @@ class _AddrWithPort(Protocol):
 APP: Quart


+def _parse_env_bool(value: str | None, default: bool) -> bool:
+    if value is None:
+        return default
+    return value.strip().lower() in {"1", "true", "yes", "on"}
+
+
 class AstrBotDashboard:
    def __init__(
        self,
@@ -201,19 +208,33 @@ class AstrBotDashboard:

    def run(self):
        ip_addr = []
-        if p := os.environ.get("DASHBOARD_PORT"):
-            port = p
-        else:
-            port = self.core_lifecycle.astrbot_config["dashboard"].get("port", 6185)
-        host = self.core_lifecycle.astrbot_config["dashboard"].get("host", "0.0.0.0")
-        enable = self.core_lifecycle.astrbot_config["dashboard"].get("enable", True)
+        dashboard_config = self.core_lifecycle.astrbot_config.get("dashboard", {})
+        port = (
+            os.environ.get("DASHBOARD_PORT")
+            or os.environ.get("ASTRBOT_DASHBOARD_PORT")
+            or dashboard_config.get("port", 6185)
+        )
+        host = (
+            os.environ.get("DASHBOARD_HOST")
+            or os.environ.get("ASTRBOT_DASHBOARD_HOST")
+            or dashboard_config.get("host", "0.0.0.0")
+        )
+        enable = dashboard_config.get("enable", True)
+        ssl_config = dashboard_config.get("ssl", {})
+        if not isinstance(ssl_config, dict):
+            ssl_config = {}
+        ssl_enable = _parse_env_bool(
+            os.environ.get("DASHBOARD_SSL_ENABLE")
+            or os.environ.get("ASTRBOT_DASHBOARD_SSL_ENABLE"),
+            bool(ssl_config.get("enable", False)),
+        )
+        scheme = "https" if ssl_enable else "http"

        if not enable:
            logger.info("WebUI 已被禁用")
            return None

-        logger.info(f"正在启动 WebUI, 监听地址: http://{host}:{port}")
-
+        logger.info(f"正在启动 WebUI, 监听地址: {scheme}://{host}:{port}")
        if host == "0.0.0.0":
            logger.info(
                "提示: WebUI 将监听所有网络接口，请注意安全。（可在 data/cmd_config.json 中配置 dashboard.host 以修改 host）",
@@ -241,9 +262,9 @@ class AstrBotDashboard:
            raise Exception(f"端口 {port} 已被占用")

        parts = [f"\n ✨✨✨\n  AstrBot v{VERSION} WebUI 已启动，可访问\n\n"]
-        parts.append(f"   ➜  本地: http://localhost:{port}\n")
+        parts.append(f"   ➜  本地: {scheme}://localhost:{port}\n")
        for ip in ip_addr:
-            parts.append(f"   ➜  网络: http://{ip}:{port}\n")
+            parts.append(f"   ➜  网络: {scheme}://{ip}:{port}\n")
        parts.append("   ➜  默认用户名和密码: astrbot\n ✨✨✨\n")
        display = "".join(parts)

@@ -257,11 +278,45 @@ class AstrBotDashboard:
        # 配置 Hypercorn
        config = HyperConfig()
        config.bind = [f"{host}:{port}"]
+        if ssl_enable:
+            cert_file = (
+                os.environ.get("DASHBOARD_SSL_CERT")
+                or os.environ.get("ASTRBOT_DASHBOARD_SSL_CERT")
+                or ssl_config.get("cert_file", "")
+            )
+            key_file = (
+                os.environ.get("DASHBOARD_SSL_KEY")
+                or os.environ.get("ASTRBOT_DASHBOARD_SSL_KEY")
+                or ssl_config.get("key_file", "")
+            )
+            ca_certs = (
+                os.environ.get("DASHBOARD_SSL_CA_CERTS")
+                or os.environ.get("ASTRBOT_DASHBOARD_SSL_CA_CERTS")
+                or ssl_config.get("ca_certs", "")
+            )
+
+            cert_path = Path(cert_file).expanduser()
+            key_path = Path(key_file).expanduser()
+            if not cert_file or not key_file:
+                raise ValueError(
+                    "dashboard.ssl.enable 为 true 时，必须配置 cert_file 和 key_file。",
+                )
+            if not cert_path.is_file():
+                raise ValueError(f"SSL 证书文件不存在: {cert_path}")
+            if not key_path.is_file():
+                raise ValueError(f"SSL 私钥文件不存在: {key_path}")
+
+            config.certfile = str(cert_path.resolve())
+            config.keyfile = str(key_path.resolve())
+
+            if ca_certs:
+                ca_path = Path(ca_certs).expanduser()
+                if not ca_path.is_file():
+                    raise ValueError(f"SSL CA 证书文件不存在: {ca_path}")
+                config.ca_certs = str(ca_path.resolve())

        # 根据配置决定是否禁用访问日志
-        disable_access_log = self.core_lifecycle.astrbot_config.get(
-            "dashboard", {}
-        ).get("disable_access_log", True)
+        disable_access_log = dashboard_config.get("disable_access_log", True)
        if disable_access_log:
            config.accesslog = None
        else:
@@ -0,0 +1,62 @@
+## What's Changed
+
+### 新增
+- QQ 官方机器人平台支持主动推送消息，私聊场景支持接收文件 ([#5066](https://github.com/AstrBotDevs/AstrBot/issues/5066))
+- 为 Telegram 平台适配器新增等待 AI 回复时自动展示 “正在输入”、“正在上传图片” 等状态的功能 ([#5037](https://github.com/AstrBotDevs/AstrBot/issues/5037))
+- 为飞书适配器增加接收文件、读取引用消息的内容（包括引用的图片、视频、文件、文字等） ([#5018](https://github.com/AstrBotDevs/AstrBot/issues/5018))
+- 新增自定义平台适配器 i18n 支持 ([#5045](https://github.com/AstrBotDevs/AstrBot/issues/5045))
+- 新增临时文件处理能力，可在系统配置中限制 data/temp 目录的最大大小。 ([#5026](https://github.com/AstrBotDevs/AstrBot/issues/5026))
+- 增加首次启动公告功能，支持多语言与 WebUI 集成
+
+### 修复
+
+- 修复 OpenRouter DeepSeek 场景下的 chunk 错误 ([#5069](https://github.com/AstrBotDevs/AstrBot/issues/5069))
+- 修复备份时人格文件夹映射缺失问题 ([#5042](https://github.com/AstrBotDevs/AstrBot/issues/5042))
+- 修复更新日志与官方文档弹窗双滚动条问题 ([#5060](https://github.com/AstrBotDevs/AstrBot/issues/5060))
+- 修复 provider 额外参数弹窗 key 显示异常
+- 修复连接失败时错误日志提示不准确的问题
+- 修复提前返回时未等待 reset 协程导致的资源清理问题 ([#5033](https://github.com/AstrBotDevs/AstrBot/issues/5033))
+- 提升打包版桌面端启动稳定性并优化插件依赖处理 ([#5031](https://github.com/AstrBotDevs/AstrBot/issues/5031))
+- 为 Electron 与后端日志增加按大小轮转 ([#5029](https://github.com/AstrBotDevs/AstrBot/issues/5029))
+- 加固冻结运行时（frozen app runtime）插件依赖加载流程 ([#5015](https://github.com/AstrBotDevs/AstrBot/issues/5015))
+
+### 优化
+- 完善合并消息、引用解析与图片回退，并支持配置化控制 ([#5054](https://github.com/AstrBotDevs/AstrBot/issues/5054))
+- 配置页面支持通过侧边栏子项切换普通配置/系统配置，并补充相关路由修复
+- 优化分段回复间隔时间初始化逻辑 ([#5068](https://github.com/AstrBotDevs/AstrBot/issues/5068))
+
+### 文档与维护
+- 同步并修正 README 文档内容与拼写 ([#5055](https://github.com/AstrBotDevs/AstrBot/issues/5055), [#5014](https://github.com/AstrBotDevs/AstrBot/issues/5014))
+- 新增 AUR 安装方式说明 ([#4879](https://github.com/AstrBotDevs/AstrBot/issues/4879))
+- 执行代码格式化（ruff）
+
+## What's Changed (EN)
+
+### New Features
+- Added proactive message push and private-chat file receiving support for the QQ official bot adapter ([#5066](https://github.com/AstrBotDevs/AstrBot/issues/5066))
+- Added automatic "typing..." and "uploading image..." status display while waiting for AI response in the Telegram adapter ([#5037](https://github.com/AstrBotDevs/AstrBot/issues/5037))
+- Added file receiving and quoted message content reading (including quoted images, videos, files, text, etc.) for the Feishu adapter ([#5018](https://github.com/AstrBotDevs/AstrBot/issues/5018))
+- Added i18n support for custom platform adapters ([#5045](https://github.com/AstrBotDevs/AstrBot/issues/5045))
+- Introduced temporary file handling and `TempDirCleaner` ([#5026](https://github.com/AstrBotDevs/AstrBot/issues/5026))
+- Added a first-launch notice feature with multilingual content and WebUI integration
+
+### Fixes
+- Added sidebar child-tab switching for normal/system config and fixed related routing behavior on the config page
+- Fixed chunk errors when using OpenRouter DeepSeek ([#5069](https://github.com/AstrBotDevs/AstrBot/issues/5069))
+- Improved forwarded-quote parsing and image fallback with configurable controls ([#5054](https://github.com/AstrBotDevs/AstrBot/issues/5054))
+- Fixed missing persona-folder mapping in backup exports ([#5042](https://github.com/AstrBotDevs/AstrBot/issues/5042))
+- Fixed double scrollbar issue in changelog and official docs dialogs ([#5060](https://github.com/AstrBotDevs/AstrBot/issues/5060))
+- Fixed key rendering issues in the provider extra-params dialog
+- Improved error log wording for connection failures
+- Fixed unawaited reset coroutine cleanup on early returns ([#5033](https://github.com/AstrBotDevs/AstrBot/issues/5033))
+- Improved packaged desktop startup stability and plugin dependency handling ([#5031](https://github.com/AstrBotDevs/AstrBot/issues/5031))
+- Added size-based log rotation for Electron and backend logs ([#5029](https://github.com/AstrBotDevs/AstrBot/issues/5029))
+- Hardened plugin dependency loading in frozen app runtime ([#5015](https://github.com/AstrBotDevs/AstrBot/issues/5015))
+
+### Improvements
+- Optimized initialization logic for segmented-reply interval timing ([#5068](https://github.com/AstrBotDevs/AstrBot/issues/5068))
+
+### Docs & Maintenance
+- Synced and fixed README docs and typos ([#5055](https://github.com/AstrBotDevs/AstrBot/issues/5055), [#5014](https://github.com/AstrBotDevs/AstrBot/issues/5014))
+- Added AUR installation instructions ([#4879](https://github.com/AstrBotDevs/AstrBot/issues/4879))
+- Applied code formatting with ruff
@@ -0,0 +1,29 @@
+## What's Changed
+
+### 新增
+- 新增 LINE 平台适配器与相关配置支持 ([#5085](https://github.com/AstrBotDevs/AstrBot/issues/5085))
+- 新增备用回退聊天模型列表，当主模型报错时自动切换到备用模型 ([#5109](https://github.com/AstrBotDevs/AstrBot/issues/5109))
+- 新增插件加载失败后的热重载支持，便于插件修复后快速恢复 ([#5043](https://github.com/AstrBotDevs/AstrBot/issues/5043))
+- WebUI 新增 SSL 配置选项并同步更新相关日志行为 ([#5117](https://github.com/AstrBotDevs/AstrBot/issues/5117))
+
+### 修复
+- 修复 Dockerfile 中依赖导出流程，增加 `uv lock` 步骤并移除不必要的 `--frozen` 参数，提升构建稳定性 ([#5091](https://github.com/AstrBotDevs/AstrBot/issues/5091), [#5089](https://github.com/AstrBotDevs/AstrBot/issues/5089))
+- 修复首次启动公告 `FIRST_NOTICE.md` 的本地化路径解析问题，补充兼容路径处理 ([#5083](https://github.com/AstrBotDevs/AstrBot/issues/5083), [#5082](https://github.com/AstrBotDevs/AstrBot/issues/5082))
+
+### 优化
+- 日志系统由 `colorlog` 切换为 `loguru`，增强日志输出与展示能力 ([#5115](https://github.com/AstrBotDevs/AstrBot/issues/5115))
+
+## What's Changed (EN)
+
+### New Features
+- Added LINE platform adapter support with related configuration options ([#5085](https://github.com/AstrBotDevs/AstrBot/issues/5085))
+- Added fallback chat model chain support in tool loop runner, with corresponding config and improved provider selection display ([#5109](https://github.com/AstrBotDevs/AstrBot/issues/5109))
+- Added hot reload support after plugin load failure for faster recovery during plugin development and maintenance ([#5043](https://github.com/AstrBotDevs/AstrBot/issues/5043))
+- Added SSL configuration options for WebUI and updated related logging behavior ([#5117](https://github.com/AstrBotDevs/AstrBot/issues/5117))
+
+### Fixes
+- Fixed Dockerfile dependency export flow by adding a `uv lock` step and removing unnecessary `--frozen` flag to improve build stability ([#5091](https://github.com/AstrBotDevs/AstrBot/issues/5091), [#5089](https://github.com/AstrBotDevs/AstrBot/issues/5089))
+- Fixed locale path resolution for `FIRST_NOTICE.md` and added compatible fallback handling ([#5083](https://github.com/AstrBotDevs/AstrBot/issues/5083), [#5082](https://github.com/AstrBotDevs/AstrBot/issues/5082))
+
+### Improvements
+- Replaced `colorlog` with `loguru` to improve logging capabilities and console display ([#5115](https://github.com/AstrBotDevs/AstrBot/issues/5115))
@@ -0,0 +1,34 @@
+## What's Changed
+
+hotfix of 4.17.0
+
+- 修复：当开启了 “启用文件日志” 后，无法启动 AstrBot，报错 `ValueError: Invalid unit value while parsing duration: 'files'`。这是由于日志轮转设置中保留配置错误导致的，已通过根据备份数量正确设置保留参数进行修复。
+- fix: When "Enable file logging" is turned on, AstrBot fails to start with error `ValueError: Invalid unit value while parsing duration: 'files'`. This is due to an incorrect retention configuration in the log rotation setup, which has been fixed by properly setting the retention parameter based on backup count. 
+
+### 新增
+- 新增 LINE 平台适配器与相关配置支持 ([#5085](https://github.com/AstrBotDevs/AstrBot/issues/5085))
+- 新增备用回退聊天模型列表，当主模型报错时自动切换到备用模型 ([#5109](https://github.com/AstrBotDevs/AstrBot/issues/5109))
+- 新增插件加载失败后的热重载支持，便于插件修复后快速恢复 ([#5043](https://github.com/AstrBotDevs/AstrBot/issues/5043))
+- WebUI 新增 SSL 配置选项并同步更新相关日志行为 ([#5117](https://github.com/AstrBotDevs/AstrBot/issues/5117))
+
+### 修复
+- 修复 Dockerfile 中依赖导出流程，增加 `uv lock` 步骤并移除不必要的 `--frozen` 参数，提升构建稳定性 ([#5091](https://github.com/AstrBotDevs/AstrBot/issues/5091), [#5089](https://github.com/AstrBotDevs/AstrBot/issues/5089))
+- 修复首次启动公告 `FIRST_NOTICE.md` 的本地化路径解析问题，补充兼容路径处理 ([#5083](https://github.com/AstrBotDevs/AstrBot/issues/5083), [#5082](https://github.com/AstrBotDevs/AstrBot/issues/5082))
+
+### 优化
+- 日志系统由 `colorlog` 切换为 `loguru`，增强日志输出与展示能力 ([#5115](https://github.com/AstrBotDevs/AstrBot/issues/5115))
+
+## What's Changed (EN)
+
+### New Features
+- Added LINE platform adapter support with related configuration options ([#5085](https://github.com/AstrBotDevs/AstrBot/issues/5085))
+- Added fallback chat model chain support in tool loop runner, with corresponding config and improved provider selection display ([#5109](https://github.com/AstrBotDevs/AstrBot/issues/5109))
+- Added hot reload support after plugin load failure for faster recovery during plugin development and maintenance ([#5043](https://github.com/AstrBotDevs/AstrBot/issues/5043))
+- Added SSL configuration options for WebUI and updated related logging behavior ([#5117](https://github.com/AstrBotDevs/AstrBot/issues/5117))
+
+### Fixes
+- Fixed Dockerfile dependency export flow by adding a `uv lock` step and removing unnecessary `--frozen` flag to improve build stability ([#5091](https://github.com/AstrBotDevs/AstrBot/issues/5091), [#5089](https://github.com/AstrBotDevs/AstrBot/issues/5089))
+- Fixed locale path resolution for `FIRST_NOTICE.md` and added compatible fallback handling ([#5083](https://github.com/AstrBotDevs/AstrBot/issues/5083), [#5082](https://github.com/AstrBotDevs/AstrBot/issues/5082))
+
+### Improvements
+- Replaced `colorlog` with `loguru` to improve logging capabilities and console display ([#5115](https://github.com/AstrBotDevs/AstrBot/issues/5115))
@@ -0,0 +1,8 @@
+## What's Changed
+
+hotfix of 4.17.0
+
+- 修复：MCP 服务器的 Tools 没有被正确添加到上下文中。
+- 修复：Electron 桌面应用部署时，系统自带插件未被正确加载的问题。
+- fix: Tools from MCP server were not properly added to context.
+- fix: built-in plugins were not properly loaded in Electron desktop application deployment.
@@ -1,5 +1,6 @@
 <template>
  <RouterView></RouterView>
+  <WaitingForRestart ref="globalWaitingRef" />

  <!-- 全局唯一 snackbar -->
  <v-snackbar v-if="toastStore.current" v-model="snackbarShow" :color="toastStore.current.color"
@@ -14,10 +15,14 @@

 <script setup>
 import { RouterView } from 'vue-router';
-import { computed } from 'vue'
+import { computed, onBeforeUnmount, onMounted, ref } from 'vue'
 import { useToastStore } from '@/stores/toast'
+import WaitingForRestart from '@/components/shared/WaitingForRestart.vue'
+import { restartAstrBot } from '@/utils/restartAstrBot'

 const toastStore = useToastStore()
+const globalWaitingRef = ref(null)
+let disposeTrayRestartListener = null

 const snackbarShow = computed({
  get: () => !!toastStore.current,
@@ -25,4 +30,26 @@ const snackbarShow = computed({
    if (!val) toastStore.shift()
  }
 })
+
+onMounted(() => {
+  const desktopBridge = window.astrbotDesktop
+  if (!desktopBridge?.isElectron || !desktopBridge.onTrayRestartBackend) {
+    return
+  }
+  disposeTrayRestartListener = desktopBridge.onTrayRestartBackend(async () => {
+    try {
+      await restartAstrBot(globalWaitingRef.value)
+    } catch (error) {
+      globalWaitingRef.value?.stop?.()
+      console.error('Tray restart backend failed:', error)
+    }
+  })
+})
+
+onBeforeUnmount(() => {
+  if (disposeTrayRestartListener) {
+    disposeTrayRestartListener()
+    disposeTrayRestartListener = null
+  }
+})
 </script>
@@ -522,7 +522,14 @@ export default {
    }
  },
  methods: {
-    getPlatformIcon,
+    getPlatformIcon(platformType) {
+      // Check for plugin-provided logo_token first
+      const template = this.platformTemplates?.[platformType];
+      if (template && template.logo_token) {
+        return `/api/file/${template.logo_token}`;
+      }
+      return getPlatformIcon(platformType);
+    },
    getPlatformDescription,
    resetForm() {
      this.selectedPlatformType = null;
@@ -151,7 +151,7 @@ getCurrentVersion();
          <v-icon>mdi-close</v-icon>
        </v-btn>
      </v-card-title>
-      <v-card-text>
+      <v-card-text class="pb-5">
        <!-- 版本选择器 -->
        <div class="mb-4">
          <v-select
@@ -10,6 +10,14 @@
    <template v-else-if="itemMeta?._special === 'select_provider_tts'">
      <ProviderSelector :model-value="modelValue" @update:model-value="emitUpdate" :provider-type="'text_to_speech'" />
    </template>
+    <template v-else-if="itemMeta?._special === 'select_providers'">
+      <ProviderSelector
+        :model-value="modelValue"
+        @update:model-value="emitUpdate"
+        :provider-type="'chat_completion'"
+        :multiple="true"
+      />
+    </template>
    <template v-else-if="getSpecialName(itemMeta?._special) === 'select_agent_runner_provider'">
      <ProviderSelector
        :model-value="modelValue"
@@ -27,7 +27,7 @@ export default {
    return {
      autoScroll: true,
      logColorAnsiMap: {
-        '\u001b[1;34m': 'color: #0000FF; font-weight: bold;',
+        '\u001b[1;34m': 'color: #39C5BB; font-weight: bold;',
        '\u001b[1;36m': 'color: #00FFFF; font-weight: bold;',
        '\u001b[1;33m': 'color: #FFFF00; font-weight: bold;',
        '\u001b[31m': 'color: #FF0000;',
@@ -2,7 +2,7 @@
  <div class="d-flex align-center justify-space-between">
    <div>
      <span v-if="!modelValue || Object.keys(modelValue).length === 0" style="color: rgb(var(--v-theme-primaryText));">
-        暂无项目
+        {{ t('core.common.objectEditor.noItems') }}
      </span>
      <div v-else class="d-flex flex-wrap ga-2">
        <v-chip v-for="key in displayKeys" :key="key" size="x-small" label color="primary">
@@ -14,7 +14,7 @@
      </div>
    </div>
    <v-btn size="small" color="primary" variant="tonal" @click="openDialog">
-      {{ buttonText }}
+      {{ resolveButtonText }}
    </v-btn>
  </div>

@@ -22,7 +22,7 @@
  <v-dialog v-model="dialog" max-width="600px">
    <v-card>
      <v-card-title class="text-h3 py-4" style="font-weight: normal;">
-        {{ dialogTitle }}
+        {{ resolveDialogTitle }}
      </v-card-title>

      <v-card-text class="pa-4" style="max-height: 400px; overflow-y: auto;">
@@ -36,7 +36,7 @@
                  density="compact"
                  variant="outlined"
                  hide-details
-                  placeholder="键名"
+                  :placeholder="t('core.common.objectEditor.placeholders.keyName')"
                  @blur="updateKey(index, pair.key)"
                ></v-text-field>
              </v-col>
@@ -47,7 +47,7 @@
                  density="compact"
                  variant="outlined"
                  hide-details
-                  placeholder="字符串值"
+                  :placeholder="t('core.common.objectEditor.placeholders.stringValue')"
                ></v-text-field>
                <div v-else-if="pair.type === 'number' || pair.type === 'float' || pair.type === 'int'" class="d-flex align-center gap-2 flex-grow-1">
                  <v-slider
@@ -68,7 +68,7 @@
                    density="compact"
                    variant="outlined"
                    hide-details
-                    placeholder="数值"
+                    :placeholder="t('core.common.objectEditor.placeholders.numberValue')"
                    :style="pair.slider ? 'max-width: 120px;' : ''"
                  ></v-text-field>
                </div>
@@ -85,7 +85,7 @@
                  density="compact"
                  variant="outlined"
                  hide-details="auto"
-                  placeholder="JSON"
+                  :placeholder="t('core.common.objectEditor.placeholders.jsonValue')"
                  @blur="updateJSON(index, pair.value)"
                  :error-messages="pair.jsonError"
                ></v-text-field>
@@ -108,13 +108,13 @@
        <!-- Template schema fields -->
        <div v-if="hasTemplateSchema" class="mt-4">
          <v-divider class="mb-3"></v-divider>
-          <div class="text-caption text-grey mb-2">预设</div>
+          <div class="text-caption text-grey mb-2">{{ t('core.common.objectEditor.presets') }}</div>
          <div v-for="(template, templateKey) in templateSchema" :key="templateKey" class="template-field" :class="{ 'template-field-inactive': !isTemplateKeyAdded(templateKey) }">
            <v-row no-gutters align="center" class="mb-2">
              <v-col cols="4">
                <div class="d-flex flex-column">
-                  <span class="text-caption font-weight-medium">{{ template.name || template.description || templateKey }}</span>
-                  <span v-if="template.hint" class="text-caption text-grey" style="font-size: 0.7rem;">{{ template.hint }}</span>
+                  <span class="text-caption font-weight-medium">{{ getTemplateTitle(template, templateKey) }}</span>
+                  <span v-if="template.hint" class="text-caption text-grey" style="font-size: 0.7rem;">{{ translateIfKey(template.hint) }}</span>
                </div>
              </v-col>
              <v-col cols="7" class="pl-2 d-flex align-center justify-end">
@@ -125,7 +125,7 @@
                  density="compact"
                  variant="outlined"
                  hide-details
-                  placeholder="字符串值"
+                  :placeholder="t('core.common.objectEditor.placeholders.stringValue')"
                ></v-text-field>
                <div v-else-if="template.type === 'number' || template.type === 'float' || template.type === 'int'" class="d-flex align-center ga-4 flex-grow-1">
                  <v-slider
@@ -147,7 +147,7 @@
                    density="compact"
                    variant="outlined"
                    hide-details
-                    placeholder="数值"
+                    :placeholder="t('core.common.objectEditor.placeholders.numberValue')"
                    :style="template.slider ? 'max-width: 120px;' : ''"
                  ></v-text-field>
                </div>
@@ -178,7 +178,7 @@

        <div v-if="localKeyValuePairs.length === 0 && !hasTemplateSchema" class="text-center py-8">
          <v-icon size="64" color="grey-lighten-1">mdi-code-json</v-icon>
-          <p class="text-grey mt-4">暂无参数</p>
+          <p class="text-grey mt-4">{{ t('core.common.objectEditor.noParams') }}</p>
        </div>
      </v-card-text>

@@ -187,7 +187,7 @@
        <div class="d-flex align-center ga-2">
          <v-text-field
            v-model="newKey"
-            label="新键名"
+            :label="t('core.common.objectEditor.newKeyLabel')"
            density="compact"
            variant="outlined"
            hide-details
@@ -196,7 +196,7 @@
          <v-select
            v-model="newValueType"
            :items="['string', 'number', 'boolean', 'json']"
-            label="值类型"
+            :label="t('core.common.objectEditor.valueTypeLabel')"
            density="compact"
            variant="outlined"
            hide-details
@@ -204,15 +204,15 @@
          ></v-select>
          <v-btn @click="addKeyValuePair" variant="tonal" color="primary">
            <v-icon>mdi-plus</v-icon>
-            添加
+            {{ t('core.common.add') }}
          </v-btn>
        </div>
      </v-card-text>

      <v-card-actions class="pa-4">
        <v-spacer></v-spacer>
-        <v-btn variant="text" @click="cancelDialog">取消</v-btn>
-        <v-btn color="primary" @click="confirmDialog">确认</v-btn>
+        <v-btn variant="text" @click="cancelDialog">{{ t('core.common.cancel') }}</v-btn>
+        <v-btn color="primary" @click="confirmDialog">{{ t('core.common.confirm') }}</v-btn>
      </v-card-actions>
    </v-card>
  </v-dialog>
@@ -220,9 +220,10 @@

 <script setup>
 import { ref, computed, watch } from 'vue'
-import { useI18n } from '@/i18n/composables'
+import { useI18n, useModuleI18n } from '@/i18n/composables'

 const { t } = useI18n()
+const { tm, getRaw } = useModuleI18n('features/config-metadata')

 const props = defineProps({
  modelValue: {
@@ -235,11 +236,11 @@ const props = defineProps({
  },
  buttonText: {
    type: String,
-    default: '修改'
+    default: ''
  },
  dialogTitle: {
    type: String,
-    default: '修改键值对'
+    default: ''
  },
  maxDisplayItems: {
    type: Number,
@@ -249,6 +250,9 @@ const props = defineProps({

 const emit = defineEmits(['update:modelValue'])

+const resolveButtonText = computed(() => props.buttonText || t('core.common.list.modifyButton'))
+const resolveDialogTitle = computed(() => props.dialogTitle || t('core.common.objectEditor.dialogTitle'))
+
 const dialog = ref(false)
 const localKeyValuePairs = ref([])
 const originalKeyValuePairs = ref([])
@@ -320,7 +324,7 @@ function addKeyValuePair() {
  if (key !== '') {
    const isKeyExists = localKeyValuePairs.value.some(pair => pair.key === key)
    if (isKeyExists) {
-      alert('键名已存在')
+      alert(t('core.common.objectEditor.keyExists'))
      return
    }

@@ -354,7 +358,7 @@ function updateJSON(index, newValue) {
    JSON.parse(newValue)
    localKeyValuePairs.value[index].jsonError = ''
  } catch (e) {
-    localKeyValuePairs.value[index].jsonError = 'JSON 格式错误'
+    localKeyValuePairs.value[index].jsonError = t('core.common.objectEditor.invalidJson')
  }
 }

@@ -374,7 +378,7 @@ function updateKey(index, newKey) {
  const isKeyExists = localKeyValuePairs.value.some((pair, i) => i !== index && pair.key === newKey)
  if (isKeyExists) {
    // 如果键名已存在，提示用户并恢复原值
-    alert('键名已存在')
+    alert(t('core.common.objectEditor.keyExists'))
    // 将键名恢复为修改前的原始值
    localKeyValuePairs.value[index].key = originalKey
    return
@@ -501,6 +505,15 @@ function cancelDialog() {
  localKeyValuePairs.value = JSON.parse(JSON.stringify(originalKeyValuePairs.value))
  dialog.value = false
 }
+
+function translateIfKey(value) {
+  if (!value || typeof value !== 'string') return value
+  return getRaw(value) ? tm(value) : value
+}
+
+function getTemplateTitle(template, templateKey) {
+  return translateIfKey(template?.name || template?.description || templateKey)
+}
 </script>

 <style scoped>
@@ -515,4 +528,4 @@ function cancelDialog() {
 .template-field-inactive {
  opacity: 0.8;
 }
-</style>
+</style>
@@ -1,16 +1,35 @@
 <template>
  <div class="d-flex align-center justify-space-between">
-    <span v-if="!modelValue" style="color: rgb(var(--v-theme-primaryText));">
+    <span v-if="!hasSelection" style="color: rgb(var(--v-theme-primaryText));">
      {{ tm('providerSelector.notSelected') }}
    </span>
    <span v-else class="provider-name-text">
-      {{ modelValue }}
+      <template v-if="multiple">
+        {{ tm('providerSelector.selectedCount', { count: selectedProviders.length }) }}
+      </template>
+      <template v-else>
+        {{ modelValue }}
+      </template>
    </span>
    <v-btn size="small" color="primary" variant="tonal" @click="openDialog">
      {{ buttonText || tm('providerSelector.buttonText') }}
    </v-btn>
  </div>

+  <div v-if="multiple && selectedProviders.length > 0" class="selected-preview mt-2">
+    <v-chip
+      v-for="providerId in selectedProviders"
+      :key="`preview-${providerId}`"
+      size="x-small"
+      color="primary"
+      variant="tonal"
+      class="mr-1 mb-1"
+      label
+    >
+      {{ providerId }}
+    </v-chip>
+  </div>
+
  <!-- Provider Selection Dialog -->
  <v-dialog v-model="dialog" max-width="600px">
    <v-card>
@@ -32,10 +51,52 @@
      
      <v-card-text class="pa-0" style="max-height: 400px; overflow-y: auto;">
        <v-progress-linear v-if="loading" indeterminate color="primary"></v-progress-linear>
+
+        <div v-if="multiple && selectedProviders.length > 0" class="pa-3">
+          <div class="text-caption text-medium-emphasis mb-2">
+            {{ tm('providerSelector.selectedCount', { count: selectedProviders.length }) }}
+          </div>
+          <v-list density="compact" class="selected-order-list">
+            <v-list-item
+              v-for="(providerId, index) in selectedProviders"
+              :key="`selected-${providerId}-${index}`"
+              rounded="md"
+              class="ma-1"
+            >
+              <v-list-item-title>{{ providerId }}</v-list-item-title>
+              <template #append>
+                <div class="d-flex ga-1">
+                  <v-btn
+                    icon="mdi-arrow-up"
+                    size="x-small"
+                    variant="text"
+                    :disabled="index === 0"
+                    @click.stop="moveSelected(index, -1)"
+                  />
+                  <v-btn
+                    icon="mdi-arrow-down"
+                    size="x-small"
+                    variant="text"
+                    :disabled="index === selectedProviders.length - 1"
+                    @click.stop="moveSelected(index, 1)"
+                  />
+                  <v-btn
+                    icon="mdi-close"
+                    size="x-small"
+                    variant="text"
+                    @click.stop="removeSelected(providerId)"
+                  />
+                </div>
+              </template>
+            </v-list-item>
+          </v-list>
+          <v-divider class="ma-1"></v-divider>
+        </div>
        
        <v-list v-if="!loading && providerList.length > 0" density="compact">
          <!-- 不选择选项 -->
          <v-list-item
+            v-if="!multiple"
            key="none"
            value=""
            @click="selectProvider({ id: '' })"
@@ -57,7 +118,7 @@
            :key="provider.id"
            :value="provider.id"
            @click="selectProvider(provider)"
-            :active="selectedProvider === provider.id"
+            :active="isProviderSelected(provider.id)"
            rounded="md"
            class="ma-1">
            <v-list-item-title>{{ provider.id }}</v-list-item-title>
@@ -67,7 +128,7 @@
            </v-list-item-subtitle>
            
            <template v-slot:append>
-              <v-icon v-if="selectedProvider === provider.id" color="primary">mdi-check-circle</v-icon>
+              <v-icon v-if="isProviderSelected(provider.id)" color="primary">mdi-check-circle</v-icon>
            </template>
          </v-list-item>
        </v-list>
@@ -121,7 +182,7 @@ import ProviderPage from '@/views/ProviderPage.vue'

 const props = defineProps({
  modelValue: {
-    type: String,
+    type: [String, Array],
    default: ''
  },
  providerType: {
@@ -135,6 +196,10 @@ const props = defineProps({
  buttonText: {
    type: String,
    default: ''
+  },
+  multiple: {
+    type: Boolean,
+    default: false
  }
 })

@@ -145,8 +210,16 @@ const dialog = ref(false)
 const providerList = ref([])
 const loading = ref(false)
 const selectedProvider = ref('')
+const selectedProviders = ref([])
 const providerDrawer = ref(false)

+const hasSelection = computed(() => {
+  if (props.multiple) {
+    return selectedProviders.value.length > 0
+  }
+  return Boolean(props.modelValue)
+})
+
 const defaultTab = computed(() => {
  if (props.providerType === 'agent_runner' && props.providerSubtype) {
    return `select_agent_runner_provider:${props.providerSubtype}`
@@ -156,7 +229,13 @@ const defaultTab = computed(() => {

 // 监听 modelValue 变化，同步到 selectedProvider
 watch(() => props.modelValue, (newValue) => {
-  selectedProvider.value = newValue || ''
+  if (props.multiple) {
+    selectedProviders.value = Array.isArray(newValue)
+      ? [...newValue.filter((v) => typeof v === 'string' && v)]
+      : []
+    return
+  }
+  selectedProvider.value = typeof newValue === 'string' ? newValue : ''
 }, { immediate: true })

 watch(providerDrawer, (isOpen, wasOpen) => {
@@ -166,7 +245,13 @@ watch(providerDrawer, (isOpen, wasOpen) => {
 })

 async function openDialog() {
-  selectedProvider.value = props.modelValue || ''
+  if (props.multiple) {
+    selectedProviders.value = Array.isArray(props.modelValue)
+      ? [...props.modelValue.filter((v) => typeof v === 'string' && v)]
+      : []
+  } else {
+    selectedProvider.value = typeof props.modelValue === 'string' ? props.modelValue : ''
+  }
  dialog.value = true
  await loadProviders()
 }
@@ -205,19 +290,72 @@ function matchesProviderSubtype(provider, subtype) {
 }

 function selectProvider(provider) {
+  if (props.multiple) {
+    if (!provider.id) {
+      selectedProviders.value = []
+      return
+    }
+    const idx = selectedProviders.value.indexOf(provider.id)
+    if (idx >= 0) {
+      selectedProviders.value.splice(idx, 1)
+    } else {
+      selectedProviders.value.push(provider.id)
+    }
+    return
+  }
  selectedProvider.value = provider.id
 }

 function confirmSelection() {
-  emit('update:modelValue', selectedProvider.value)
+  if (props.multiple) {
+    emit('update:modelValue', [...selectedProviders.value])
+  } else {
+    emit('update:modelValue', selectedProvider.value)
+  }
  dialog.value = false
 }

 function cancelSelection() {
-  selectedProvider.value = props.modelValue || ''
+  if (props.multiple) {
+    selectedProviders.value = Array.isArray(props.modelValue)
+      ? [...props.modelValue.filter((v) => typeof v === 'string' && v)]
+      : []
+  } else {
+    selectedProvider.value = typeof props.modelValue === 'string' ? props.modelValue : ''
+  }
  dialog.value = false
 }

+function isProviderSelected(providerId) {
+  if (props.multiple) {
+    return selectedProviders.value.includes(providerId)
+  }
+  return selectedProvider.value === providerId
+}
+
+function removeSelected(providerId) {
+  const idx = selectedProviders.value.indexOf(providerId)
+  if (idx >= 0) {
+    selectedProviders.value.splice(idx, 1)
+  }
+}
+
+function moveSelected(index, delta) {
+  const targetIndex = index + delta
+  if (
+    targetIndex < 0
+    || targetIndex >= selectedProviders.value.length
+    || index < 0
+    || index >= selectedProviders.value.length
+  ) {
+    return
+  }
+  const copied = [...selectedProviders.value]
+  const [item] = copied.splice(index, 1)
+  copied.splice(targetIndex, 0, item)
+  selectedProviders.value = copied
+}
+
 function openProviderDrawer() {
  providerDrawer.value = true
 }
@@ -236,6 +374,16 @@ function closeProviderDrawer() {
  display: inline-block;
 }

+.selected-preview {
+  width: 100%;
+  max-width: 100%;
+}
+
+.selected-order-list {
+  background: rgba(var(--v-theme-surface-variant), 0.15);
+  border-radius: 10px;
+}
+
 .v-list-item {
  transition: all 0.2s ease;
 }
@@ -35,7 +35,7 @@ const props = defineProps({
  mode: {
    type: String,
    default: "readme",
-    validator: (value) => ["readme", "changelog"].includes(value),
+    validator: (value) => ["readme", "changelog", "first-notice"].includes(value),
  },
 });

@@ -166,19 +166,50 @@ const renderedHtml = computed(() => {
 });

 const modeConfig = computed(() => {
-  const isChangelog = props.mode === "changelog";
-  const keyBase = `core.common.${isChangelog ? "changelog" : "readme"}`;
+  if (props.mode === "changelog") {
+    return {
+      title: t("core.common.changelog.title"),
+      loading: t("core.common.changelog.loading"),
+      emptyTitle: t("core.common.changelog.empty.title"),
+      emptySubtitle: t("core.common.changelog.empty.subtitle"),
+      apiPath: "/api/plugin/changelog",
+      showGithubButton: false,
+      showRefreshButton: true,
+      refreshLabel: t("core.common.readme.buttons.refresh"),
+    };
+  }
+
+  if (props.mode === "first-notice") {
+    return {
+      title: t("core.common.firstNotice.title"),
+      loading: t("core.common.firstNotice.loading"),
+      emptyTitle: t("core.common.firstNotice.empty.title"),
+      emptySubtitle: t("core.common.firstNotice.empty.subtitle"),
+      apiPath: "/api/stat/first-notice",
+      showGithubButton: false,
+      showRefreshButton: false,
+      refreshLabel: "",
+    };
+  }
+
  return {
-    title: t(`${keyBase}.title`),
-    loading: t(`${keyBase}.loading`),
-    emptyTitle: t(`${keyBase}.empty.title`),
-    emptySubtitle: t(`${keyBase}.empty.subtitle`),
-    apiPath: `/api/plugin/${isChangelog ? "changelog" : "readme"}`,
+    title: t("core.common.readme.title"),
+    loading: t("core.common.readme.loading"),
+    emptyTitle: t("core.common.readme.empty.title"),
+    emptySubtitle: t("core.common.readme.empty.subtitle"),
+    apiPath: "/api/plugin/readme",
+    showGithubButton: true,
+    showRefreshButton: true,
+    refreshLabel: t("core.common.readme.buttons.refresh"),
  };
 });

+const requiresPluginName = computed(
+  () => props.mode === "readme" || props.mode === "changelog",
+);
+
 async function fetchContent() {
-  if (!props.pluginName) return;
+  if (requiresPluginName.value && !props.pluginName) return;
  const requestId = ++lastRequestId.value;
  loading.value = true;
  content.value = null;
@@ -186,9 +217,13 @@ async function fetchContent() {
  isEmpty.value = false;

  try {
-    const res = await axios.get(
-      `${modeConfig.value.apiPath}?name=${props.pluginName}`,
-    );
+    let params;
+    if (requiresPluginName.value) {
+      params = { name: props.pluginName };
+    } else if (props.mode === "first-notice") {
+      params = { locale: locale.value };
+    }
+    const res = await axios.get(modeConfig.value.apiPath, { params });
    if (requestId !== lastRequestId.value) return;

    if (res.data.status === "ok") {
@@ -207,7 +242,9 @@ async function fetchContent() {
 watch(
  [() => props.show, () => props.pluginName, () => props.mode],
  ([show, name]) => {
-    if (show && name) fetchContent();
+    if (!show) return;
+    if (requiresPluginName.value && !name) return;
+    fetchContent();
  },
  { immediate: true },
 );
@@ -273,22 +310,26 @@ function openExternalLink(url) {
  if (!url) return;
  window.open(url, "_blank", "noopener,noreferrer");
 }
+
+const showActionArea = computed(() => {
+  const hasGithub = modeConfig.value.showGithubButton && !!props.repoUrl;
+  return hasGithub || modeConfig.value.showRefreshButton;
+});
 </script>

 <template>
  <v-dialog v-model="_show" width="800">
    <v-card>
      <v-card-title class="d-flex justify-space-between align-center">
-        <span class="text-h5">{{ modeConfig.title }}</span>
+        <span class="text-h2 pa-2">{{ modeConfig.title }}</span>
        <v-btn icon @click="_show = false" variant="text">
          <v-icon>mdi-close</v-icon>
        </v-btn>
      </v-card-title>
-      <v-divider></v-divider>
-      <v-card-text style="height: 70vh; overflow-y: auto">
-        <div class="d-flex justify-space-between mb-4">
+      <v-card-text style="overflow-y: auto">
+        <div v-if="showActionArea" class="d-flex justify-space-between mb-4">
          <v-btn
-            v-if="repoUrl"
+            v-if="modeConfig.showGithubButton && repoUrl"
            color="primary"
            prepend-icon="mdi-github"
            @click="openExternalLink(repoUrl)"
@@ -296,11 +337,12 @@ function openExternalLink(url) {
            {{ t("core.common.readme.buttons.viewOnGithub") }}
          </v-btn>
          <v-btn
+            v-if="modeConfig.showRefreshButton"
            color="secondary"
            prepend-icon="mdi-refresh"
            @click="fetchContent"
          >
-            {{ t("core.common.readme.buttons.refresh") }}
+            {{ modeConfig.refreshLabel }}
          </v-btn>
        </div>

@@ -357,7 +399,6 @@ function openExternalLink(url) {
          </p>
        </div>
      </v-card-text>
-      <v-divider></v-divider>
      <v-card-actions>
        <v-spacer></v-spacer>
        <v-btn color="primary" variant="tonal" @click="_show = false">
@@ -19,8 +19,8 @@
            :key="option.value"
            @click="addEntry(option.value)"
          >
-            <v-list-item-title>{{ option.label }}</v-list-item-title>
-            <v-list-item-subtitle v-if="option.hint">{{ option.hint }}</v-list-item-subtitle>
+            <v-list-item-title>{{ translateIfKey(option.label) }}</v-list-item-title>
+            <v-list-item-subtitle v-if="option.hint">{{ translateIfKey(option.hint) }}</v-list-item-subtitle>
          </v-list-item>
        </v-list>
      </v-menu>
@@ -58,7 +58,7 @@
          <div class="d-flex flex-column">
            <v-list-item-title class="property-name">{{ templateLabel(entry.__template_key) }}</v-list-item-title>
            <v-list-item-subtitle class="property-hint" v-if="getTemplate(entry)?.hint || getTemplate(entry)?.description">
-              {{ getTemplate(entry)?.hint || getTemplate(entry)?.description }}
+              {{ translateIfKey(getTemplate(entry)?.hint || getTemplate(entry)?.description) }}
            </v-list-item-subtitle>
          </div>
        </div>
@@ -82,10 +82,10 @@
              >
                <div class="config-section mb-2">
                  <v-list-item-title class="config-title">
-                    {{ itemMeta?.description || itemKey }}
+                    {{ translateIfKey(itemMeta?.description) || itemKey }}
                  </v-list-item-title>
                  <v-list-item-subtitle class="config-hint" v-if="itemMeta?.hint">
-                    {{ itemMeta.hint }}
+                    {{ translateIfKey(itemMeta.hint) }}
                  </v-list-item-subtitle>
                </div>
                <div v-for="(childMeta, childKey, childIndex) in itemMeta.items" :key="childKey">
@@ -94,10 +94,10 @@
                      <v-col cols="12" sm="6" class="property-info">
                        <v-list-item density="compact">
                          <v-list-item-title class="property-name">
-                            {{ childMeta?.description || childKey }}
+                            {{ translateIfKey(childMeta?.description) || childKey }}
                          </v-list-item-title>
                          <v-list-item-subtitle class="property-hint">
-                            {{ childMeta?.hint }}
+                            {{ translateIfKey(childMeta?.hint) }}
                          </v-list-item-subtitle>
                        </v-list-item>
                      </v-col>
@@ -122,11 +122,11 @@
                  <v-col cols="12" sm="6" class="property-info">
                    <v-list-item density="compact">
                      <v-list-item-title class="property-name">
-                        <span v-if="itemMeta?.description">{{ itemMeta?.description }} <span class="property-key">({{ itemKey }})</span></span>
+                        <span v-if="itemMeta?.description">{{ translateIfKey(itemMeta?.description) }} <span class="property-key">({{ itemKey }})</span></span>
                        <span v-else>{{ itemKey }}</span>
                      </v-list-item-title>
                      <v-list-item-subtitle class="property-hint">
-                        {{ itemMeta?.hint }}
+                        {{ translateIfKey(itemMeta?.hint) }}
                      </v-list-item-subtitle>
                    </v-list-item>
                  </v-col>
@@ -153,7 +153,7 @@
 <script setup>
 import { computed, ref, watch } from 'vue'
 import ConfigItemRenderer from './ConfigItemRenderer.vue'
-import { useI18n } from '@/i18n/composables'
+import { useI18n, useModuleI18n } from '@/i18n/composables'

 const props = defineProps({
  modelValue: {
@@ -168,6 +168,7 @@ const props = defineProps({

 const emit = defineEmits(['update:modelValue'])
 const { t } = useI18n()
+const { tm, getRaw } = useModuleI18n('features/config-metadata')

 const expandedEntries = ref({})

@@ -195,7 +196,12 @@ const templateOptions = computed(() => {

 function templateLabel(key) {
  if (!key) return t('core.common.templateList.unknownTemplate') || '未指定模板'
-  return props.templates?.[key]?.name || key
+  return translateIfKey(props.templates?.[key]?.name || key)
+}
+
+function translateIfKey(value) {
+  if (!value || typeof value !== 'string') return value
+  return getRaw(value) ? tm(value) : value
 }

 function buildDefaults(itemsMeta = {}) {
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Soulter	572689b416	feat: implement retry mechanism for model requests in anthropic and openai providers	2026-02-17 13:18:00 +08:00
Soulter	97c9e95211	chore: ruff format	2026-02-17 02:31:38 +08:00
Soulter	a4be369e43	chore: bump version to 4.17.1	2026-02-17 02:30:13 +08:00
Soulter	bdaca78750	fix: add support for collecting data from builtin stars in electron pyinstaller build (#5145 )	2026-02-17 02:27:07 +08:00
Soulter	6326d7e4ba	fix: add MCP tools to function tool set in _plugin_tool_fix (#5144 )	2026-02-17 02:19:36 +08:00
Soulter	a809a09e55	docs: Added instructions for deploying AstrBot using AstrBot Launcher. (#5136 ) Added instructions for deploying AstrBot using AstrBot Launcher.	2026-02-16 17:06:56 +08:00
Soulter	52c4ef2d87	chore: bump version to 4.17.1	2026-02-15 23:45:34 +08:00
Soulter	52c31fabe2	fix: update retention logic in LogManager to handle backup count correctly	2026-02-15 23:42:12 +08:00
NayukiMeko	79e239ad97	fix: handle list format content from OpenAI-compatible APIs (#5128 ) * fix: handle list format content from OpenAI-compatible APIs Some LLM providers (e.g., GLM-4.5V via SiliconFlow) return content as list[dict] format like [{'type': 'text', 'text': '...'}] instead of plain string. This causes the raw list representation to be displayed to users. Changes: - Add _normalize_content() helper to extract text from various content formats - Use json.loads instead of ast.literal_eval for safer parsing - Add size limit check (8KB) before attempting JSON parsing - Only convert lists that match OpenAI content-part schema (has 'type': 'text') to avoid collapsing legitimate list-literal replies like ['foo', 'bar'] - Add strip parameter to preserve whitespace in streaming chunks - Clean up orphan </think> tags that may leak from some models Fixes #5124 * fix: improve content normalization safety - Try json.loads first, fallback to ast.literal_eval for single-quoted Python literals to avoid corrupting apostrophes (e.g., "don't") - Coerce text values to str to handle null or non-string text fields	2026-02-15 23:30:47 +08:00
Soulter	8abaf1015d	chore: bump version to 4.17.0	2026-02-15 21:51:00 +08:00
Soulter	9a0c814fd4	feat: add SSL configuration options for WebUI and update related logging (#5117 )	2026-02-15 17:43:36 +08:00
Soulter	c64e1b42a4	feat: replace colorlog with loguru for enhanced logging support (#5115 )	2026-02-15 17:11:03 +08:00
Soulter	2d23c36067	feat: add Afdian support card to resources section in WelcomePage	2026-02-15 16:20:34 +08:00
Soulter	754144ad99	feat: add fallback chat model chain in tool loop runner (#5109 ) * feat: implement fallback provider support for chat models and update configuration * feat: enhance provider selection display with count and chips for selected providers * feat: update fallback chat providers to use provider settings and add warning for non-list fallback models	2026-02-15 11:51:34 +08:00
Waterwzy	0faf109c2a	feat: support hot reload after plugin load failure (#5043 ) * add :Support hot reload after plugin load failure * Apply suggestions from code review Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> * fix:reformat code * fix:reformat code --------- Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>	2026-02-13 18:37:20 +08:00
evpeople	7d1eff3ec4	fix #5089 : add uv lock step in Dockerfile before export (#5091 ) Co-authored-by: Soulter <37870767+Soulter@users.noreply.github.com>	2026-02-13 18:34:26 +08:00
Soulter	e295c470a5	fix: remove unnecessary frozen flag from requirements export in Dockerfile fixes: #5089	2026-02-13 18:09:49 +08:00
Li-shi-ling	935168c024	fix-correct-FIRST_NOTICE.md-locale-path-resolution (#5083 ) (#5082 ) * fix:修改配置文件目录 * fix:添加备选的FIRST_NOTICE.zh-CN.md用于兼容	2026-02-13 13:15:08 +08:00
Soulter	f44961d065	feat: add LINE platform support with adapter and configuration (#5085 )	2026-02-13 13:01:48 +08:00
Soulter	0c7a95ccd8	chore: bump version to 4.16.0 (#5074 )	2026-02-12 22:55:42 +08:00
Soulter	09215bad57	fix: add config tabs and routing for normal and system configurations	2026-02-12 22:48:17 +08:00
エイカク	4ff07e3c74	fix: 完善转发引用解析与图片回退并支持配置化控制 (#5054 ) * feat: support fallback image parsing for quoted messages * fix: fallback parse quoted images when reply chain has placeholders * style: format network utils with ruff * test: expand quoted parser coverage and improve fallback diagnostics * fix: fallback to text-only retry when image requests fail * fix: tighten image fallback and resolve nested quoted forwards * refactor: simplify quoted message extraction and dedupe images * fix: harden quoted parsing and openai error candidates * fix: harden quoted image ref normalization * refactor: organize quoted parser settings and logging * fix: cap quoted fallback images and avoid retry loops * refactor: split quoted message parser into focused modules * refactor: share onebot segment parsing logic * refactor: unify quoted message parsing flow * feat: move quoted parser tuning to provider settings * fix: add missing i18n metadata for quoted parser settings * chore: refine forwarded message setting labels	2026-02-12 23:42:29 +09:00
stevessr	473e01aadd	feat: add i18n supports for custom platform adapters (#5045 ) * Feat: 为插件提供的适配器的元数据&i18n提供数据通路 * chore: update docstrings with pull request references Added references to pull request 5045 in docstrings. --------- Co-authored-by: Soulter <37870767+Soulter@users.noreply.github.com>	2026-02-12 21:49:12 +08:00
Soulter	cd5312ba77	fix: chunk err when using openrouter deepseek (#5069 )	2026-02-12 19:26:12 +08:00
Soulter	d87bfb0d5d	perf: 优化分段回复间隔时间的初始化逻辑 (#5068 ) fixes: #5059	2026-02-12 19:18:23 +08:00
Soulter	d2de0ea5ad	feat: QQ 官方机器人平台支持主动推送消息、私聊场景下支持接收文件 (#5066 ) * feat: QQ 官方机器人平台支持主动推送消息、私聊场景下支持接收文件 * feat: enhance QQOfficialWebhook to remember session scenes for group, channel, and friend messages	2026-02-12 19:07:53 +08:00
xzj0898	4af064fd17	fix: 修复备份时缺失的人格文件夹映射 (#5042 )	2026-02-12 16:40:45 +08:00
letr	8ab2b515f6	docs: sync and fix readme typo (#5055 ) * docs: fix index typo * docs: fix typo in README_en.md - 移除英文README中意外出现的俄语，并替换为英语 * docs: fix html typo - remove unused '</p>' * docs: sync table with README * docs: sync README header format - keep the README header format consistent * doc: sync key features * style: format files - Fix formatting issues from previous PR * fix: correct md anchor link * docs: correct typo in README_fr.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> * docs: correct typo in README_zh-TW.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> --------- Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>	2026-02-12 16:37:25 +08:00
SnowNightt	51a1c0e375	fix: 修复更新日志、官方文档弹窗双滚动条问题 (#5060 )	2026-02-12 16:36:05 +08:00
evpeople	30a0098b2a	feat: add send_chat_action for Telegram platform adapter (#5037 ) * feat: add send_chat_action for Telegram platform adapter Add typing/upload indicator when sending messages via Telegram. - Added _send_chat_action helper method for sending chat actions - Send appropriate action (typing, upload_photo, upload_document, upload_voice) before sending different message types - Support streaming mode with typing indicator - Support supergroup with message_thread_id * refactor(telegram): extract chat action helpers and add throttling - Add ACTION_BY_TYPE mapping for message type to action priority - Add _get_chat_action_for_chain() to determine action from message chain - Add _send_media_with_action() for upload → send → restore typing pattern - Add _ensure_typing() helper for typing status - Add chat action throttling (0.5s) in streaming mode to avoid rate limits - Update type annotation to ChatAction \| str for better static checking * feat(telegram): implement send_typing method for Telegram platform --------- Co-authored-by: Soulter <905617992@qq.com>	2026-02-12 14:46:06 +08:00
Soulter	e3cb9eb8af	chore: ruff format	2026-02-12 13:56:33 +08:00
Soulter	b0de33c801	fix: provider extra param dialog key display error	2026-02-12 13:09:34 +08:00
letr	bcdd8c463c	docs: clean and sync README (#5014 ) * fix: close missing div in README * fix: sync README_zh-TW with README * fix: sync README * fix: correct typo correct url in README_en README_fr README_ru * docs: sync README_en with README * Update README_en.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> --------- Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> Co-authored-by: Soulter <37870767+Soulter@users.noreply.github.com>	2026-02-12 01:17:15 +08:00
Soulter	336e2a2c40	fix: update error logging message for connection failures	2026-02-12 01:14:49 +08:00
Limitless	338d8a6610	fix: close unawaited reset coroutine on early return (#5033 ) When an OnLLMRequestEvent hook stops event propagation, the reset_coro created by build_main_agent was never awaited, causing a RuntimeWarning. Close the coroutine explicitly before returning. Fixes #5032 Co-authored-by: Limitless2023 <limitless@users.noreply.github.com>	2026-02-12 01:07:13 +08:00
Soulter	9d93bda3fe	feat: temporary file handling and introduce TempDirCleaner (#5026 ) * feat: temporary file handling and introduce TempDirCleaner - Updated various modules to use `get_astrbot_temp_path()` instead of `get_astrbot_data_path()` for temporary file storage. - Renamed temporary files for better identification and organization. - Introduced `TempDirCleaner` to manage the size of the temporary directory, ensuring it does not exceed a specified limit by deleting the oldest files. - Added configuration option for maximum temporary directory size in the dashboard. - Implemented tests for `TempDirCleaner` to verify cleanup functionality and size management. * ruff	2026-02-12 01:04:48 +08:00
エイカク	a8dda20a30	fix: 提升打包版桌面端启动稳定性并优化插件依赖处理 (#5031 ) * fix(desktop): rotate electron and backend logs * refactor(desktop): centralize log rotation defaults and debug fs errors * fix(desktop): harden rotation fs ops and buffer backend log writes * refactor(desktop): extract buffered logger and reduce sync stat calls * refactor(desktop): simplify rotation flow and harden logger config * fix(desktop): make app logging async and flush-safe * fix: harden app log path switching and debug-gated rotation errors * fix: cap buffered log chunk size during path switch * fix: avoid redundant plugin reinstall and upgrade electron * fix: stop webchat tasks cleanly and bind packaged backend to localhost * fix: unify platform shutdown and await webchat listener cleanup * fix: improve startup logs for dashboard and onebot listeners * fix: revert extra startup service logs * fix: harden plugin import recovery and webchat listener cleanup * fix: pin dashboard ci node version to 24.13.0 * fix: avoid duplicate webchat listener cleanup on terminate * refactor: clarify platform task lifecycle management * fix: continue platform shutdown when terminate fails	2026-02-12 01:04:04 +08:00
Soulter	cd7755fe07	feat: add first notice feature with multilingual support and UI integration	2026-02-12 00:00:53 +08:00
エイカク	dc995af34b	fix(desktop): 为 Electron 与后端日志增加按大小轮转 (#5029 ) * fix(desktop): rotate electron and backend logs * refactor(desktop): centralize log rotation defaults and debug fs errors * fix(desktop): harden rotation fs ops and buffer backend log writes * refactor(desktop): extract buffered logger and reduce sync stat calls * refactor(desktop): simplify rotation flow and harden logger config * fix(desktop): make app logging async and flush-safe * fix: harden app log path switching and debug-gated rotation errors * fix: cap buffered log chunk size during path switch	2026-02-11 20:17:57 +09:00
LIghtJUNction	331ada02fd	docs: add AUR installation method (#4879 ) * docs: sync system package manager installation instructions to all languages * Update README.md Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> * Update README.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> * fix/typo * refactor: update system package manager installation instructions for Arch Linux across multiple language README files * feat: add installation command for AstrBot in multiple language README files --------- Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> Co-authored-by: Soulter <37870767+Soulter@users.noreply.github.com> Co-authored-by: Soulter <905617992@qq.com>	2026-02-11 13:36:17 +08:00
Soulter	80e1231e9a	feat: adding support for media and quoted message attachments for feishu (#5018 )	2026-02-11 13:26:27 +08:00
エイカク	e61b29ec6a	fix: harden plugin dependency loading in frozen app runtime (#5015 ) * fix: compare plugin versions semantically in market updates * fix: prioritize plugin site-packages for in-process pip * fix: reload starlette from plugin target site-packages * fix: harden plugin dependency import precedence in frozen runtime * fix: improve plugin dependency conflict handling * refactor: simplify plugin conflict checks and version utils * fix: expand transitive plugin dependencies for conflict checks * fix: recover conflicting plugin dependencies during module prefer * fix: reuse renderer restart flow for tray backend restart * fix: add recoverable plugin dependency conflict handling * revert: remove plugin version comparison changes * fix: add missing tray restart backend labels	2026-02-11 13:01:44 +09:00