feat: add loading state for dashboard update process

Refactor: 将 LLM Request 部分抽象为 AgentRunner 并优化多轮工具调用

.github/ISSUE_TEMPLATE/PLUGIN_PUBLISH.md

@@ -0,0 +1,30 @@
+---
+name: '🥳 发布插件'
+title: "[Plugin] 插件名"
+about: 提交插件到插件市场
+labels: [ "plugin-publish" ]
+assignees: ''
+
+---
+
+欢迎发布插件到插件市场！
+
+## 插件基本信息
+
+请将插件信息填写到下方的 Json 代码块中。`tags`（插件标签）和 `social_link`（社交链接）选填。
+
+```json
+{
+  "name": "插件名",
+  "desc": "插件介绍",
+  "repo": "插件仓库链接",
+  "tags": [],
+  "social_link": ""
+}
+```
+
+## 检查
+
+- [ ] 我的插件经过完整的测试
+- [ ] 我的插件不包含恶意代码
+- [ ] 我已阅读并同意遵守该项目的 [行为准则](https://docs.github.com/zh/site-policy/github-terms/github-community-code-of-conduct)。

.github/ISSUE_TEMPLATE/PLUGIN_PUBLISH.yml

@@ -1,40 +0,0 @@
-name: '🥳 发布插件'
-title: "[Plugin] 插件名"
-description: 提交插件到插件市场
-labels: [ "plugin-publish" ]
-body:
-  - type: markdown
-    attributes:
-      value: |
-        欢迎发布插件到插件市场！请确保您的插件经过**完整的**测试。
-  
-  - type: textarea
-    attributes:
-      label: 插件仓库
-      description: 插件的 GitHub 仓库链接
-      placeholder: >
-        如 https://github.com/Soulter/astrbot-github-cards
-
-  - type: textarea
-    attributes:
-      label: 描述
-      value: |
-        插件名: 
-        插件作者: 
-        插件简介: 
-        支持的消息平台:(必填，如 QQ、微信、飞书)
-        标签:(可选)
-        社交链接:(可选, 将会在插件市场作者名称上作为可点击的链接)
-      description: 必填。请以列表的字段按顺序将插件名、插件作者、插件简介放在这里。如果您不知道支持哪些消息平台，请填写测试过的消息平台。
-
-  - type: checkboxes
-    attributes:
-      label: Code of Conduct
-      options:
-        - label: >
-            我已阅读并同意遵守该项目的 [行为准则](https://docs.github.com/zh/site-policy/github-terms/github-community-code-of-conduct)。
-          required: true
-
-  - type: markdown
-    attributes:
-      value: "❤️"

.github/workflows/auto_release.yml

@@ -23,6 +23,36 @@ jobs:
           echo "COMMIT_SHA=$(git rev-parse HEAD)" >> $GITHUB_ENV
           echo ${{ github.ref_name }} > dist/assets/version
           zip -r dist.zip dist
+      
+      - name: Upload to Cloudflare R2
+        env:
+          R2_ACCOUNT_ID: ${{ secrets.R2_ACCOUNT_ID }}
+          R2_ACCESS_KEY_ID: ${{ secrets.R2_ACCESS_KEY_ID }}
+          R2_SECRET_ACCESS_KEY: ${{ secrets.R2_SECRET_ACCESS_KEY }}
+          R2_BUCKET_NAME: "astrbot"
+          R2_OBJECT_NAME: "astrbot-webui-latest.zip"
+          VERSION_TAG: ${{ github.ref_name }}
+        run: |
+          echo "Installing rclone..."
+          curl https://rclone.org/install.sh | sudo bash
+
+          echo "Configuring rclone remote..."
+          mkdir -p ~/.config/rclone
+          cat <<EOF > ~/.config/rclone/rclone.conf
+          [r2]
+          type = s3
+          provider = Cloudflare
+          access_key_id = $R2_ACCESS_KEY_ID
+          secret_access_key = $R2_SECRET_ACCESS_KEY
+          endpoint = https://${R2_ACCOUNT_ID}.r2.cloudflarestorage.com
+          EOF
+
+          echo "Uploading dist.zip to R2 bucket: $R2_BUCKET_NAME/$R2_OBJECT_NAME"
+          mv dashboard/dist.zip dashboard/$R2_OBJECT_NAME
+          rclone copy dashboard/$R2_OBJECT_NAME r2:$R2_BUCKET_NAME --progress
+          mv dashboard/$R2_OBJECT_NAME dashboard/astrbot-webui-${VERSION_TAG}.zip
+          rclone copy dashboard/astrbot-webui-${VERSION_TAG}.zip r2:$R2_BUCKET_NAME --progress
+          mv dashboard/astrbot-webui-${VERSION_TAG}.zip dashboard/dist.zip
 
       - name: Fetch Changelog
         run: |

.github/workflows/dashboard_ci.yml

@@ -1,6 +1,10 @@
 name: AstrBot Dashboard CI
 
-on: [push]
+on:
+  push:
+    branches: [ "master" ]
+  pull_request:
+    branches: [ "master" ]
 
 jobs:
   build:

.github/workflows/docker-image.yml

@@ -11,24 +11,42 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - name: 拉取源码
+      - name: Pull The Codes
         uses: actions/checkout@v3
         with:
-          fetch-depth: 1
+          fetch-depth: 0  # Must be 0 so we can fetch tags
 
-      - name: 设置 QEMU
+      - name: Get latest tag (only on manual trigger)
+        id: get-latest-tag
+        if: github.event_name == 'workflow_dispatch'
+        run: |
+          tag=$(git describe --tags --abbrev=0)
+          echo "latest_tag=$tag" >> $GITHUB_OUTPUT
+
+      - name: Checkout to latest tag (only on manual trigger)
+        if: github.event_name == 'workflow_dispatch'
+        run: git checkout ${{ steps.get-latest-tag.outputs.latest_tag }}
+
+      - name: Set QEMU
         uses: docker/setup-qemu-action@v3
 
-      - name: 设置 Docker Buildx
+      - name: Set Docker Buildx
         uses: docker/setup-buildx-action@v3
 
-      - name: 登录到 DockerHub
+      - name: Log in to DockerHub
         uses: docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_HUB_USERNAME }}
           password: ${{ secrets.DOCKER_HUB_PASSWORD }}
 
-      - name: 构建和推送 Docker hub
+      - name: Login to GitHub Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: Soulter
+          password: ${{ secrets.GHCR_GITHUB_TOKEN }}
+
+      - name: Build and Push Docker to DockerHub and Github GHCR
         uses: docker/build-push-action@v6
         with:
           context: .
@@ -36,8 +54,9 @@ jobs:
           push: true
           tags: |
             ${{ secrets.DOCKER_HUB_USERNAME }}/astrbot:latest
-            ${{ secrets.DOCKER_HUB_USERNAME }}/astrbot:${{ github.ref_name }}
+            ${{ secrets.DOCKER_HUB_USERNAME }}/astrbot:${{ github.event_name == 'workflow_dispatch' && steps.get-latest-tag.outputs.latest_tag || github.ref_name }}
+            ghcr.io/soulter/astrbot:latest
+            ghcr.io/soulter/astrbot:${{ github.event_name == 'workflow_dispatch' && steps.get-latest-tag.outputs.latest_tag || github.ref_name }}
 
       - name: Post build notifications
         run: echo "Docker image has been built and pushed successfully"
-

Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.10-slim
+FROM python:3.11-slim
 WORKDIR /AstrBot
 
 COPY . /AstrBot/

README.md

@@ -1,6 +1,6 @@
 <p align="center">
-
-![yjtp](https://github.com/user-attachments/assets/dcc74009-c57e-4b66-9ae3-0a81fc001255)
+ 
+![AstrBot-Logo-Simplified](https://github.com/user-attachments/assets/ffd99b6b-3272-4682-beaa-6fe74250f7d9)
 
 </p>
 
@@ -31,13 +31,21 @@ AstrBot 是一个松耦合、异步、支持多消息平台部署、具有易用
 <!-- [![codecov](https://img.shields.io/codecov/c/github/soulter/astrbot?style=for-the-badge)](https://codecov.io/gh/Soulter/AstrBot)
  -->
 
-> [!NOTE]
+> [!WARNING]
 > 
-> 个人微信接入所依赖的开源项目 Gewechat 近期已停止维护，`v3.5.10` 已经支持接入 WeChatPadPro 替换 gewechat 方式。详见文档 [WeChatPadPro](https://astrbot.app/deploy/platform/wechat/wechatpadpro.html)
+> 请务必修改默认密码以及保证 AstrBot 版本 >= 3.5.13。
 
 ## ✨ 近期更新
 
-1. AstrBot 现已支持接入 [MCP](https://modelcontextprotocol.io/) 服务器！
+<details><summary>1. AstrBot 现已自带知识库能力</summary>
+
+ 📚 详见[文档](https://astrbot.app/use/knowledge-base.html)
+
+ ![image](https://github.com/user-attachments/assets/28b639b0-bb5c-4958-8e94-92ae8cfd1ab4)
+
+</details>
+
+2. AstrBot 现已支持接入 [MCP](https://modelcontextprotocol.io/) 服务器！
 
 ## ✨ 主要功能
 
@@ -45,7 +53,7 @@ AstrBot 是一个松耦合、异步、支持多消息平台部署、具有易用
 > 🪧 我们正基于前沿科研成果，设计并实现适用于角色扮演和情感陪伴的长短期记忆模型及情绪控制模型，旨在提升对话的真实性与情感表达能力。敬请期待 `v3.6.0` 版本！
 
 1. **大语言模型对话**。支持各种大语言模型，包括 OpenAI API、Google Gemini、Llama、Deepseek、ChatGLM 等，支持接入本地部署的大模型，通过 Ollama、LLMTuner。具有多轮对话、人格情境、多模态能力，支持图片理解、语音转文字（Whisper）。
-2. **多消息平台接入**。支持接入 QQ（OneBot）、QQ 频道、微信（Gewechat）、飞书、Telegram。后续将支持钉钉、Discord、WhatsApp、小爱音响。支持速率限制、白名单、关键词过滤、百度内容审核。
+2. **多消息平台接入**。支持接入 QQ（OneBot、QQ 官方机器人平台）、QQ 频道、微信、企业微信、微信公众号、飞书、Telegram、钉钉、Discord、KOOK、VoceChat。支持速率限制、白名单、关键词过滤、百度内容审核。
 3. **Agent**。原生支持部分 Agent 能力，如代码执行器、自然语言待办、网页搜索。对接 [Dify 平台](https://dify.ai/)，便捷接入 Dify 智能助手、知识库和 Dify 工作流。
 4. **插件扩展**。深度优化的插件机制，支持[开发插件](https://astrbot.app/dev/plugin.html)扩展功能，极简开发。已支持安装多个插件。
 5. **可视化管理面板**。支持可视化修改配置、插件管理、日志查看等功能，降低配置难度。集成 WebChat，可在面板上与大模型对话。
@@ -109,20 +117,24 @@ uvx astrbot init
 
 ## ⚡ 消息平台支持情况
 
-| 平台    | 支持性 | 详情 | 消息类型 |
-| -------- | ------- | ------- | ------ |
-| QQ(官方机器人接口) | ✔    | 私聊、群聊，QQ 频道私聊、群聊 | 文字、图片 |
-| QQ(OneBot)      | ✔    | 私聊、群聊 | 文字、图片、语音 |
-| 微信个人号    | ✔    | 微信个人号私聊、群聊 | 文字、图片、语音 |
-| Telegram   | ✔    | 私聊、群聊 | 文字、图片 |
-| 企业微信    | ✔    | 私聊 | 文字、图片、语音 |
-| 微信客服    | ✔    | 私聊 | 文字、图片 |
-| 飞书   | ✔    | 私聊、群聊 | 文字、图片 |
-| 钉钉   | ✔    | 私聊、群聊 | 文字、图片 |
-| 微信对话开放平台 | 🚧    | 计划内 | - |
-| Discord   | 🚧    | 计划内 | - |
-| WhatsApp   | 🚧    | 计划内 | - |
-| 小爱音响   | 🚧    | 计划内 | - |
+| 平台    | 支持性 |
+| -------- | ------- |
+| QQ(官方机器人接口) | ✔    |
+| QQ(OneBot)      | ✔    |
+| 微信个人号    | ✔    |
+| Telegram   | ✔    |
+| 企业微信    | ✔    |
+| 微信客服    | ✔    |
+| 微信公众号    | ✔    |
+| 飞书   | ✔    |
+| 钉钉   | ✔    |
+| Slack   | ✔    |
+| Discord   | ✔    |
+| [KOOK](https://github.com/wuyan1003/astrbot_plugin_kook_adapter)   | ✔    |
+| [VoceChat](https://github.com/HikariFroya/astrbot_plugin_vocechat)   | ✔    |
+| 微信对话开放平台 | 🚧    |
+| WhatsApp   | 🚧    |
+| 小爱音响   | 🚧    |
 
 ## ⚡ 提供商支持情况
 
@@ -143,6 +155,7 @@ uvx astrbot init
 | SenseVoice | ✔    | 语音转文本 | 本地部署 |
 | OpenAI TTS API | ✔    | 文本转语音 |  |
 | GSVI | ✔    | 文本转语音 | GPT-Sovits-Inference |
+| GPT-SoVITs | ✔    | 文本转语音 | GPT-Sovits-Inference |
 | FishAudio | ✔    | 文本转语音 | GPT-Sovits 作者参与的项目 |
 | Edge TTS | ✔    | 文本转语音 | Edge 浏览器的免费 TTS |
 | 阿里云百炼 TTS | ✔    | 文本转语音 |  |
@@ -171,7 +184,6 @@ pre-commit install
 
 - Star 这个项目！
 - 在[爱发电](https://afdian.com/a/soulter)支持我！
-- 在[微信](https://drive.soulter.top/f/pYfA/d903f4fa49a496fda3f16d2be9e023b5.png)支持我~
 
 ## ✨ Demo
 
@@ -211,7 +223,7 @@ _✨ WebUI ✨_
 
 此外，本项目的诞生离不开以下开源项目：
 
-- [NapNeko/NapCatQQ](https://github.com/NapNeko/NapCatQQ)
+- [NapNeko/NapCatQQ](https://github.com/NapNeko/NapCatQQ) - 伟大的猫猫框架
 - [wechatpy/wechatpy](https://github.com/wechatpy/wechatpy)
 
 ## ⭐ Star History
@@ -225,6 +237,9 @@ _✨ WebUI ✨_
 
 </div>
 
+![10k-star-banner-credit-by-kevin](https://github.com/user-attachments/assets/c97fc5fb-20b9-4bc8-9998-c20b930ab097)
+
+
 ## Disclaimer
 
 1. The project is protected under the `AGPL-v3` opensource license.

astrbot/cli/utils/plugin.py

@@ -3,7 +3,6 @@ import tempfile
 
 import httpx
 import yaml
-import re
 from enum import Enum
 from io import BytesIO
 from pathlib import Path
@@ -59,7 +58,16 @@ def get_git_repo(url: str, target_path: Path, proxy: str | None = None):
             proxy=proxy if proxy else None, follow_redirects=True
         ) as client:
             resp = client.get(download_url)
-            resp.raise_for_status()
+            if (
+                resp.status_code == 404
+                and "archive/refs/heads/master.zip" in download_url
+            ):
+                alt_url = download_url.replace("master.zip", "main.zip")
+                click.echo("master 分支不存在，尝试下载 main 分支")
+                resp = client.get(alt_url)
+                resp.raise_for_status()
+            else:
+                resp.raise_for_status()
             zip_content = BytesIO(resp.content)
         with ZipFile(zip_content) as z:
             z.extractall(temp_dir)
@@ -91,39 +99,6 @@ def load_yaml_metadata(plugin_dir: Path) -> dict:
     return {}
 
 
-def extract_py_metadata(plugin_dir: Path) -> dict:
-    """从 Python 文件中提取插件元数据
-
-    Args:
-        plugin_dir: 插件目录路径
-
-    Returns:
-        dict: 包含元数据的字典，如果提取失败则返回空字典
-    """
-    # 检查 main.py 或与目录同名的 py 文件
-    for pattern in ["main.py", f"{plugin_dir.name}.py"]:
-        for py_file in plugin_dir.glob(pattern):
-            try:
-                content = py_file.read_text(encoding="utf-8")
-                register_match = re.search(
-                    r'@register_star\s*\(\s*"([^"]+)"\s*,\s*"([^"]+)"\s*,\s*"([^"]+)"\s*,\s*"([^"]+)"(?:\s*,\s*"?([^")]+)"?)?\s*\)',
-                    content,
-                )
-                if register_match:
-                    # 映射匹配组到元数据键
-                    metadata = {}
-                    keys = ["name", "author", "desc", "version", "repo"]
-                    for i, key in enumerate(keys):
-                        if i + 1 <= len(
-                            register_match.groups()
-                        ) and register_match.group(i + 1):
-                            metadata[key] = register_match.group(i + 1)
-                    return metadata
-            except Exception as e:
-                click.echo(f"读取 {py_file} 失败: {e}", err=True)
-    return {}
-
-
 def build_plug_list(plugins_dir: Path) -> list:
     """构建插件列表，包含本地和在线插件信息
 
@@ -139,31 +114,22 @@ def build_plug_list(plugins_dir: Path) -> list:
         for plugin_name in [d.name for d in plugins_dir.glob("*") if d.is_dir()]:
             plugin_dir = plugins_dir / plugin_name
 
-            # 从不同来源加载元数据
+            # 从 metadata.yaml 加载元数据
             metadata = load_yaml_metadata(plugin_dir)
 
-            # 如果元数据不完整，尝试从 Python 文件提取
-            if not metadata or not all(
+            # 如果成功加载元数据，添加到结果列表
+            if metadata and all(
                 k in metadata for k in ["name", "desc", "version", "author", "repo"]
             ):
-                py_metadata = extract_py_metadata(plugin_dir)
-                # 合并元数据，保留已有的值
-                for key, value in py_metadata.items():
-                    if key not in metadata or not metadata[key]:
-                        metadata[key] = value
-            # 如果成功提取元数据，添加到结果列表
-            if metadata:
-                result.append(
-                    {
-                        "name": str(metadata.get("name", "")),
-                        "desc": str(metadata.get("desc", "")),
-                        "version": str(metadata.get("version", "")),
-                        "author": str(metadata.get("author", "")),
-                        "repo": str(metadata.get("repo", "")),
-                        "status": PluginStatus.INSTALLED,
-                        "local_path": str(plugin_dir),
-                    }
-                )
+                result.append({
+                    "name": str(metadata.get("name", "")),
+                    "desc": str(metadata.get("desc", "")),
+                    "version": str(metadata.get("version", "")),
+                    "author": str(metadata.get("author", "")),
+                    "repo": str(metadata.get("repo", "")),
+                    "status": PluginStatus.INSTALLED,
+                    "local_path": str(plugin_dir),
+                })
 
     # 获取在线插件列表
     online_plugins = []
@@ -173,17 +139,15 @@ def build_plug_list(plugins_dir: Path) -> list:
             resp.raise_for_status()
             data = resp.json()
             for plugin_id, plugin_info in data.items():
-                online_plugins.append(
-                    {
-                        "name": str(plugin_id),
-                        "desc": str(plugin_info.get("desc", "")),
-                        "version": str(plugin_info.get("version", "")),
-                        "author": str(plugin_info.get("author", "")),
-                        "repo": str(plugin_info.get("repo", "")),
-                        "status": PluginStatus.NOT_INSTALLED,
-                        "local_path": None,
-                    }
-                )
+                online_plugins.append({
+                    "name": str(plugin_id),
+                    "desc": str(plugin_info.get("desc", "")),
+                    "version": str(plugin_info.get("version", "")),
+                    "author": str(plugin_info.get("author", "")),
+                    "repo": str(plugin_info.get("repo", "")),
+                    "status": PluginStatus.NOT_INSTALLED,
+                    "local_path": None,
+                })
     except Exception as e:
         click.echo(f"获取在线插件列表失败: {e}", err=True)
 

astrbot/core/__init__.py

@@ -13,7 +13,6 @@ from .utils.astrbot_path import get_astrbot_data_path
 # 初始化数据存储文件夹
 os.makedirs(get_astrbot_data_path(), exist_ok=True)
 
-WEBUI_SK = "Advanced_System_for_Text_Response_and_Bot_Operations_Tool"
 DEMO_MODE = os.getenv("DEMO_MODE", False)
 
 astrbot_config = AstrBotConfig()
@@ -31,4 +30,3 @@ pip_installer = PipInstaller(
 )
 web_chat_queue = asyncio.Queue(maxsize=32)
 web_chat_back_queue = asyncio.Queue(maxsize=32)
-

astrbot/core/config/astrbot_config.py

@@ -43,6 +43,7 @@ class AstrBotConfig(dict):
             """不存在时载入默认配置"""
             with open(config_path, "w", encoding="utf-8-sig") as f:
                 json.dump(default_config, f, indent=4, ensure_ascii=False)
+                object.__setattr__(self, "first_deploy", True)  # 标记第一次部署
 
         with open(config_path, "r", encoding="utf-8-sig") as f:
             conf_str = f.read()
@@ -82,23 +83,61 @@ class AstrBotConfig(dict):
         return conf
 
     def check_config_integrity(self, refer_conf: Dict, conf: Dict, path=""):
-        """检查配置完整性，如果有新的配置项则返回 True"""
+        """检查配置完整性，如果有新的配置项或顺序不一致则返回 True"""
         has_new = False
+
+        # 创建一个新的有序字典以保持参考配置的顺序
+        new_conf = {}
+
+        # 先按照参考配置的顺序添加配置项
         for key, value in refer_conf.items():
             if key not in conf:
-                # logger.info(f"检查到配置项 {path + "." + key if path else key} 不存在，已插入默认值 {value}")
+                # 配置项不存在，插入默认值
                 path_ = path + "." + key if path else key
                 logger.info(f"检查到配置项 {path_} 不存在，已插入默认值 {value}")
-                conf[key] = value
+                new_conf[key] = value
                 has_new = True
             else:
                 if conf[key] is None:
-                    conf[key] = value
+                    # 配置项为 None，使用默认值
+                    new_conf[key] = value
                     has_new = True
                 elif isinstance(value, dict):
-                    has_new |= self.check_config_integrity(
-                        value, conf[key], path + "." + key if path else key
-                    )
+                    # 递归检查子配置项
+                    if not isinstance(conf[key], dict):
+                        # 类型不匹配，使用默认值
+                        new_conf[key] = value
+                        has_new = True
+                    else:
+                        # 递归检查并同步顺序
+                        child_has_new = self.check_config_integrity(
+                            value, conf[key], path + "." + key if path else key
+                        )
+                        new_conf[key] = conf[key]
+                        has_new |= child_has_new
+                else:
+                    # 直接使用现有配置
+                    new_conf[key] = conf[key]
+
+        # 检查是否存在参考配置中没有的配置项
+        for key in list(conf.keys()):
+            if key not in refer_conf:
+                path_ = path + "." + key if path else key
+                logger.info(f"检查到配置项 {path_} 不存在，将从当前配置中删除")
+                has_new = True
+
+        # 顺序不一致也算作变更
+        if list(conf.keys()) != list(new_conf.keys()):
+            if path:
+                logger.info(f"检查到配置项 {path} 的子项顺序不一致，已重新排序")
+            else:
+                logger.info("检查到配置项顺序不一致，已重新排序")
+            has_new = True
+
+        # 更新原始配置
+        conf.clear()
+        conf.update(new_conf)
+
         return has_new
 
     def save_config(self, replace_config: Dict = None):

astrbot/core/config/default.py

@@ -3,15 +3,17 @@
 """
 
 import os
+
 from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 
-VERSION = "3.5.10"
+VERSION = "3.5.18"
 DB_PATH = os.path.join(get_astrbot_data_path(), "data_v3.db")
 
 # 默认配置
 DEFAULT_CONFIG = {
     "config_version": 2,
     "platform_settings": {
+        "plugin_enable": [],
         "unique_session": False,
         "rate_limit": {
             "time": 60,
@@ -40,12 +42,15 @@ DEFAULT_CONFIG = {
         },
         "no_permission_reply": True,
         "empty_mention_waiting": True,
+        "empty_mention_waiting_need_reply": True,
         "friend_message_needs_wake_prefix": False,
         "ignore_bot_self_message": False,
+        "ignore_at_all": False,
     },
     "provider": [],
     "provider_settings": {
         "enable": True,
+        "default_provider_id": "",
         "wake_prefix": "",
         "web_search": False,
         "web_search_link": False,
@@ -56,7 +61,9 @@ DEFAULT_CONFIG = {
         "max_context_length": -1,
         "dequeue_context_length": 1,
         "streaming_response": False,
+        "show_tool_use_status": False,
         "streaming_segmented": False,
+        "separate_provider": False,
     },
     "provider_stt_settings": {
         "enable": False,
@@ -66,6 +73,7 @@ DEFAULT_CONFIG = {
         "enable": False,
         "provider_id": "",
         "dual_output": False,
+        "use_file_service": False,
     },
     "provider_ltm_settings": {
         "group_icl_enable": False,
@@ -91,11 +99,13 @@ DEFAULT_CONFIG = {
     "t2i_word_threshold": 150,
     "t2i_strategy": "remote",
     "t2i_endpoint": "",
+    "t2i_use_file_service": False,
     "http_proxy": "",
     "dashboard": {
         "enable": True,
         "username": "astrbot",
         "password": "77b90590a8945a7d36c963981a307dc9",
+        "jwt_secret": "",
         "host": "0.0.0.0",
         "port": 6185,
     },
@@ -120,7 +130,7 @@ CONFIG_METADATA_2 = {
                 "description": "消息平台适配器",
                 "type": "list",
                 "config_template": {
-                    "qq_official(QQ)": {
+                    "QQ 官方机器人(WebSocket)": {
                         "id": "default",
                         "type": "qq_official",
                         "enable": False,
@@ -129,7 +139,7 @@ CONFIG_METADATA_2 = {
                         "enable_group_c2c": True,
                         "enable_guild_direct_message": True,
                     },
-                    "qq_official_webhook(QQ)": {
+                    "QQ 官方机器人(Webhook)": {
                         "id": "default",
                         "type": "qq_official_webhook",
                         "enable": False,
@@ -138,7 +148,7 @@ CONFIG_METADATA_2 = {
                         "callback_server_host": "0.0.0.0",
                         "port": 6196,
                     },
-                    "aiocqhttp(OneBotv11)": {
+                    "QQ 个人号(aiocqhttp)": {
                         "id": "default",
                         "type": "aiocqhttp",
                         "enable": False,
@@ -146,7 +156,7 @@ CONFIG_METADATA_2 = {
                         "ws_reverse_port": 6199,
                         "ws_reverse_token": "",
                     },
-                    "gewechat(微信)": {
+                    "微信个人号(Gewechat)": {
                         "id": "gwchat",
                         "type": "gewechat",
                         "enable": False,
@@ -155,7 +165,7 @@ CONFIG_METADATA_2 = {
                         "host": "这里填写你的局域网IP或者公网服务器IP",
                         "port": 11451,
                     },
-                    "wechatpadpro(微信)": {
+                    "微信个人号(WeChatPadPro)": {
                         "id": "wechatpadpro",
                         "type": "wechatpadpro",
                         "enable": False,
@@ -165,7 +175,7 @@ CONFIG_METADATA_2 = {
                         "wpp_active_message_poll": False,
                         "wpp_active_message_poll_interval": 3,
                     },
-                    "weixin_official_account(微信公众平台)": {
+                    "微信公众平台": {
                         "id": "weixin_official_account",
                         "type": "weixin_official_account",
                         "enable": False,
@@ -176,8 +186,9 @@ CONFIG_METADATA_2 = {
                         "api_base_url": "https://api.weixin.qq.com/cgi-bin/",
                         "callback_server_host": "0.0.0.0",
                         "port": 6194,
+                        "active_send_mode": False,
                     },
-                    "wecom(企业微信)": {
+                    "企业微信(含微信客服)": {
                         "id": "wecom",
                         "type": "wecom",
                         "enable": False,
@@ -190,7 +201,7 @@ CONFIG_METADATA_2 = {
                         "callback_server_host": "0.0.0.0",
                         "port": 6195,
                     },
-                    "lark(飞书)": {
+                    "飞书(Lark)": {
                         "id": "lark",
                         "type": "lark",
                         "enable": False,
@@ -199,14 +210,14 @@ CONFIG_METADATA_2 = {
                         "app_secret": "",
                         "domain": "https://open.feishu.cn",
                     },
-                    "dingtalk(钉钉)": {
+                    "钉钉(DingTalk)": {
                         "id": "dingtalk",
                         "type": "dingtalk",
                         "enable": False,
                         "client_id": "",
                         "client_secret": "",
                     },
-                    "telegram": {
+                    "Telegram": {
                         "id": "telegram",
                         "type": "telegram",
                         "enable": False,
@@ -218,22 +229,70 @@ CONFIG_METADATA_2 = {
                         "telegram_command_auto_refresh": True,
                         "telegram_command_register_interval": 300,
                     },
+                    "Discord": {
+                        "id": "discord",
+                        "type": "discord",
+                        "enable": False,
+                        "discord_token": "",
+                        "discord_proxy": "",
+                        "discord_command_register": True,
+                        "discord_guild_id_for_debug": "",
+                        "discord_activity_name": "",
+                    },
+                    "Slack": {
+                        "id": "slack",
+                        "type": "slack",
+                        "enable": False,
+                        "bot_token": "",
+                        "app_token": "",
+                        "signing_secret": "",
+                        "slack_connection_mode": "socket",  # webhook, socket
+                        "slack_webhook_host": "0.0.0.0",
+                        "slack_webhook_port": 6197,
+                        "slack_webhook_path": "/astrbot-slack-webhook/callback",
+                    },
                 },
                 "items": {
+                    "slack_connection_mode": {
+                        "description": "Slack Connection Mode",
+                        "type": "string",
+                        "options": ["webhook", "socket"],
+                        "hint": "The connection mode for Slack. `webhook` uses a webhook server, `socket` uses Slack's Socket Mode.",
+                    },
+                    "slack_webhook_host": {
+                        "description": "Slack Webhook Host",
+                        "type": "string",
+                        "hint": "Only valid when Slack connection mode is `webhook`.",
+                    },
+                    "slack_webhook_port": {
+                        "description": "Slack Webhook Port",
+                        "type": "int",
+                        "hint": "Only valid when Slack connection mode is `webhook`.",
+                    },
+                    "slack_webhook_path": {
+                        "description": "Slack Webhook Path",
+                        "type": "string",
+                        "hint": "Only valid when Slack connection mode is `webhook`.",
+                    },
+                    "active_send_mode": {
+                        "description": "是否换用主动发送接口",
+                        "type": "bool",
+                        "desc": "只有企业认证的公众号才能主动发送。主动发送接口的限制会少一些。",
+                    },
                     "wpp_active_message_poll": {
-                      "description": "是否启用主动消息轮询",
-                      "type": "bool",
-                      "hint": "只有当你发现微信消息没有按时同步到 AstrBot 时，才需要启用这个功能，默认不启用。"
+                        "description": "是否启用主动消息轮询",
+                        "type": "bool",
+                        "hint": "只有当你发现微信消息没有按时同步到 AstrBot 时，才需要启用这个功能，默认不启用。",
                     },
                     "wpp_active_message_poll_interval": {
-                      "description": "主动消息轮询间隔",
-                      "type": "int",
-                      "hint": "主动消息轮询间隔，单位为秒，默认 3 秒，最大不要超过 60 秒，否则可能被认为是旧消息。"
+                        "description": "主动消息轮询间隔",
+                        "type": "int",
+                        "hint": "主动消息轮询间隔，单位为秒，默认 3 秒，最大不要超过 60 秒，否则可能被认为是旧消息。",
                     },
                     "kf_name": {
-                      "description": "微信客服账号名",
-                      "type": "string",
-                      "hint": "可选。微信客服账号名(不是 ID)。可在 https://kf.weixin.qq.com/kf/frame#/accounts 获取"
+                        "description": "微信客服账号名",
+                        "type": "string",
+                        "hint": "可选。微信客服账号名(不是 ID)。可在 https://kf.weixin.qq.com/kf/frame#/accounts 获取",
                     },
                     "telegram_token": {
                         "description": "Bot Token",
@@ -256,10 +315,10 @@ CONFIG_METADATA_2 = {
                         "hint": "Telegram 命令自动刷新间隔，单位为秒。",
                     },
                     "id": {
-                        "description": "ID",
+                        "description": "机器人名称",
                         "type": "string",
                         "obvious_hint": True,
-                        "hint": "ID 不能和其它的平台适配器重复，否则将发生严重冲突。",
+                        "hint": "机器人名称(ID)不能和其它的平台适配器重复。",
                     },
                     "type": {
                         "description": "适配器类型",
@@ -312,6 +371,25 @@ CONFIG_METADATA_2 = {
                         "hint": "请务必填对，否则 @ 机器人将无法唤醒，只能通过前缀唤醒。",
                         "obvious_hint": True,
                     },
+                    "discord_token": {
+                        "description": "Discord Bot Token",
+                        "type": "string",
+                        "hint": "在此处填入你的Discord Bot Token",
+                    },
+                    "discord_proxy": {
+                        "description": "Discord 代理地址",
+                        "type": "string",
+                        "hint": "可选的代理地址：http://ip:port",
+                    },
+                    "discord_command_register": {
+                        "description": "是否自动将插件指令注册为 Discord 斜杠指令",
+                        "type": "bool",
+                    },
+                    "discord_activity_name": {
+                        "description": "Discord 活动名称",
+                        "type": "string",
+                        "hint": "可选的 Discord 活动名称。留空则不设置活动。",
+                    },
                 },
             },
             "platform_settings": {
@@ -347,9 +425,14 @@ CONFIG_METADATA_2 = {
                         "hint": "启用后，当用户没有权限执行某个操作时，机器人会回复一条消息。",
                     },
                     "empty_mention_waiting": {
-                        "description": "只 @ 机器人是否触发等待回复",
+                        "description": "只 @ 机器人是否触发等待",
                         "type": "bool",
-                        "hint": "启用后，当消息内容只有 @ 机器人时，会触发等待回复，在 60 秒内的该用户的任意一条消息均会唤醒机器人。这在某些平台不支持 @ 和语音/图片等消息同时发送时特别有用。",
+                        "hint": "启用后，当消息内容只有 @ 机器人时，会触发等待，在 60 秒内的该用户的任意一条消息均会唤醒机器人。这在某些平台不支持 @ 和语音/图片等消息同时发送时特别有用。",
+                    },
+                    "empty_mention_waiting_need_reply": {
+                        "description": "只 @ 机器人触发等待时是否需要回复提醒",
+                        "type": "bool",
+                        "hint": "在上面一个配置项中，如果启用了触发等待，启用此项后，机器人会使用 LLM 生成一条回复。否则，将不回复而只是等待。",
                     },
                     "friend_message_needs_wake_prefix": {
                         "description": "私聊消息是否需要唤醒前缀",
@@ -359,7 +442,12 @@ CONFIG_METADATA_2 = {
                     "ignore_bot_self_message": {
                         "description": "是否忽略机器人自身的消息",
                         "type": "bool",
-                        "hint": "某些平台如 gewechat 会将自身账号在其他 APP 端发送的消息也当做消息事件下发导致给自己发消息时唤醒机器人",
+                        "hint": "某些平台会将自身账号在其他 APP 端发送的消息也当做消息事件下发导致给自己发消息时唤醒机器人",
+                    },
+                    "ignore_at_all": {
+                        "description": "是否忽略 @ 全体成员",
+                        "type": "bool",
+                        "hint": "启用后，机器人会忽略 @ 全体成员 的消息事件。",
                     },
                     "segmented_reply": {
                         "description": "分段回复",
@@ -612,6 +700,7 @@ CONFIG_METADATA_2 = {
                         "gm_resp_image_modal": False,
                         "gm_native_search": False,
                         "gm_native_coderunner": False,
+                        "gm_url_context": False,
                         "gm_safety_settings": {
                             "harassment": "BLOCK_MEDIUM_AND_ABOVE",
                             "hate_speech": "BLOCK_MEDIUM_AND_ABOVE",
@@ -682,17 +771,6 @@ CONFIG_METADATA_2 = {
                             "model": "deepseek/deepseek-r1",
                         },
                     },
-                    "LLMTuner": {
-                        "id": "llmtuner_default",
-                        "type": "llm_tuner",
-                        "provider_type": "chat_completion",
-                        "enable": True,
-                        "base_model_path": "",
-                        "adapter_model_path": "",
-                        "llmtuner_template": "",
-                        "finetuning_type": "lora",
-                        "quantization_bit": 4,
-                    },
                     "Dify": {
                         "id": "dify_app_default",
                         "type": "dify",
@@ -777,6 +855,37 @@ CONFIG_METADATA_2 = {
                         "edge-tts-voice": "zh-CN-XiaoxiaoNeural",
                         "timeout": 20,
                     },
+                    "GSV TTS(本地加载)": {
+                        "id": "gsv_tts",
+                        "enable": False,
+                        "type": "gsv_tts_selfhost",
+                        "provider_type": "text_to_speech",
+                        "api_base": "http://127.0.0.1:9880",
+                        "gpt_weights_path": "",
+                        "sovits_weights_path": "",
+                        "timeout": 60,
+                        "gsv_default_parms": {
+                            "gsv_ref_audio_path": "",
+                            "gsv_prompt_text": "",
+                            "gsv_prompt_lang": "zh",
+                            "gsv_aux_ref_audio_paths": "",
+                            "gsv_text_lang": "zh",
+                            "gsv_top_k": 5,
+                            "gsv_top_p": 1.0,
+                            "gsv_temperature": 1.0,
+                            "gsv_text_split_method": "cut3",
+                            "gsv_batch_size": 1,
+                            "gsv_batch_threshold": 0.75,
+                            "gsv_split_bucket": True,
+                            "gsv_speed_factor": 1,
+                            "gsv_fragment_interval": 0.3,
+                            "gsv_streaming_mode": False,
+                            "gsv_seed": -1,
+                            "gsv_parallel_infer": True,
+                            "gsv_repetition_penalty": 1.35,
+                            "gsv_media_type": "wav",
+                        },
+                    },
                     "GSVI TTS(API)": {
                         "id": "gsvi_tts",
                         "type": "gsvi_tts_api",
@@ -818,7 +927,7 @@ CONFIG_METADATA_2 = {
                         "azure_tts_rate": "1",
                         "azure_tts_volume": "100",
                         "azure_tts_subscription_key": "",
-                        "azure_tts_region": "eastus"
+                        "azure_tts_region": "eastus",
                     },
                     "MiniMax TTS(API)": {
                         "id": "minimax_tts",
@@ -841,44 +950,294 @@ CONFIG_METADATA_2 = {
                         "minimax-voice-english-normalization": False,
                         "timeout": 20,
                     },
+                    "火山引擎_TTS(API)": {
+                        "id": "volcengine_tts",
+                        "type": "volcengine_tts",
+                        "provider_type": "text_to_speech",
+                        "enable": False,
+                        "api_key": "",
+                        "appid": "",
+                        "volcengine_cluster": "volcano_tts",
+                        "volcengine_voice_type": "",
+                        "volcengine_speed_ratio": 1.0,
+                        "api_base": "https://openspeech.bytedance.com/api/v1/tts",
+                        "timeout": 20,
+                    },
+                    "Gemini TTS": {
+                        "id": "gemini_tts",
+                        "type": "gemini_tts",
+                        "provider_type": "text_to_speech",
+                        "enable": False,
+                        "gemini_tts_api_key": "",
+                        "gemini_tts_api_base": "",
+                        "gemini_tts_timeout": 20,
+                        "gemini_tts_model": "gemini-2.5-flash-preview-tts",
+                        "gemini_tts_prefix": "",
+                        "gemini_tts_voice_name": "Leda",
+                    },
+                    "OpenAI Embedding": {
+                        "id": "openai_embedding",
+                        "type": "openai_embedding",
+                        "provider_type": "embedding",
+                        "enable": True,
+                        "embedding_api_key": "",
+                        "embedding_api_base": "",
+                        "embedding_model": "",
+                        "embedding_dimensions": 1536,
+                        "timeout": 20,
+                    },
+                    "Gemini Embedding": {
+                        "id": "gemini_embedding",
+                        "type": "gemini_embedding",
+                        "provider_type": "embedding",
+                        "enable": True,
+                        "embedding_api_key": "",
+                        "embedding_api_base": "",
+                        "embedding_model": "gemini-embedding-exp-03-07",
+                        "embedding_dimensions": 768,
+                        "timeout": 20,
+                    },
                 },
                 "items": {
+                    "gpt_weights_path": {
+                        "description": "GPT模型文件路径",
+                        "type": "string",
+                        "hint": "即“.ckpt”后缀的文件，请使用绝对路径，路径两端不要带双引号，不填则默认用GPT_SoVITS内置的SoVITS模型(建议直接在GPT_SoVITS中改默认模型)",
+                        "obvious_hint": True,
+                    },
+                    "sovits_weights_path": {
+                        "description": "SoVITS模型文件路径",
+                        "type": "string",
+                        "hint": "即“.pth”后缀的文件，请使用绝对路径，路径两端不要带双引号，不填则默认用GPT_SoVITS内置的SoVITS模型(建议直接在GPT_SoVITS中改默认模型)",
+                        "obvious_hint": True,
+                    },
+                    "gsv_default_parms": {
+                        "description": "GPT_SoVITS默认参数",
+                        "hint": "参考音频文件路径、参考音频文本必填，其他参数根据个人爱好自行填写",
+                        "type": "object",
+                        "items": {
+                            "gsv_ref_audio_path": {
+                                "description": "参考音频文件路径",
+                                "type": "string",
+                                "hint": "必填！请使用绝对路径！路径两端不要带双引号！",
+                                "obvious_hint": True,
+                            },
+                            "gsv_prompt_text": {
+                                "description": "参考音频文本",
+                                "type": "string",
+                                "hint": "必填！请填写参考音频讲述的文本",
+                                "obvious_hint": True,
+                            },
+                            "gsv_prompt_lang": {
+                                "description": "参考音频文本语言",
+                                "type": "string",
+                                "hint": "请填写参考音频讲述的文本的语言，默认为中文",
+                            },
+                            "gsv_aux_ref_audio_paths": {
+                                "description": "辅助参考音频文件路径",
+                                "type": "string",
+                                "hint": "辅助参考音频文件，可不填",
+                            },
+                            "gsv_text_lang": {
+                                "description": "文本语言",
+                                "type": "string",
+                                "hint": "默认为中文",
+                            },
+                            "gsv_top_k": {
+                                "description": "生成语音的多样性",
+                                "type": "int",
+                                "hint": "",
+                            },
+                            "gsv_top_p": {
+                                "description": "核采样的阈值",
+                                "type": "float",
+                                "hint": "",
+                            },
+                            "gsv_temperature": {
+                                "description": "生成语音的随机性",
+                                "type": "float",
+                                "hint": "",
+                            },
+                            "gsv_text_split_method": {
+                                "description": "切分文本的方法",
+                                "type": "string",
+                                "hint": "可选值：  `cut0`：不切分    `cut1`：四句一切   `cut2`：50字一切    `cut3`：按中文句号切    `cut4`：按英文句号切    `cut5`：按标点符号切",
+                                "options": [
+                                    "cut0",
+                                    "cut1",
+                                    "cut2",
+                                    "cut3",
+                                    "cut4",
+                                    "cut5",
+                                ],
+                            },
+                            "gsv_batch_size": {
+                                "description": "批处理大小",
+                                "type": "int",
+                                "hint": "",
+                            },
+                            "gsv_batch_threshold": {
+                                "description": "批处理阈值",
+                                "type": "float",
+                                "hint": "",
+                            },
+                            "gsv_split_bucket": {
+                                "description": "将文本分割成桶以便并行处理",
+                                "type": "bool",
+                                "hint": "",
+                            },
+                            "gsv_speed_factor": {
+                                "description": "语音播放速度",
+                                "type": "float",
+                                "hint": "1为原始语速",
+                            },
+                            "gsv_fragment_interval": {
+                                "description": "语音片段之间的间隔时间",
+                                "type": "float",
+                                "hint": "",
+                            },
+                            "gsv_streaming_mode": {
+                                "description": "启用流模式",
+                                "type": "bool",
+                                "hint": "",
+                            },
+                            "gsv_seed": {
+                                "description": "随机种子",
+                                "type": "int",
+                                "hint": "用于结果的可重复性",
+                            },
+                            "gsv_parallel_infer": {
+                                "description": "并行执行推理",
+                                "type": "bool",
+                                "hint": "",
+                            },
+                            "gsv_repetition_penalty": {
+                                "description": "重复惩罚因子",
+                                "type": "float",
+                                "hint": "",
+                            },
+                            "gsv_media_type": {
+                                "description": "输出媒体的类型",
+                                "type": "string",
+                                "hint": "建议用wav",
+                            },
+                        },
+                    },
+                    "embedding_dimensions": {
+                        "description": "嵌入维度",
+                        "type": "int",
+                        "hint": "嵌入向量的维度。根据模型不同，可能需要调整，请参考具体模型的文档。此配置项请务必填写正确，否则将导致向量数据库无法正常工作。",
+                    },
+                    "embedding_model": {
+                        "description": "嵌入模型",
+                        "type": "string",
+                        "hint": "嵌入模型名称。",
+                    },
+                    "embedding_api_key": {
+                        "description": "API Key",
+                        "type": "string",
+                    },
+                    "embedding_api_base": {
+                        "description": "API Base URL",
+                        "type": "string",
+                    },
+                    "volcengine_cluster": {
+                        "type": "string",
+                        "description": "火山引擎集群",
+                        "hint": "若使用语音复刻大模型，可选volcano_icl或volcano_icl_concurr，默认使用volcano_tts",
+                    },
+                    "volcengine_voice_type": {
+                        "type": "string",
+                        "description": "火山引擎音色",
+                        "hint": "输入声音id(Voice_type)",
+                    },
+                    "volcengine_speed_ratio": {
+                        "type": "float",
+                        "description": "语速设置",
+                        "hint": "语速设置，范围为 0.2 到 3.0,默认值为 1.0",
+                    },
+                    "volcengine_volume_ratio": {
+                        "type": "float",
+                        "description": "音量设置",
+                        "hint": "音量设置，范围为 0.0 到 2.0,默认值为 1.0",
+                    },
                     "azure_tts_voice": {
                         "type": "string",
                         "description": "音色设置",
-                        "hint": "API 音色"
+                        "hint": "API 音色",
                     },
                     "azure_tts_style": {
                         "type": "string",
                         "description": "风格设置",
-                        "hint": "声音特定的讲话风格。 可以表达快乐、同情和平静等情绪。"
+                        "hint": "声音特定的讲话风格。 可以表达快乐、同情和平静等情绪。",
                     },
                     "azure_tts_role": {
                         "type": "string",
                         "description": "模仿设置（可选）",
                         "hint": "讲话角色扮演。 声音可以模仿不同的年龄和性别，但声音名称不会更改。 例如，男性语音可以提高音调和改变语调来模拟女性语音，但语音名称不会更改。 如果角色缺失或不受声音的支持，则会忽略此属性。",
-                        "options": ["Boy","Girl","YoungAdultFemale","YoungAdultMale","OlderAdultFemale","OlderAdultMale","SeniorFemale","SeniorMale","禁用"]
+                        "options": [
+                            "Boy",
+                            "Girl",
+                            "YoungAdultFemale",
+                            "YoungAdultMale",
+                            "OlderAdultFemale",
+                            "OlderAdultMale",
+                            "SeniorFemale",
+                            "SeniorMale",
+                            "禁用",
+                        ],
                     },
                     "azure_tts_rate": {
                         "type": "string",
                         "description": "语速设置",
-                        "hint": "指示文本的讲出速率。可在字词或句子层面应用语速。 速率变化应为原始音频的 0.5 到 2 倍。"
+                        "hint": "指示文本的讲出速率。可在字词或句子层面应用语速。 速率变化应为原始音频的 0.5 到 2 倍。",
                     },
                     "azure_tts_volume": {
                         "type": "string",
                         "description": "语音音量设置",
-                        "hint": "指示语音的音量级别。 可在句子层面应用音量的变化。以从 0.0 到 100.0（从最安静到最大声，例如 75）的数字表示。 默认值为 100.0。"
+                        "hint": "指示语音的音量级别。 可在句子层面应用音量的变化。以从 0.0 到 100.0（从最安静到最大声，例如 75）的数字表示。 默认值为 100.0。",
                     },
                     "azure_tts_region": {
                         "type": "string",
                         "description": "API 地区",
                         "hint": "Azure_TTS 处理数据所在区域，具体参考 https://learn.microsoft.com/zh-cn/azure/ai-services/speech-service/regions",
-                        "options": ["southafricanorth", "eastasia", "southeastasia", "australiaeast", "centralindia", "japaneast", "japanwest", "koreacentral", "canadacentral", "northeurope", "westeurope", "francecentral", "germanywestcentral", "norwayeast", "swedencentral", "switzerlandnorth", "switzerlandwest", "uksouth", "uaenorth", "brazilsouth", "qatarcentral", "centralus", "eastus", "eastus2", "northcentralus", "southcentralus", "westcentralus", "westus", "westus2", "westus3"]
+                        "options": [
+                            "southafricanorth",
+                            "eastasia",
+                            "southeastasia",
+                            "australiaeast",
+                            "centralindia",
+                            "japaneast",
+                            "japanwest",
+                            "koreacentral",
+                            "canadacentral",
+                            "northeurope",
+                            "westeurope",
+                            "francecentral",
+                            "germanywestcentral",
+                            "norwayeast",
+                            "swedencentral",
+                            "switzerlandnorth",
+                            "switzerlandwest",
+                            "uksouth",
+                            "uaenorth",
+                            "brazilsouth",
+                            "qatarcentral",
+                            "centralus",
+                            "eastus",
+                            "eastus2",
+                            "northcentralus",
+                            "southcentralus",
+                            "westcentralus",
+                            "westus",
+                            "westus2",
+                            "westus3",
+                        ],
                     },
                     "azure_tts_subscription_key": {
                         "type": "string",
                         "description": "服务订阅密钥",
-                        "hint": "Azure_TTS 服务的订阅密钥（注意不是令牌）"
+                        "hint": "Azure_TTS 服务的订阅密钥（注意不是令牌）",
                     },
                     "dashscope_tts_voice": {
                         "description": "语音合成模型",
@@ -902,6 +1261,12 @@ CONFIG_METADATA_2 = {
                         "hint": "启用后所有函数工具将全部失效",
                         "obvious_hint": True,
                     },
+                    "gm_url_context": {
+                        "description": "启用URL上下文功能",
+                        "type": "bool",
+                        "hint": "启用后所有函数工具将全部失效",
+                        "obvious_hint": True,
+                    },
                     "gm_safety_settings": {
                         "description": "安全过滤器",
                         "type": "object",
@@ -973,7 +1338,33 @@ CONFIG_METADATA_2 = {
                         "type": "string",
                         "description": "指定语言/方言",
                         "hint": "增强对指定的小语种和方言的识别能力，设置后可以提升在指定小语种/方言场景下的语音表现",
-                        "options": [ "Chinese","Chinese,Yue","English","Arabic","Russian","Spanish","French","Portuguese","German","Turkish","Dutch","Ukrainian","Vietnamese","Indonesian","Japanese","Italian","Korean","Thai","Polish","Romanian","Greek","Czech","Finnish","Hindi","auto",],
+                        "options": [
+                            "Chinese",
+                            "Chinese,Yue",
+                            "English",
+                            "Arabic",
+                            "Russian",
+                            "Spanish",
+                            "French",
+                            "Portuguese",
+                            "German",
+                            "Turkish",
+                            "Dutch",
+                            "Ukrainian",
+                            "Vietnamese",
+                            "Indonesian",
+                            "Japanese",
+                            "Italian",
+                            "Korean",
+                            "Thai",
+                            "Polish",
+                            "Romanian",
+                            "Greek",
+                            "Czech",
+                            "Finnish",
+                            "Hindi",
+                            "auto",
+                        ],
                     },
                     "minimax-voice-speed": {
                         "type": "float",
@@ -1010,7 +1401,15 @@ CONFIG_METADATA_2 = {
                         "type": "string",
                         "description": "情绪",
                         "hint": "控制合成语音的情绪",
-                        "options": ["happy","sad","angry","fearful","disgusted","surprised","neutral",],
+                        "options": [
+                            "happy",
+                            "sad",
+                            "angry",
+                            "fearful",
+                            "disgusted",
+                            "surprised",
+                            "neutral",
+                        ],
                     },
                     "minimax-voice-latex": {
                         "type": "bool",
@@ -1223,9 +1622,19 @@ CONFIG_METADATA_2 = {
                     "enable": {
                         "description": "启用大语言模型聊天",
                         "type": "bool",
-                        "hint": "如需切换大语言模型提供商，请使用 `/provider` 命令。",
+                        "hint": "如需切换大语言模型提供商，请使用 /provider 命令。",
                         "obvious_hint": True,
                     },
+                    "separate_provider": {
+                        "description": "提供商会话隔离",
+                        "type": "bool",
+                        "hint": "启用后，每个会话支持独立选择文本生成、STT、TTS 等提供商。如果会话在使用 /provider 指令时提示无权限，可以将会话加入管理员名单或者使用 /alter_cmd provider member 将指令设为非管理员指令。",
+                    },
+                    "default_provider_id": {
+                        "description": "默认模型提供商 ID",
+                        "type": "string",
+                        "hint": "可选。每个聊天会话的默认提供商 ID。",
+                    },
                     "wake_prefix": {
                         "description": "LLM 聊天额外唤醒前缀",
                         "type": "string",
@@ -1280,10 +1689,15 @@ CONFIG_METADATA_2 = {
                         "type": "bool",
                         "hint": "启用后，将会流式输出 LLM 的响应。目前仅支持 OpenAI API提供商 以及 Telegram、QQ Official 私聊 两个平台",
                     },
+                    "show_tool_use_status": {
+                        "description": "函数调用状态输出",
+                        "type": "bool",
+                        "hint": "在触发函数调用时输出其函数名和内容。",
+                    },
                     "streaming_segmented": {
                         "description": "不支持流式回复的平台分段输出",
                         "type": "bool",
-                        "hint": "启用后，若平台不支持流式回复，会分段输出。目前仅支持 aiocqhttp 和 gewechat 两个平台，不支持或无需使用流式分段输出的平台会静默忽略此选项",
+                        "hint": "启用后，若平台不支持流式回复，会分段输出。目前仅支持 aiocqhttp 两个平台，不支持或无需使用流式分段输出的平台会静默忽略此选项",
                     },
                 },
             },
@@ -1338,7 +1752,7 @@ CONFIG_METADATA_2 = {
                         "obvious_hint": True,
                     },
                     "provider_id": {
-                        "description": "提供商 ID，不填则默认第一个STT提供商",
+                        "description": "提供商 ID",
                         "type": "string",
                         "hint": "语音转文本提供商 ID。如果不填写将使用载入的第一个提供商。",
                     },
@@ -1355,7 +1769,7 @@ CONFIG_METADATA_2 = {
                         "obvious_hint": True,
                     },
                     "provider_id": {
-                        "description": "提供商 ID，不填则默认第一个TTS提供商",
+                        "description": "提供商 ID",
                         "type": "string",
                         "hint": "文本转语音提供商 ID。如果不填写将使用载入的第一个提供商。",
                     },
@@ -1365,6 +1779,11 @@ CONFIG_METADATA_2 = {
                         "hint": "启用后，Bot 将同时输出语音和文字消息。",
                         "obvious_hint": True,
                     },
+                    "use_file_service": {
+                        "description": "使用文件服务提供 TTS 语音文件",
+                        "type": "bool",
+                        "hint": "启用后，如已配置 callback_api_base ，将会使用文件服务提供TTS语音文件",
+                    },
                 },
             },
             "provider_ltm_settings": {
@@ -1481,7 +1900,7 @@ CONFIG_METADATA_2 = {
                 "description": "对外可达的回调接口地址",
                 "type": "string",
                 "obvious_hint": True,
-                "hint": "外部服务可能会通过 AstrBot 生成的回调链接（如文件下载链接）访问 AstrBot 后端。由于 AstrBot 无法自动判断部署环境中对外可达的主机地址（host），因此需要通过此配置项显式指定 “外部服务如何访问 AstrBot” 的地址。如 http://localhost:6185，https://example.com 等。"
+                "hint": "外部服务可能会通过 AstrBot 生成的回调链接（如文件下载链接）访问 AstrBot 后端。由于 AstrBot 无法自动判断部署环境中对外可达的主机地址（host），因此需要通过此配置项显式指定 “外部服务如何访问 AstrBot” 的地址。如 http://localhost:6185，https://example.com 等。",
             },
             "log_level": {
                 "description": "控制台日志级别",
@@ -1500,6 +1919,11 @@ CONFIG_METADATA_2 = {
                 "type": "string",
                 "hint": "当 t2i_strategy 为 remote 时生效。为空时使用 AstrBot API 服务",
             },
+            "t2i_use_file_service": {
+                "description": "本地文本转图像使用文件服务提供文件",
+                "type": "bool",
+                "hint": "当 t2i_strategy 为 local 并且配置 callback_api_base 时生效。是否使用文件服务提供文件。",
+            },
             "pip_install_arg": {
                 "description": "pip 安装参数",
                 "type": "string",

astrbot/core/core_lifecycle.py

@@ -1,6 +1,6 @@
 """
 Astrbot 核心生命周期管理类, 负责管理 AstrBot 的启动、停止、重启等操作。
-该类负责初始化各个组件, 包括 ProviderManager、PlatformManager、KnowledgeDBManager、ConversationManager、PluginManager、PipelineScheduler、EventBus等。
+该类负责初始化各个组件, 包括 ProviderManager、PlatformManager、ConversationManager、PluginManager、PipelineScheduler、EventBus等。
 该类还负责加载和执行插件, 以及处理事件总线的分发。
 
 工作流程:
@@ -28,7 +28,6 @@ from astrbot.core.db import BaseDatabase
 from astrbot.core.updator import AstrBotUpdator
 from astrbot.core import logger
 from astrbot.core.config.default import VERSION
-from astrbot.core.rag.knowledge_db_mgr import KnowledgeDBManager
 from astrbot.core.conversation_mgr import ConversationManager
 from astrbot.core.star.star_handler import star_handlers_registry, EventType
 from astrbot.core.star.star_handler import star_map
@@ -37,7 +36,7 @@ from astrbot.core.star.star_handler import star_map
 class AstrBotCoreLifecycle:
     """
     AstrBot 核心生命周期管理类, 负责管理 AstrBot 的启动、停止、重启等操作。
-    该类负责初始化各个组件, 包括 ProviderManager、PlatformManager、KnowledgeDBManager、ConversationManager、PluginManager、PipelineScheduler、
+    该类负责初始化各个组件, 包括 ProviderManager、PlatformManager、ConversationManager、PluginManager、PipelineScheduler、
     EventBus 等。
     该类还负责加载和执行插件, 以及处理事件总线的分发。
     """
@@ -47,14 +46,17 @@ class AstrBotCoreLifecycle:
         self.astrbot_config = astrbot_config  # 初始化配置
         self.db = db  # 初始化数据库
 
-        # 根据环境变量设置代理
-        os.environ["https_proxy"] = self.astrbot_config["http_proxy"]
-        os.environ["http_proxy"] = self.astrbot_config["http_proxy"]
+        # 设置代理
+        if self.astrbot_config.get("http_proxy", ""):
+            os.environ["https_proxy"] = self.astrbot_config["http_proxy"]
+            os.environ["http_proxy"] = self.astrbot_config["http_proxy"]
+        if proxy := os.environ.get("https_proxy"):
+            logger.debug(f"Using proxy: {proxy}")
         os.environ["no_proxy"] = "localhost"
 
     async def initialize(self):
         """
-        初始化 AstrBot 核心生命周期管理类, 负责初始化各个组件, 包括 ProviderManager、PlatformManager、KnowledgeDBManager、ConversationManager、PluginManager、PipelineScheduler、EventBus、AstrBotUpdator等。
+        初始化 AstrBot 核心生命周期管理类, 负责初始化各个组件, 包括 ProviderManager、PlatformManager、ConversationManager、PluginManager、PipelineScheduler、EventBus、AstrBotUpdator等。
         """
 
         # 初始化日志代理
@@ -73,9 +75,6 @@ class AstrBotCoreLifecycle:
         # 初始化平台管理器
         self.platform_manager = PlatformManager(self.astrbot_config, self.event_queue)
 
-        # 初始化知识库管理器
-        self.knowledge_db_manager = KnowledgeDBManager(self.astrbot_config)
-
         # 初始化对话管理器
         self.conversation_manager = ConversationManager(self.db)
 
@@ -87,7 +86,6 @@ class AstrBotCoreLifecycle:
             self.provider_manager,
             self.platform_manager,
             self.conversation_manager,
-            self.knowledge_db_manager,
         )
 
         # 初始化插件管理器

astrbot/core/db/plugin/sqlite_impl.py

@@ -1,113 +0,0 @@
-import json
-import aiosqlite
-import os
-from typing import Any
-from .plugin_storage import PluginStorage
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
-
-DBPATH = os.path.join(get_astrbot_data_path(), "plugin_data", "sqlite", "plugin_data.db")
-
-
-class SQLitePluginStorage(PluginStorage):
-    """插件数据的 SQLite 存储实现类。
-
-    该类提供异步方式将插件数据存储到 SQLite 数据库中，支持数据的增删改查操作。
-    所有数据以 (plugin, key) 作为复合主键进行索引。
-    """
-
-    _instance = None  # Standalone instance of the class
-    _db_conn = None
-    db_path = None
-
-    def __new__(cls):
-        """
-        创建或获取 SQLitePluginStorage 的单例实例。
-        如果实例已存在，则返回现有实例；否则创建一个新实例。
-        数据在 `data/plugin_data/sqlite/plugin_data.db` 下。
-        """
-        os.makedirs(os.path.dirname(DBPATH), exist_ok=True)
-        if cls._instance is None:
-            cls._instance = super(SQLitePluginStorage, cls).__new__(cls)
-            cls._instance.db_path = DBPATH
-        return cls._instance
-
-    async def _init_db(self):
-        """初始化数据库连接（只执行一次）"""
-        if SQLitePluginStorage._db_conn is None:
-            SQLitePluginStorage._db_conn = await aiosqlite.connect(self.db_path)
-            await self._setup_db()
-
-    async def _setup_db(self):
-        """
-        异步初始化数据库。
-
-        创建插件数据表，如果表不存在则创建，表结构包含 plugin、key 和 value 字段，
-        其中 plugin 和 key 组合作为主键。
-        """
-        await self._db_conn.execute("""
-            CREATE TABLE IF NOT EXISTS plugin_data (
-                plugin TEXT,
-                key TEXT,
-                value TEXT,
-                PRIMARY KEY (plugin, key)
-            )
-        """)
-        await self._db_conn.commit()
-
-    async def set(self, plugin: str, key: str, value: Any):
-        """
-        异步存储数据。
-
-        将指定插件的键值对存入数据库，如果键已存在则更新值。
-        值会被序列化为 JSON 字符串后存储。
-
-        Args:
-            plugin: 插件标识符
-            key: 数据键名
-            value: 要存储的数据值（任意类型，将被 JSON 序列化）
-        """
-        await self._init_db()
-        await self._db_conn.execute(
-            "INSERT INTO plugin_data (plugin, key, value) VALUES (?, ?, ?) "
-            "ON CONFLICT(plugin, key) DO UPDATE SET value = excluded.value",
-            (plugin, key, json.dumps(value)),
-        )
-        await self._db_conn.commit()
-
-    async def get(self, plugin: str, key: str) -> Any:
-        """
-        异步获取数据。
-
-        从数据库中获取指定插件和键名对应的值，
-        返回的值会从 JSON 字符串反序列化为原始数据类型。
-
-        Args:
-            plugin: 插件标识符
-            key: 数据键名
-
-        Returns:
-            Any: 存储的数据值，如果未找到则返回 None
-        """
-        await self._init_db()
-        async with self._db_conn.execute(
-            "SELECT value FROM plugin_data WHERE plugin = ? AND key = ?",
-            (plugin, key),
-        ) as cursor:
-            row = await cursor.fetchone()
-            return json.loads(row[0]) if row else None
-
-    async def delete(self, plugin: str, key: str):
-        """
-        异步删除数据。
-
-        从数据库中删除指定插件和键名对应的数据项。
-
-        Args:
-            plugin: 插件标识符
-            key: 要删除的数据键名
-        """
-        await self._init_db()
-        await self._db_conn.execute(
-            "DELETE FROM plugin_data WHERE plugin = ? AND key = ?", (plugin, key)
-        )
-        await self._db_conn.commit()

astrbot/core/db/sqlite.py

@@ -11,7 +11,9 @@ class SQLiteDatabase(BaseDatabase):
         super().__init__()
         self.db_path = db_path
 
-        with open(os.path.dirname(__file__) + "/sqlite_init.sql", "r") as f:
+        with open(
+            os.path.dirname(__file__) + "/sqlite_init.sql", "r", encoding="utf-8"
+        ) as f:
             sql = f.read()
 
         # 初始化数据库

astrbot/core/db/vec_db/base.py

@@ -0,0 +1,46 @@
+import abc
+from dataclasses import dataclass
+
+
+@dataclass
+class Result:
+    similarity: float
+    data: dict
+
+
+class BaseVecDB:
+    async def initialize(self):
+        """
+        初始化向量数据库
+        """
+        pass
+
+    @abc.abstractmethod
+    async def insert(self, content: str, metadata: dict = None, id: str = None) -> int:
+        """
+        插入一条文本和其对应向量，自动生成 ID 并保持一致性。
+        """
+        ...
+
+    @abc.abstractmethod
+    async def retrieve(self, query: str, top_k: int = 5) -> list[Result]:
+        """
+        搜索最相似的文档。
+        Args:
+            query (str): 查询文本
+            top_k (int): 返回的最相似文档的数量
+        Returns:
+            List[Result]: 查询结果
+        """
+        ...
+
+    @abc.abstractmethod
+    async def delete(self, doc_id: str) -> bool:
+        """
+        删除指定文档。
+        Args:
+            doc_id (str): 要删除的文档 ID
+        Returns:
+            bool: 删除是否成功
+        """
+        ...

astrbot/core/db/vec_db/faiss_impl/__init__.py

@@ -0,0 +1,3 @@
+from .vec_db import FaissVecDB
+
+__all__ = ["FaissVecDB"]

astrbot/core/db/vec_db/faiss_impl/document_storage.py

@@ -0,0 +1,121 @@
+import aiosqlite
+import os
+
+
+class DocumentStorage:
+    def __init__(self, db_path: str):
+        self.db_path = db_path
+        self.connection = None
+        self.sqlite_init_path = os.path.join(
+            os.path.dirname(__file__), "sqlite_init.sql"
+        )
+
+    async def initialize(self):
+        """Initialize the SQLite database and create the documents table if it doesn't exist."""
+        if not os.path.exists(self.db_path):
+            await self.connect()
+            async with self.connection.cursor() as cursor:
+                with open(self.sqlite_init_path, "r", encoding="utf-8") as f:
+                    sql_script = f.read()
+                await cursor.executescript(sql_script)
+            await self.connection.commit()
+        else:
+            await self.connect()
+
+    async def connect(self):
+        """Connect to the SQLite database."""
+        self.connection = await aiosqlite.connect(self.db_path)
+
+    async def get_documents(self, metadata_filters: dict, ids: list = None):
+        """Retrieve documents by metadata filters and ids.
+
+        Args:
+            metadata_filters (dict): The metadata filters to apply.
+
+        Returns:
+            list: The list of document IDs(primary key, not doc_id) that match the filters.
+        """
+        # metadata filter -> SQL WHERE clause
+        where_clauses = []
+        values = []
+        for key, val in metadata_filters.items():
+            where_clauses.append(f"json_extract(metadata, '$.{key}') = ?")
+            values.append(val)
+        if ids is not None and len(ids) > 0:
+            ids = [str(i) for i in ids if i != -1]
+            where_clauses.append("id IN ({})".format(",".join("?" * len(ids))))
+            values.extend(ids)
+        where_sql = " AND ".join(where_clauses) or "1=1"
+
+        result = []
+        async with self.connection.cursor() as cursor:
+            sql = "SELECT * FROM documents WHERE " + where_sql
+            await cursor.execute(sql, values)
+            for row in await cursor.fetchall():
+                result.append(await self.tuple_to_dict(row))
+        return result
+
+    async def get_document_by_doc_id(self, doc_id: str):
+        """Retrieve a document by its doc_id.
+
+        Args:
+            doc_id (str): The doc_id of the document to retrieve.
+
+        Returns:
+            dict: The document data.
+        """
+        async with self.connection.cursor() as cursor:
+            await cursor.execute("SELECT * FROM documents WHERE doc_id = ?", (doc_id,))
+            row = await cursor.fetchone()
+            if row:
+                return await self.tuple_to_dict(row)
+            else:
+                return None
+
+    async def update_document_by_doc_id(self, doc_id: str, new_text: str):
+        """Retrieve a document by its doc_id.
+
+        Args:
+            doc_id (str): The doc_id.
+            new_text (str): The new text to update the document with.
+        """
+        async with self.connection.cursor() as cursor:
+            await cursor.execute(
+                "UPDATE documents SET text = ? WHERE doc_id = ?", (new_text, doc_id)
+            )
+            await self.connection.commit()
+
+    async def get_user_ids(self) -> list[str]:
+        """Retrieve all user IDs from the documents table.
+
+        Returns:
+            list: A list of user IDs.
+        """
+        async with self.connection.cursor() as cursor:
+            await cursor.execute("SELECT DISTINCT user_id FROM documents")
+            rows = await cursor.fetchall()
+            return [row[0] for row in rows]
+
+    async def tuple_to_dict(self, row):
+        """Convert a tuple to a dictionary.
+
+        Args:
+            row (tuple): The row to convert.
+
+        Returns:
+            dict: The converted dictionary.
+        """
+        return {
+            "id": row[0],
+            "doc_id": row[1],
+            "text": row[2],
+            "metadata": row[3],
+            "created_at": row[4],
+            "updated_at": row[5],
+        }
+
+    async def close(self):
+        """Close the connection to the SQLite database."""
+        if self.connection:
+            await self.connection.close()
+            self.connection = None

astrbot/core/db/vec_db/faiss_impl/embedding_storage.py

@@ -0,0 +1,59 @@
+try:
+    import faiss
+except ModuleNotFoundError:
+    raise ImportError(
+        "faiss 未安装。请使用 'pip install faiss-cpu' 或 'pip install faiss-gpu' 安装。"
+    )
+import os
+import numpy as np
+
+
+class EmbeddingStorage:
+    def __init__(self, dimension: int, path: str = None):
+        self.dimension = dimension
+        self.path = path
+        self.index = None
+        if path and os.path.exists(path):
+            self.index = faiss.read_index(path)
+        else:
+            base_index = faiss.IndexFlatL2(dimension)
+            self.index = faiss.IndexIDMap(base_index)
+        self.storage = {}
+
+    async def insert(self, vector: np.ndarray, id: int):
+        """插入向量
+
+        Args:
+            vector (np.ndarray): 要插入的向量
+            id (int): 向量的ID
+        Raises:
+            ValueError: 如果向量的维度与存储的维度不匹配
+        """
+        if vector.shape[0] != self.dimension:
+            raise ValueError(
+                f"向量维度不匹配, 期望: {self.dimension}, 实际: {vector.shape[0]}"
+            )
+        self.index.add_with_ids(vector.reshape(1, -1), np.array([id]))
+        self.storage[id] = vector
+        await self.save_index()
+
+    async def search(self, vector: np.ndarray, k: int) -> tuple:
+        """搜索最相似的向量
+
+        Args:
+            vector (np.ndarray): 查询向量
+            k (int): 返回的最相似向量的数量
+        Returns:
+            tuple: (距离, 索引)
+        """
+        faiss.normalize_L2(vector)
+        distances, indices = self.index.search(vector, k)
+        return distances, indices
+
+    async def save_index(self):
+        """保存索引
+
+        Args:
+            path (str): 保存索引的路径
+        """
+        faiss.write_index(self.index, self.path)

astrbot/core/db/vec_db/faiss_impl/sqlite_init.sql

@@ -0,0 +1,17 @@
+-- 创建文档存储表，包含 faiss 中文档的 id，文档文本，create_at，updated_at
+CREATE TABLE documents (
+    id INTEGER PRIMARY KEY AUTOINCREMENT,
+    doc_id TEXT NOT NULL,
+    text TEXT NOT NULL,
+    metadata TEXT,
+    created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
+    updated_at DATETIME DEFAULT CURRENT_TIMESTAMP
+);
+
+ALTER TABLE documents
+ADD COLUMN group_id TEXT GENERATED ALWAYS AS (json_extract(metadata, '$.group_id')) STORED;
+ALTER TABLE documents
+ADD COLUMN user_id TEXT GENERATED ALWAYS AS (json_extract(metadata, '$.user_id')) STORED;
+
+CREATE INDEX idx_documents_user_id ON documents(user_id);
+CREATE INDEX idx_documents_group_id ON documents(group_id);

astrbot/core/db/vec_db/faiss_impl/vec_db.py

@@ -0,0 +1,117 @@
+import uuid
+import json
+import numpy as np
+from .document_storage import DocumentStorage
+from .embedding_storage import EmbeddingStorage
+from ..base import Result, BaseVecDB
+from astrbot.core.provider.provider import EmbeddingProvider
+
+
+class FaissVecDB(BaseVecDB):
+    """
+    A class to represent a vector database.
+    """
+
+    def __init__(
+        self,
+        doc_store_path: str,
+        index_store_path: str,
+        embedding_provider: EmbeddingProvider,
+    ):
+        self.doc_store_path = doc_store_path
+        self.index_store_path = index_store_path
+        self.embedding_provider = embedding_provider
+        self.document_storage = DocumentStorage(doc_store_path)
+        self.embedding_storage = EmbeddingStorage(
+            embedding_provider.get_dim(), index_store_path
+        )
+        self.embedding_provider = embedding_provider
+
+    async def initialize(self):
+        await self.document_storage.initialize()
+
+    async def insert(self, content: str, metadata: dict = None, id: str = None) -> int:
+        """
+        插入一条文本和其对应向量，自动生成 ID 并保持一致性。
+        """
+        metadata = metadata or {}
+        str_id = id or str(uuid.uuid4())  # 使用 UUID 作为原始 ID
+
+        vector = await self.embedding_provider.get_embedding(content)
+        vector = np.array(vector, dtype=np.float32)
+        async with self.document_storage.connection.cursor() as cursor:
+            await cursor.execute(
+                "INSERT INTO documents (doc_id, text, metadata) VALUES (?, ?, ?)",
+                (str_id, content, json.dumps(metadata)),
+            )
+            await self.document_storage.connection.commit()
+            result = await self.document_storage.get_document_by_doc_id(str_id)
+            int_id = result["id"]
+
+            # 插入向量到 FAISS
+            await self.embedding_storage.insert(vector, int_id)
+            return int_id
+
+    async def retrieve(
+        self, query: str, k: int = 5, fetch_k: int = 20, metadata_filters: dict = None
+    ) -> list[Result]:
+        """
+        搜索最相似的文档。
+
+        Args:
+            query (str): 查询文本
+            k (int): 返回的最相似文档的数量
+            fetch_k (int): 在根据 metadata 过滤前从 FAISS 中获取的数量
+            metadata_filters (dict): 元数据过滤器
+
+        Returns:
+            List[Result]: 查询结果
+        """
+        embedding = await self.embedding_provider.get_embedding(query)
+        scores, indices = await self.embedding_storage.search(
+            vector=np.array([embedding]).astype("float32"),
+            k=fetch_k if metadata_filters else k,
+        )
+        # TODO: rerank
+        if len(indices[0]) == 0 or indices[0][0] == -1:
+            return []
+        # normalize scores
+        scores[0] = 1.0 - (scores[0] / 2.0)
+        # NOTE: maybe the size is less than k.
+        fetched_docs = await self.document_storage.get_documents(
+            metadata_filters=metadata_filters or {}, ids=indices[0]
+        )
+        if not fetched_docs:
+            return []
+        result_docs = []
+
+        idx_pos = {fetch_doc["id"]: idx for idx, fetch_doc in enumerate(fetched_docs)}
+        for i, indice_idx in enumerate(indices[0]):
+            pos = idx_pos.get(indice_idx)
+            if pos is None:
+                continue
+            fetch_doc = fetched_docs[pos]
+            score = scores[0][i]
+            result_docs.append(Result(similarity=float(score), data=fetch_doc))
+        return result_docs[:k]
+
+    async def delete(self, doc_id: int):
+        """
+        删除一条文档
+        """
+        await self.document_storage.connection.execute(
+            "DELETE FROM documents WHERE doc_id = ?", (doc_id,)
+        )
+        await self.document_storage.connection.commit()
+
+    async def close(self):
+        await self.document_storage.close()
+
+    async def count_documents(self) -> int:
+        """
+        计算文档数量
+        """
+        async with self.document_storage.connection.cursor() as cursor:
+            await cursor.execute("SELECT COUNT(*) FROM documents")
+            count = await cursor.fetchone()
+            return count[0] if count else 0

astrbot/core/initial_loader.py

@@ -26,13 +26,14 @@ class InitialLoader:
     async def start(self):
         core_lifecycle = AstrBotCoreLifecycle(self.log_broker, self.db)
 
-        core_task = []
         try:
             await core_lifecycle.initialize()
-            core_task = core_lifecycle.start()
         except Exception as e:
             logger.critical(traceback.format_exc())
             logger.critical(f"😭 初始化 AstrBot 失败：{e} !!!")
+            return
+
+        core_task = core_lifecycle.start()
 
         self.dashboard_server = AstrBotDashboard(
             core_lifecycle, self.db, core_lifecycle.dashboard_shutdown_event

astrbot/core/message/components.py

@@ -102,6 +102,10 @@ class BaseMessageComponent(BaseModel):
             data[k] = v
         return {"type": self.type.lower(), "data": data}
 
+    async def to_dict(self) -> dict:
+        # 默认情况下，回退到旧的同步 toDict()
+        return self.toDict()
+
 
 class Plain(BaseMessageComponent):
     type: ComponentType = "Plain"
@@ -118,6 +122,11 @@ class Plain(BaseMessageComponent):
             self.text.replace("&", "&").replace("[", "[").replace("]", "]")
         )
 
+    def toDict(self):
+        return {"type": "text", "data": {"text": self.text.strip()}}
+
+    async def to_dict(self):
+        return {"type": "text", "data": {"text": self.text}}
 
 class Face(BaseMessageComponent):
     type: ComponentType = "Face"
@@ -235,9 +244,6 @@ class Video(BaseMessageComponent):
     path: T.Optional[str] = ""
 
     def __init__(self, file: str, **_):
-        # for k in _.keys():
-        #     if k == "c" and _[k] not in [2, 3]:
-        #         logger.warn(f"Protocol: {k}={_[k]} doesn't match values")
         super().__init__(file=file, **_)
 
     @staticmethod
@@ -250,6 +256,70 @@ class Video(BaseMessageComponent):
             return Video(file=url, **_)
         raise Exception("not a valid url")
 
+    async def convert_to_file_path(self) -> str:
+        """将这个视频统一转换为本地文件路径。这个方法避免了手动判断视频数据类型，直接返回视频数据的本地路径（如果是网络 URL，则会自动进行下载）。
+
+        Returns:
+            str: 视频的本地路径，以绝对路径表示。
+        """
+        url = self.file
+        if url and url.startswith("file:///"):
+            return url[8:]
+        elif 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}")
+            await download_file(url, video_file_path)
+            if os.path.exists(video_file_path):
+                return os.path.abspath(video_file_path)
+            else:
+                raise Exception(f"download failed: {url}")
+        elif os.path.exists(url):
+            return os.path.abspath(url)
+        else:
+            raise Exception(f"not a valid file: {url}")
+
+    async def register_to_file_service(self):
+        """
+        将视频注册到文件服务。
+
+        Returns:
+            str: 注册后的URL
+
+        Raises:
+            Exception: 如果未配置 callback_api_base
+        """
+        callback_host = astrbot_config.get("callback_api_base")
+
+        if not callback_host:
+            raise Exception("未配置 callback_api_base，文件服务不可用")
+
+        file_path = await self.convert_to_file_path()
+
+        token = await file_token_service.register_file(file_path)
+
+        logger.debug(f"已注册：{callback_host}/api/file/{token}")
+
+        return f"{callback_host}/api/file/{token}"
+
+    async def to_dict(self):
+        """需要和 toDict 区分开，toDict 是同步方法"""
+        url_or_path = self.file
+        if url_or_path.startswith("http"):
+            payload_file = url_or_path
+        elif callback_host := astrbot_config.get("callback_api_base"):
+            callback_host = str(callback_host).removesuffix("/")
+            token = await file_token_service.register_file(url_or_path)
+            payload_file = f"{callback_host}/api/file/{token}"
+            logger.debug(f"Generated video file callback link: {payload_file}")
+        else:
+            payload_file = url_or_path
+        return {
+            "type": "video",
+            "data": {
+                "file": payload_file,
+            },
+        }
+
 
 class At(BaseMessageComponent):
     type: ComponentType = "At"
@@ -259,6 +329,12 @@ class At(BaseMessageComponent):
     def __init__(self, **_):
         super().__init__(**_)
 
+    def toDict(self):
+        return {
+            "type": "at",
+            "data": {"qq": str(self.qq)},
+        }
+
 
 class AtAll(At):
     qq: str = "all"
@@ -514,27 +590,51 @@ class Node(BaseMessageComponent):
     id: T.Optional[int] = 0  # 忽略
     name: T.Optional[str] = ""  # qq昵称
     uin: T.Optional[str] = "0"  # qq号
-    content: T.Optional[T.Union[str, list, dict]] = ""  # 子消息段列表
+    content: T.Optional[list[BaseMessageComponent]] = []
     seq: T.Optional[T.Union[str, list]] = ""  # 忽略
     time: T.Optional[int] = 0  # 忽略
 
-    def __init__(self, content: T.Union[str, list, dict, "Node", T.List["Node"]], **_):
-        if isinstance(content, list):
-            _content = None
-            if all(isinstance(item, Node) for item in content):
-                _content = [node.toDict() for node in content]
-            else:
-                _content = ""
-                for chain in content:
-                    _content += chain.toString()
-            content = _content
-        elif isinstance(content, Node):
-            content = content.toDict()
+    def __init__(self, content: list[BaseMessageComponent], **_):
+        if isinstance(content, Node):
+            # back
+            content = [content]
         super().__init__(content=content, **_)
 
-    def toString(self):
-        # logger.warn("Protocol: node doesn't support stringify")
-        return ""
+    async def to_dict(self):
+        data_content = []
+        for comp in self.content:
+            if isinstance(comp, (Image, Record)):
+                # For Image and Record segments, we convert them to base64
+                bs64 = await comp.convert_to_base64()
+                data_content.append(
+                    {
+                        "type": comp.type.lower(),
+                        "data": {"file": f"base64://{bs64}"},
+                    }
+                )
+            elif isinstance(comp, Plain):
+                # For Plain segments, we need to handle the plain differently
+                d = await comp.to_dict()
+                data_content.append(d)
+            elif isinstance(comp, File):
+                # For File segments, we need to handle the file differently
+                d = await comp.to_dict()
+                data_content.append(d)
+            elif isinstance(comp, (Node, Nodes)):
+                # For Node segments, we recursively convert them to dict
+                d = await comp.to_dict()
+                data_content.append(d)
+            else:
+                d = comp.toDict()
+                data_content.append(d)
+        return {
+            "type": "node",
+            "data": {
+                "user_id": str(self.uin),
+                "nickname": self.name,
+                "content": data_content,
+            },
+        }
 
 
 class Nodes(BaseMessageComponent):
@@ -545,12 +645,20 @@ class Nodes(BaseMessageComponent):
         super().__init__(nodes=nodes, **_)
 
     def toDict(self):
+        """Deprecated. Use to_dict instead"""
         ret = {
             "messages": [],
         }
         for node in self.nodes:
             d = node.toDict()
-            d["data"]["uin"] = str(node.uin)  # 转为字符串
+            ret["messages"].append(d)
+        return ret
+
+    async def to_dict(self):
+        """将 Nodes 转换为字典格式，适用于 OneBot JSON 格式"""
+        ret = {"messages": []}
+        for node in self.nodes:
+            d = await node.to_dict()
             ret["messages"].append(d)
         return ret
 
@@ -723,6 +831,26 @@ class File(BaseMessageComponent):
 
         return f"{callback_host}/api/file/{token}"
 
+    async def to_dict(self):
+        """需要和 toDict 区分开，toDict 是同步方法"""
+        url_or_path = await self.get_file(allow_return_url=True)
+        if url_or_path.startswith("http"):
+            payload_file = url_or_path
+        elif callback_host := astrbot_config.get("callback_api_base"):
+            callback_host = str(callback_host).removesuffix("/")
+            token = await file_token_service.register_file(url_or_path)
+            payload_file = f"{callback_host}/api/file/{token}"
+            logger.debug(f"Generated file callback link: {payload_file}")
+        else:
+            payload_file = url_or_path
+        return {
+            "type": "file",
+            "data": {
+                "name": self.name,
+                "file": payload_file,
+            },
+        }
+
 
 class WechatEmoji(BaseMessageComponent):
     type: ComponentType = "WechatEmoji"

astrbot/core/message/message_event_result.py

@@ -24,6 +24,8 @@ class MessageChain:
 
     chain: List[BaseMessageComponent] = field(default_factory=list)
     use_t2i_: Optional[bool] = None  # None 为跟随用户设置
+    type: Optional[str] = None
+    """消息链承载的消息的类型。可选，用于让消息平台区分不同业务场景的消息链。"""
 
     def message(self, message: str):
         """添加一条文本消息到消息链 `chain` 中。
@@ -98,6 +100,15 @@ class MessageChain:
         self.chain.append(Image.fromFileSystem(path))
         return self
 
+    def base64_image(self, base64_str: str):
+        """添加一条图片消息（base64 编码字符串）到消息链 `chain` 中。
+        Example:
+
+            CommandResult().base64_image("iVBORw0KGgoAAAANSUhEUgAAAAUA...")
+        """
+        self.chain.append(Image.fromBase64(base64_str))
+        return self
+
     def use_t2i(self, use_t2i: bool):
         """设置是否使用文本转图片服务。
 
@@ -157,7 +168,7 @@ class ResultContentType(enum.Enum):
     """普通的消息结果"""
     STREAMING_RESULT = enum.auto()
     """调用 LLM 产生的流式结果"""
-    STREAMING_FINISH= enum.auto()
+    STREAMING_FINISH = enum.auto()
     """流式输出完成"""
 
 

astrbot/core/pipeline/context.py

@@ -1,6 +1,14 @@
+import inspect
+import traceback
+import typing as T
 from dataclasses import dataclass
 from astrbot.core.config.astrbot_config import AstrBotConfig
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.star import PluginManager
+from astrbot.api import logger
+from astrbot.core.star.star_handler import star_handlers_registry, EventType
+from astrbot.core.star.star import star_map
+from astrbot.core.message.message_event_result import MessageEventResult, CommandResult
 
 
 @dataclass
@@ -9,3 +17,91 @@ class PipelineContext:
 
     astrbot_config: AstrBotConfig  # AstrBot 配置对象
     plugin_manager: PluginManager  # 插件管理器对象
+
+    async def call_event_hook(
+        self,
+        event: AstrMessageEvent,
+        hook_type: EventType,
+        *args,
+    ):
+        platform_id = event.get_platform_id()
+        handlers = star_handlers_registry.get_handlers_by_event_type(
+            hook_type, platform_id=platform_id
+        )
+        for handler in handlers:
+            try:
+                logger.debug(
+                    f"hook(on_llm_request) -> {star_map[handler.handler_module_path].name} - {handler.handler_name}"
+                )
+                await handler.handler(event, *args)
+            except BaseException:
+                logger.error(traceback.format_exc())
+
+            if event.is_stopped():
+                logger.info(
+                    f"{star_map[handler.handler_module_path].name} - {handler.handler_name} 终止了事件传播。"
+                )
+                return
+
+    async def call_handler(
+        self,
+        event: AstrMessageEvent,
+        handler: T.Awaitable,
+        *args,
+        **kwargs,
+    ) -> T.AsyncGenerator[None, None]:
+        """执行事件处理函数并处理其返回结果
+
+        该方法负责调用处理函数并处理不同类型的返回值。它支持两种类型的处理函数:
+        1. 异步生成器: 实现洋葱模型，每次 yield 都会将控制权交回上层
+        2. 协程: 执行一次并处理返回值
+
+        Args:
+            ctx (PipelineContext): 消息管道上下文对象
+            event (AstrMessageEvent): 事件对象
+            handler (Awaitable): 事件处理函数
+
+        Returns:
+            AsyncGenerator[None, None]: 异步生成器，用于在管道中传递控制流
+        """
+        ready_to_call = None  # 一个协程或者异步生成器
+
+        trace_ = None
+
+        try:
+            ready_to_call = handler(event, *args, **kwargs)
+        except TypeError as _:
+            # 向下兼容
+            trace_ = traceback.format_exc()
+            # 以前的 handler 会额外传入一个参数, 但是 context 对象实际上在插件实例中有一份
+            ready_to_call = handler(event, self.plugin_manager.context, *args, **kwargs)
+
+        if inspect.isasyncgen(ready_to_call):
+            _has_yielded = False
+            try:
+                async for ret in ready_to_call:
+                    # 这里逐步执行异步生成器, 对于每个 yield 返回的 ret, 执行下面的代码
+                    # 返回值只能是 MessageEventResult 或者 None（无返回值）
+                    _has_yielded = True
+                    if isinstance(ret, (MessageEventResult, CommandResult)):
+                        # 如果返回值是 MessageEventResult, 设置结果并继续
+                        event.set_result(ret)
+                        yield
+                    else:
+                        # 如果返回值是 None, 则不设置结果并继续
+                        # 继续执行后续阶段
+                        yield ret
+                if not _has_yielded:
+                    # 如果这个异步生成器没有执行到 yield 分支
+                    yield
+            except Exception as e:
+                logger.error(f"Previous Error: {trace_}")
+                raise e
+        elif inspect.iscoroutine(ready_to_call):
+            # 如果只是一个协程, 直接执行
+            ret = await ready_to_call
+            if isinstance(ret, (MessageEventResult, CommandResult)):
+                event.set_result(ret)
+                yield
+            else:
+                yield ret

astrbot/core/pipeline/preprocess_stage/stage.py

@@ -43,31 +43,31 @@ class PreProcessStage(Stage):
         # STT
         if self.stt_settings.get("enable", False):
             # TODO: 独立
-            stt_provider = (
-                self.plugin_manager.context.provider_manager.curr_stt_provider_inst
-            )
-            if stt_provider:
-                message_chain = event.get_messages()
-                for idx, component in enumerate(message_chain):
-                    if isinstance(component, Record) and component.url:
-                        path = component.url.removeprefix("file://")
-                        retry = 5
-                        for i in range(retry):
-                            try:
-                                result = await stt_provider.get_text(audio_url=path)
-                                if result:
-                                    logger.info("语音转文本结果: " + result)
-                                    message_chain[idx] = Plain(result)
-                                    event.message_str += result
-                                    event.message_obj.message_str += result
-                                break
-                            except FileNotFoundError as e:
-                                # napcat workaround
-                                logger.warning(e)
-                                logger.warning(f"重试中: {i + 1}/{retry}")
-                                await asyncio.sleep(0.5)
-                                continue
-                            except BaseException as e:
-                                logger.error(traceback.format_exc())
-                                logger.error(f"语音转文本失败: {e}")
-                                break
+            ctx = self.plugin_manager.context
+            stt_provider = ctx.get_using_stt_provider(event.unified_msg_origin)
+            if not stt_provider:
+                return
+            message_chain = event.get_messages()
+            for idx, component in enumerate(message_chain):
+                if isinstance(component, Record) and component.url:
+                    path = component.url.removeprefix("file://")
+                    retry = 5
+                    for i in range(retry):
+                        try:
+                            result = await stt_provider.get_text(audio_url=path)
+                            if result:
+                                logger.info("语音转文本结果: " + result)
+                                message_chain[idx] = Plain(result)
+                                event.message_str += result
+                                event.message_obj.message_str += result
+                            break
+                        except FileNotFoundError as e:
+                            # napcat workaround
+                            logger.warning(e)
+                            logger.warning(f"重试中: {i + 1}/{retry}")
+                            await asyncio.sleep(0.5)
+                            continue
+                        except BaseException as e:
+                            logger.error(traceback.format_exc())
+                            logger.error(f"语音转文本失败: {e}")
+                            break

astrbot/core/pipeline/process_stage/agent_runner/base.py

@@ -0,0 +1,57 @@
+import abc
+import typing as T
+from dataclasses import dataclass
+from astrbot.core.provider.entities import LLMResponse
+from ....message.message_event_result import MessageChain
+from enum import Enum, auto
+
+
+class AgentState(Enum):
+    """Agent 状态枚举"""
+    IDLE = auto()          # 初始状态
+    RUNNING = auto()       # 运行中
+    DONE = auto()          # 完成
+    ERROR = auto()         # 错误状态
+
+
+class AgentResponseData(T.TypedDict):
+    chain: MessageChain
+
+
+@dataclass
+class AgentResponse:
+    type: str
+    data: AgentResponseData
+
+
+class BaseAgentRunner:
+    @abc.abstractmethod
+    async def reset(self) -> None:
+        """
+        Reset the agent to its initial state.
+        This method should be called before starting a new run.
+        """
+        ...
+
+    @abc.abstractmethod
+    async def step(self) -> T.AsyncGenerator[AgentResponse, None]:
+        """
+        Process a single step of the agent.
+        """
+        ...
+
+    @abc.abstractmethod
+    def done(self) -> bool:
+        """
+        Check if the agent has completed its task.
+        Returns True if the agent is done, False otherwise.
+        """
+        ...
+
+    @abc.abstractmethod
+    def get_final_llm_resp(self) -> LLMResponse | None:
+        """
+        Get the final observation from the agent.
+        This method should be called after the agent is done.
+        """
+        ...

astrbot/core/pipeline/process_stage/agent_runner/tool_loop_agent.py

@@ -0,0 +1,300 @@
+import sys
+import traceback
+import typing as T
+from .base import BaseAgentRunner, AgentResponse, AgentResponseData, AgentState
+from ...context import PipelineContext
+from astrbot.core.provider.provider import Provider
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
+from astrbot.core.message.message_event_result import (
+    MessageChain,
+)
+from astrbot.core.provider.entities import (
+    ProviderRequest,
+    LLMResponse,
+    ToolCallMessageSegment,
+    AssistantMessageSegment,
+    ToolCallsResult,
+)
+from mcp.types import (
+    TextContent,
+    ImageContent,
+    EmbeddedResource,
+    TextResourceContents,
+    BlobResourceContents,
+)
+from astrbot.core.star.star_handler import EventType
+from astrbot import logger
+
+if sys.version_info >= (3, 12):
+    from typing import override
+else:
+    from typing_extensions import override
+
+
+# TODO:
+# 1. 处理平台不兼容的处理器
+
+
+class ToolLoopAgent(BaseAgentRunner):
+    def __init__(
+        self, provider: Provider, event: AstrMessageEvent, pipeline_ctx: PipelineContext
+    ) -> None:
+        self.provider = provider
+        self.req = None
+        self.event = event
+        self.pipeline_ctx = pipeline_ctx
+        self._state = AgentState.IDLE
+        self.final_llm_resp = None
+        self.streaming = False
+
+    @override
+    async def reset(self, req: ProviderRequest, streaming: bool) -> None:
+        self.req = req
+        self.streaming = streaming
+        self.final_llm_resp = None
+        self._state = AgentState.IDLE
+
+    def _transition_state(self, new_state: AgentState) -> None:
+        """转换 Agent 状态"""
+        if self._state != new_state:
+            logger.debug(f"Agent state transition: {self._state} -> {new_state}")
+            self._state = new_state
+
+    async def _iter_llm_responses(self) -> T.AsyncGenerator[LLMResponse, None]:
+        """Yields chunks *and* a final LLMResponse."""
+        if self.streaming:
+            stream = self.provider.text_chat_stream(**self.req.__dict__)
+            async for resp in stream:  # type: ignore
+                yield resp
+        else:
+            yield await self.provider.text_chat(**self.req.__dict__)
+
+    @override
+    async def step(self):
+        """
+        Process a single step of the agent.
+        This method should return the result of the step.
+        """
+        if not self.req:
+            raise ValueError("Request is not set. Please call reset() first.")
+
+        # 开始处理，转换到运行状态
+        self._transition_state(AgentState.RUNNING)
+        llm_resp_result = None
+
+        async for llm_response in self._iter_llm_responses():
+            assert isinstance(llm_response, LLMResponse)
+            if llm_response.is_chunk:
+                if llm_response.result_chain:
+                    yield AgentResponse(
+                        type="streaming_delta",
+                        data=AgentResponseData(chain=llm_response.result_chain),
+                    )
+                else:
+                    yield AgentResponse(
+                        type="streaming_delta",
+                        data=AgentResponseData(
+                            chain=MessageChain().message(llm_response.completion_text)
+                        ),
+                    )
+                continue
+            llm_resp_result = llm_response
+            break  # got final response
+
+        if not llm_resp_result:
+            return
+
+        # 处理 LLM 响应
+        llm_resp = llm_resp_result
+        logger.debug(f"LLMResp: {llm_resp}")
+
+        if llm_resp.role == "err":
+            # 如果 LLM 响应错误，转换到错误状态
+            self.final_llm_resp = llm_resp
+            self._transition_state(AgentState.ERROR)
+            yield AgentResponse(
+                type="err",
+                data=AgentResponseData(
+                    chain=MessageChain().message(
+                        f"LLM 响应错误: {llm_resp.completion_text or '未知错误'}"
+                    )
+                ),
+            )
+
+        if not llm_resp.tools_call_name:
+            # 如果没有工具调用，转换到完成状态
+            self.final_llm_resp = llm_resp
+            self._transition_state(AgentState.DONE)
+
+            # 执行事件钩子
+            await self.pipeline_ctx.call_event_hook(
+                self.event, EventType.OnLLMResponseEvent, llm_resp
+            )
+
+        # 返回 LLM 结果
+        if llm_resp.result_chain:
+            yield AgentResponse(
+                type="llm_result",
+                data=AgentResponseData(chain=llm_resp.result_chain),
+            )
+        elif llm_resp.completion_text:
+            yield AgentResponse(
+                type="llm_result",
+                data=AgentResponseData(
+                    chain=MessageChain().message(llm_resp.completion_text)
+                ),
+            )
+
+        # 如果有工具调用，还需处理工具调用
+        if llm_resp.tools_call_name:
+            tool_call_result_blocks = []
+            for tool_call_name in llm_resp.tools_call_name:
+                yield AgentResponse(
+                    type="tool_call",
+                    data=AgentResponseData(
+                        chain=MessageChain().message(f"🔨 调用工具: {tool_call_name}")
+                    ),
+                )
+            async for result in self._handle_function_tools(self.req, llm_resp):
+                if isinstance(result, list):
+                    tool_call_result_blocks = result
+                elif isinstance(result, MessageChain):
+                    yield AgentResponse(
+                        type="tool_call_result",
+                        data=AgentResponseData(chain=result),
+                    )
+            # 将结果添加到上下文中
+            tool_calls_result = ToolCallsResult(
+                tool_calls_info=AssistantMessageSegment(
+                    role="assistant",
+                    tool_calls=llm_resp.to_openai_tool_calls(),
+                    content=llm_resp.completion_text,
+                ),
+                tool_calls_result=tool_call_result_blocks,
+            )
+            self.req.append_tool_calls_result(tool_calls_result)
+
+    async def _handle_function_tools(
+        self,
+        req: ProviderRequest,
+        llm_response: LLMResponse,
+    ) -> T.AsyncGenerator[MessageChain | list[ToolCallMessageSegment], None]:
+        """处理函数工具调用。"""
+        tool_call_result_blocks: list[ToolCallMessageSegment] = []
+        logger.info(f"Agent 使用工具: {llm_response.tools_call_name}")
+
+        # 执行函数调用
+        for func_tool_name, func_tool_args, func_tool_id in zip(
+            llm_response.tools_call_name,
+            llm_response.tools_call_args,
+            llm_response.tools_call_ids,
+        ):
+            try:
+                if not req.func_tool:
+                    return
+                func_tool = req.func_tool.get_func(func_tool_name)
+                if func_tool.origin == "mcp":
+                    logger.info(
+                        f"从 MCP 服务 {func_tool.mcp_server_name} 调用工具函数：{func_tool.name}，参数：{func_tool_args}"
+                    )
+                    client = req.func_tool.mcp_client_dict[func_tool.mcp_server_name]
+                    res = await client.session.call_tool(func_tool.name, func_tool_args)
+                    if not res:
+                        continue
+                    if isinstance(res.content[0], TextContent):
+                        tool_call_result_blocks.append(
+                            ToolCallMessageSegment(
+                                role="tool",
+                                tool_call_id=func_tool_id,
+                                content=res.content[0].text,
+                            )
+                        )
+                        yield MessageChain().message(res.content[0].text)
+                    elif isinstance(res.content[0], ImageContent):
+                        tool_call_result_blocks.append(
+                            ToolCallMessageSegment(
+                                role="tool",
+                                tool_call_id=func_tool_id,
+                                content="返回了图片(已直接发送给用户)",
+                            )
+                        )
+                        yield MessageChain().base64_image(res.content[0].data)
+                    elif isinstance(res.content[0], EmbeddedResource):
+                        resource = res.content[0].resource
+                        if isinstance(resource, TextResourceContents):
+                            tool_call_result_blocks.append(
+                                ToolCallMessageSegment(
+                                    role="tool",
+                                    tool_call_id=func_tool_id,
+                                    content=resource.text,
+                                )
+                            )
+                            yield MessageChain().message(resource.text)
+                        elif (
+                            isinstance(resource, BlobResourceContents)
+                            and resource.mimeType
+                            and resource.mimeType.startswith("image/")
+                        ):
+                            tool_call_result_blocks.append(
+                                ToolCallMessageSegment(
+                                    role="tool",
+                                    tool_call_id=func_tool_id,
+                                    content="返回了图片(已直接发送给用户)",
+                                )
+                            )
+                            yield MessageChain().base64_image(res.content[0].data)
+                        else:
+                            tool_call_result_blocks.append(
+                                ToolCallMessageSegment(
+                                    role="tool",
+                                    tool_call_id=func_tool_id,
+                                    content="返回的数据类型不受支持",
+                                )
+                            )
+                            yield MessageChain().message("返回的数据类型不受支持。")
+                else:
+                    logger.info(f"使用工具：{func_tool_name}，参数：{func_tool_args}")
+                    # 尝试调用工具函数
+                    wrapper = self.pipeline_ctx.call_handler(
+                        self.event, func_tool.handler, **func_tool_args
+                    )
+                    async for resp in wrapper:
+                        if resp is not None:
+                            # Tool 返回结果
+                            tool_call_result_blocks.append(
+                                ToolCallMessageSegment(
+                                    role="tool",
+                                    tool_call_id=func_tool_id,
+                                    content=resp,
+                                )
+                            )
+                            yield MessageChain().message(resp)
+                        else:
+                            # Tool 直接请求发送消息给用户
+                            # 这里我们将直接结束 Agent Loop。
+                            self._transition_state(AgentState.DONE)
+                            if res := self.event.get_result():
+                                if res.chain:
+                                    yield MessageChain(chain=res.chain)
+
+                self.event.clear_result()
+            except Exception as e:
+                logger.warning(traceback.format_exc())
+                tool_call_result_blocks.append(
+                    ToolCallMessageSegment(
+                        role="tool",
+                        tool_call_id=func_tool_id,
+                        content=f"error: {str(e)}",
+                    )
+                )
+
+        # 处理函数调用响应
+        if tool_call_result_blocks:
+            yield tool_call_result_blocks
+
+    def done(self) -> bool:
+        """检查 Agent 是否已完成工作"""
+        return self._state in (AgentState.DONE, AgentState.ERROR)
+
+    def get_final_llm_resp(self) -> LLMResponse | None:
+        return self.final_llm_resp

astrbot/core/pipeline/process_stage/method/llm_request.py

@@ -3,6 +3,7 @@
 """
 
 import traceback
+import copy
 import asyncio
 import json
 from typing import Union, AsyncGenerator
@@ -20,38 +21,27 @@ from astrbot.core.utils.metrics import Metric
 from astrbot.core.provider.entities import (
     ProviderRequest,
     LLMResponse,
-    ToolCallMessageSegment,
-    AssistantMessageSegment,
-    ToolCallsResult,
-)
-from astrbot.core.star.star_handler import star_handlers_registry, EventType
-from astrbot.core.star.star import star_map
-from mcp.types import (
-    TextContent,
-    ImageContent,
-    EmbeddedResource,
-    TextResourceContents,
-    BlobResourceContents,
 )
+from astrbot.core.star.star_handler import EventType
+from astrbot.core import web_chat_back_queue
+from ..agent_runner.tool_loop_agent import ToolLoopAgent
 
 
 class LLMRequestSubStage(Stage):
     async def initialize(self, ctx: PipelineContext) -> None:
         self.ctx = ctx
-        self.bot_wake_prefixs = ctx.astrbot_config["wake_prefix"]  # list
-        self.provider_wake_prefix = ctx.astrbot_config["provider_settings"][
-            "wake_prefix"
-        ]  # str
-        self.max_context_length = ctx.astrbot_config["provider_settings"][
-            "max_context_length"
-        ]  # int
-        self.dequeue_context_length = min(
-            max(1, ctx.astrbot_config["provider_settings"]["dequeue_context_length"]),
+        conf = ctx.astrbot_config
+        settings = conf["provider_settings"]
+        self.bot_wake_prefixs: list[str] = conf["wake_prefix"]  # list
+        self.provider_wake_prefix: str = settings["wake_prefix"]  # str
+        self.max_context_length = settings["max_context_length"]  # int
+        self.dequeue_context_length: int = min(
+            max(1, settings["dequeue_context_length"]),
             self.max_context_length - 1,
-        )  # int
-        self.streaming_response = ctx.astrbot_config["provider_settings"][
-            "streaming_response"
-        ]  # bool
+        )
+        self.streaming_response: bool = settings["streaming_response"]
+        self.max_step: int = settings.get("max_agent_step", 10)
+        self.show_tool_use: bool = settings.get("show_tool_use_status", True)
 
         for bwp in self.bot_wake_prefixs:
             if self.provider_wake_prefix.startswith(bwp):
@@ -67,7 +57,11 @@ class LLMRequestSubStage(Stage):
     ) -> Union[None, AsyncGenerator[None, None]]:
         req: ProviderRequest = None
 
-        provider = self.ctx.plugin_manager.context.get_using_provider()
+        if not self.ctx.astrbot_config["provider_settings"]["enable"]:
+            logger.debug("未启用 LLM 能力，跳过处理。")
+            return
+        umo = event.unified_msg_origin
+        provider = self.ctx.plugin_manager.context.get_using_provider(umo=umo)
         if provider is None:
             return
 
@@ -78,10 +72,7 @@ class LLMRequestSubStage(Stage):
             )
 
             if req.conversation:
-                all_contexts = json.loads(req.conversation.history)
-                req.contexts = self._process_tool_message_pairs(
-                    all_contexts, remove_tags=True
-                )
+                req.contexts = json.loads(req.conversation.history)
 
         else:
             req = ProviderRequest(prompt="", image_urls=[])
@@ -122,26 +113,7 @@ class LLMRequestSubStage(Stage):
             return
 
         # 执行请求 LLM 前事件钩子。
-        # 装饰 system_prompt 等功能
-        # 获取当前平台ID
-        platform_id = event.get_platform_id()
-        handlers = star_handlers_registry.get_handlers_by_event_type(
-            EventType.OnLLMRequestEvent, platform_id=platform_id
-        )
-        for handler in handlers:
-            try:
-                logger.debug(
-                    f"hook(on_llm_request) -> {star_map[handler.handler_module_path].name} - {handler.handler_name}"
-                )
-                await handler.handler(event, req)
-            except BaseException:
-                logger.error(traceback.format_exc())
-
-            if event.is_stopped():
-                logger.info(
-                    f"{star_map[handler.handler_module_path].name} - {handler.handler_name} 终止了事件传播。"
-                )
-                return
+        await self.ctx.call_event_hook(event, EventType.OnLLMRequestEvent, req)
 
         if isinstance(req.contexts, str):
             req.contexts = json.loads(req.contexts)
@@ -171,77 +143,62 @@ class LLMRequestSubStage(Stage):
         if not req.session_id:
             req.session_id = event.unified_msg_origin
 
-        async def requesting(req: ProviderRequest):
-            try:
-                need_loop = True
-                while need_loop:
-                    need_loop = False
-                    logger.debug(f"提供商请求 Payload: {req}")
+        # fix messages
+        req.contexts = self.fix_messages(req.contexts)
 
-                    final_llm_response = None
+        # Call Agent
+        tool_loop_agent = ToolLoopAgent(
+            provider=provider,
+            event=event,
+            pipeline_ctx=self.ctx,
+        )
+        await tool_loop_agent.reset(req=req, streaming=self.streaming_response)
 
-                    if self.streaming_response:
-                        stream = provider.text_chat_stream(**req.__dict__)
-                        async for llm_response in stream:
-                            if llm_response.is_chunk:
-                                if llm_response.result_chain:
-                                    yield llm_response.result_chain  # MessageChain
-                                else:
-                                    yield MessageChain().message(
-                                        llm_response.completion_text
-                                    )
-                            else:
-                                final_llm_response = llm_response
-                    else:
-                        final_llm_response = await provider.text_chat(
-                            **req.__dict__
-                        )  # 请求 LLM
+        async def requesting():
+            step_idx = 0
+            while step_idx < self.max_step:
+                step_idx += 1
+                try:
+                    async for resp in tool_loop_agent.step():
+                        if resp.type == "tool_call_result":
+                            continue  # 跳过工具调用结果
+                        if resp.type == "tool_call":
+                            if self.streaming_response:
+                                # 用来标记流式响应需要分节
+                                yield MessageChain(chain=[], type="break")
+                            if self.show_tool_use or event.get_platform_name() == "webchat":
+                                resp.data["chain"].type = "tool_call"
+                                await event.send(resp.data["chain"])
+                            continue
 
-                    if not final_llm_response:
-                        raise Exception("LLM response is None.")
+                        if not self.streaming_response:
+                            content_typ = (
+                                ResultContentType.LLM_RESULT
+                                if resp.type == "llm_result"
+                                else ResultContentType.GENERAL_RESULT
+                            )
+                            event.set_result(
+                                MessageEventResult(
+                                    chain=resp.data["chain"].chain,
+                                    result_content_type=content_typ,
+                                )
+                            )
+                            yield
+                            event.clear_result()
+                        else:
+                            if resp.type == "streaming_delta":
+                                yield resp.data["chain"]  # MessageChain
+                    if tool_loop_agent.done():
+                        break
 
-                    # 执行 LLM 响应后的事件钩子。
-                    handlers = star_handlers_registry.get_handlers_by_event_type(
-                        EventType.OnLLMResponseEvent
+                except Exception as e:
+                    logger.error(traceback.format_exc())
+                    event.set_result(
+                        MessageEventResult().message(
+                            f"AstrBot 请求失败。\n错误类型: {type(e).__name__}\n错误信息: {str(e)}"
+                        )
                     )
-                    for handler in handlers:
-                        try:
-                            logger.debug(
-                                f"hook(on_llm_response) -> {star_map[handler.handler_module_path].name} - {handler.handler_name}"
-                            )
-                            await handler.handler(event, final_llm_response)
-                        except BaseException:
-                            logger.error(traceback.format_exc())
-
-                        if event.is_stopped():
-                            logger.info(
-                                f"{star_map[handler.handler_module_path].name} - {handler.handler_name} 终止了事件传播。"
-                            )
-                            return
-
-                    if self.streaming_response:
-                        # 流式输出的处理
-                        async for result in self._handle_llm_stream_response(
-                            event, req, final_llm_response
-                        ):
-                            if isinstance(result, ProviderRequest):
-                                # 有函数工具调用并且返回了结果，我们需要再次请求 LLM
-                                req = result
-                                need_loop = True
-                            else:
-                                yield
-                    else:
-                        # 非流式输出的处理
-                        async for result in self._handle_llm_response(
-                            event, req, final_llm_response
-                        ):
-                            if isinstance(result, ProviderRequest):
-                                # 有函数工具调用并且返回了结果，我们需要再次请求 LLM
-                                req = result
-                                need_loop = True
-                            else:
-                                yield
-
+                    return
                 asyncio.create_task(
                     Metric.upload(
                         llm_tick=1,
@@ -250,357 +207,134 @@ class LLMRequestSubStage(Stage):
                     )
                 )
 
-                # 保存到历史记录
-                await self._save_to_history(event, req, final_llm_response)
-
-            except BaseException as e:
-                logger.error(traceback.format_exc())
-                event.set_result(
-                    MessageEventResult().message(
-                        f"AstrBot 请求失败。\n错误类型: {type(e).__name__}\n错误信息: {str(e)}"
-                    )
-                )
-
-        if not self.streaming_response:
-            event.set_extra("tool_call_result", None)
-            async for _ in requesting(req):
-                yield
-        else:
+        if self.streaming_response:
+            # 流式响应
             event.set_result(
                 MessageEventResult()
                 .set_result_content_type(ResultContentType.STREAMING_RESULT)
-                .set_async_stream(requesting(req))
+                .set_async_stream(requesting())
             )
-            # 这里使用yield来暂停当前阶段，等待流式输出完成后继续处理
             yield
-
-            if event.get_extra("tool_call_result"):
-                event.set_result(event.get_extra("tool_call_result"))
-                event.set_extra("tool_call_result", None)
+            if tool_loop_agent.done():
+                if final_llm_resp := tool_loop_agent.get_final_llm_resp():
+                    if final_llm_resp.completion_text:
+                        chain = (
+                            MessageChain().message(final_llm_resp.completion_text).chain
+                        )
+                    else:
+                        chain = final_llm_resp.result_chain.chain
+                    event.set_result(
+                        MessageEventResult(
+                            chain=chain,
+                            result_content_type=ResultContentType.STREAMING_FINISH,
+                        )
+                    )
+        else:
+            async for _ in requesting():
                 yield
 
-        # 暂时直接发出去
-        if img_b64 := event.get_extra("tool_call_img_respond"):
-            await event.send(MessageChain(chain=[Image.fromBase64(img_b64)]))
-            event.set_extra("tool_call_img_respond", None)
-            yield
+        # 异步处理 WebChat 特殊情况
+        if event.get_platform_name() == "webchat":
+            asyncio.create_task(self._handle_webchat(event, req))
 
-    async def _handle_llm_response(
-        self,
-        event: AstrMessageEvent,
-        req: ProviderRequest,
-        llm_response: LLMResponse,
-    ) -> AsyncGenerator[Union[None, ProviderRequest], None]:
-        """处理非流式 LLM 响应。
+        await self._save_to_history(event, req, tool_loop_agent.get_final_llm_resp())
 
-        Returns:
-            AsyncGenerator[Union[None, ProviderRequest], None]: 如果返回 ProviderRequest，表示需要再次调用 LLM
-
-        Yields:
-            Iterator[Union[None, ProviderRequest]]: 将 event 交付给下一个 stage 或者返回 ProviderRequest 表示需要再次调用 LLM
-        """
-        if llm_response.role == "assistant":
-            # text completion
-            if llm_response.result_chain:
-                event.set_result(
-                    MessageEventResult(
-                        chain=llm_response.result_chain.chain
-                    ).set_result_content_type(ResultContentType.LLM_RESULT)
-                )
-            else:
-                event.set_result(
-                    MessageEventResult()
-                    .message(llm_response.completion_text)
-                    .set_result_content_type(ResultContentType.LLM_RESULT)
-                )
-        elif llm_response.role == "err":
-            event.set_result(
-                MessageEventResult().message(
-                    f"AstrBot 请求失败。\n错误信息: {llm_response.completion_text}"
-                )
-            )
-        elif llm_response.role == "tool":
-            # 处理函数工具调用
-            async for result in self._handle_function_tools(event, req, llm_response):
-                yield result
-
-    async def _handle_llm_stream_response(
-        self,
-        event: AstrMessageEvent,
-        req: ProviderRequest,
-        llm_response: LLMResponse,
-    ) -> AsyncGenerator[Union[None, ProviderRequest], None]:
-        """处理流式 LLM 响应。
-
-        专门用于处理流式输出完成后的响应，与非流式响应处理分离。
-
-        Returns:
-            AsyncGenerator[Union[None, ProviderRequest], None]: 如果返回 ProviderRequest，表示需要再次调用 LLM
-
-        Yields:
-            Iterator[Union[None, ProviderRequest]]: 将 event 交付给下一个 stage 或者返回 ProviderRequest 表示需要再次调用 LLM
-        """
-        if llm_response.role == "assistant":
-            # text completion
-            if llm_response.result_chain:
-                event.set_result(
-                    MessageEventResult(
-                        chain=llm_response.result_chain.chain
-                    ).set_result_content_type(ResultContentType.STREAMING_FINISH)
-                )
-            else:
-                event.set_result(
-                    MessageEventResult()
-                    .message(llm_response.completion_text)
-                    .set_result_content_type(ResultContentType.STREAMING_FINISH)
-                )
-        elif llm_response.role == "err":
-            event.set_result(
-                MessageEventResult().message(
-                    f"AstrBot 请求失败。\n错误信息: {llm_response.completion_text}"
-                )
-            )
-        elif llm_response.role == "tool":
-            # 处理函数工具调用
-            async for result in self._handle_function_tools(event, req, llm_response):
-                yield result
-
-    async def _handle_function_tools(
-        self,
-        event: AstrMessageEvent,
-        req: ProviderRequest,
-        llm_response: LLMResponse,
-    ) -> AsyncGenerator[Union[None, ProviderRequest], None]:
-        """处理函数工具调用。
-
-        Returns:
-            AsyncGenerator[Union[None, ProviderRequest], None]: 如果返回 ProviderRequest，表示需要再次调用 LLM
-        """
-        # function calling
-        tool_call_result: list[ToolCallMessageSegment] = []
-        logger.info(
-            f"触发 {len(llm_response.tools_call_name)} 个函数调用: {llm_response.tools_call_name}"
+    async def _handle_webchat(self, event: AstrMessageEvent, req: ProviderRequest):
+        """处理 WebChat 平台的特殊情况，包括第一次 LLM 对话时总结对话内容生成 title"""
+        conversation = await self.conv_manager.get_conversation(
+            event.unified_msg_origin, req.conversation.cid
         )
-        for func_tool_name, func_tool_args, func_tool_id in zip(
-            llm_response.tools_call_name,
-            llm_response.tools_call_args,
-            llm_response.tools_call_ids,
-        ):
-            try:
-                func_tool = req.func_tool.get_func(func_tool_name)
-                if func_tool.origin == "mcp":
-                    logger.info(
-                        f"从 MCP 服务 {func_tool.mcp_server_name} 调用工具函数：{func_tool.name}，参数：{func_tool_args}"
-                    )
-                    client = req.func_tool.mcp_client_dict[func_tool.mcp_server_name]
-                    res = await client.session.call_tool(func_tool.name, func_tool_args)
-                    if res:
-                        # TODO 仅对ImageContent | EmbeddedResource进行了简单的Fallback
-                        if isinstance(res.content[0], TextContent):
-                            tool_call_result.append(
-                                ToolCallMessageSegment(
-                                    role="tool",
-                                    tool_call_id=func_tool_id,
-                                    content=res.content[0].text,
-                                )
-                            )
-                        elif isinstance(res.content[0], ImageContent):
-                            tool_call_result.append(
-                                ToolCallMessageSegment(
-                                    role="tool",
-                                    tool_call_id=func_tool_id,
-                                    content="返回了图片(已直接发送给用户)",
-                                )
-                            )
-                            event.set_extra(
-                                "tool_call_img_respond",
-                                res.content[0].data,
-                            )
-                        elif isinstance(res.content[0], EmbeddedResource):
-                            resource = res.content[0].resource
-                            if isinstance(resource, TextResourceContents):
-                                tool_call_result.append(
-                                    ToolCallMessageSegment(
-                                        role="tool",
-                                        tool_call_id=func_tool_id,
-                                        content=resource.text,
-                                    )
-                                )
-                            elif (
-                                isinstance(resource, BlobResourceContents)
-                                and resource.mimeType
-                                and resource.mimeType.startswith("image/")
-                            ):
-                                tool_call_result.append(
-                                    ToolCallMessageSegment(
-                                        role="tool",
-                                        tool_call_id=func_tool_id,
-                                        content="返回了图片(已直接发送给用户)",
-                                    )
-                                )
-                                event.set_extra(
-                                    "tool_call_img_respond",
-                                    res.content[0].data,
-                                )
-                            else:
-                                tool_call_result.append(
-                                    ToolCallMessageSegment(
-                                        role="tool",
-                                        tool_call_id=func_tool_id,
-                                        content="返回的数据类型不受支持",
-                                    )
-                                )
-                else:
-                    # 获取处理器，过滤掉平台不兼容的处理器
-                    platform_id = event.get_platform_id()
-                    star_md = star_map.get(func_tool.handler_module_path)
-                    if (
-                        star_md
-                        and platform_id in star_md.supported_platforms
-                        and not star_md.supported_platforms[platform_id]
-                    ):
-                        logger.debug(
-                            f"处理器 {func_tool_name}({star_md.name}) 在当前平台不兼容或者被禁用，跳过执行"
-                        )
-                        # 直接跳过，不添加任何消息到tool_call_result
-                        continue
-
-                    logger.info(
-                        f"调用工具函数：{func_tool_name}，参数：{func_tool_args}"
-                    )
-                    # 尝试调用工具函数
-                    wrapper = self._call_handler(
-                        self.ctx, event, func_tool.handler, **func_tool_args
-                    )
-                    async for resp in wrapper:
-                        if resp is not None:  # 有 return 返回
-                            tool_call_result.append(
-                                ToolCallMessageSegment(
-                                    role="tool",
-                                    tool_call_id=func_tool_id,
-                                    content=resp,
-                                )
-                            )
-                        else:
-                            res = event.get_result()
-                            if res and res.chain:
-                                event.set_extra("tool_call_result", res)
-                            yield  # 有生成器返回
-                event.clear_result()  # 清除上一个 handler 的结果
-            except BaseException as e:
-                logger.warning(traceback.format_exc())
-                tool_call_result.append(
-                    ToolCallMessageSegment(
-                        role="tool",
-                        tool_call_id=func_tool_id,
-                        content=f"error: {str(e)}",
-                    )
-                )
-        if tool_call_result:
-            # 函数调用结果
-            req.func_tool = None  # 暂时不支持递归工具调用
-            assistant_msg_seg = AssistantMessageSegment(
-                role="assistant", tool_calls=llm_response.to_openai_tool_calls()
+        if conversation and not req.conversation.title:
+            messages = json.loads(conversation.history)
+            latest_pair = messages[-2:]
+            if not latest_pair:
+                return
+            provider = self.ctx.plugin_manager.context.get_using_provider()
+            cleaned_text = "User: " + latest_pair[0].get("content", "").strip()
+            logger.debug(f"WebChat 对话标题生成请求，清理后的文本: {cleaned_text}")
+            llm_resp = await provider.text_chat(
+                system_prompt="You are expert in summarizing user's query.",
+                prompt=(
+                    f"Please summarize the following query of user:\n"
+                    f"{cleaned_text}\n"
+                    "Only output the summary within 10 words, DO NOT INCLUDE any other text."
+                    "You must use the same language as the user."
+                    "If you think the dialog is too short to summarize, only output a special mark: `None`"
+                ),
             )
-            # 在多轮 Tool 调用的情况下，这里始终保持最新的 Tool 调用结果，减少上下文长度。
-            req.tool_calls_result = ToolCallsResult(
-                tool_calls_info=assistant_msg_seg,
-                tool_calls_result=tool_call_result,
-            )
-            yield req  # 再次执行 LLM 请求
-        else:
-            if llm_response.completion_text:
-                event.set_result(
-                    MessageEventResult().message(llm_response.completion_text)
+            if llm_resp and llm_resp.completion_text:
+                logger.debug(
+                    f"WebChat 对话标题生成响应: {llm_resp.completion_text.strip()}"
                 )
+                title = llm_resp.completion_text.strip()
+                if not title or "None" == title:
+                    return
+                await self.conv_manager.update_conversation_title(
+                    event.unified_msg_origin, title=title
+                )
+                # 由于 WebChat 平台特殊性，其有两个对话，因此我们要更新两个对话的标题
+                # webchat adapter 中，session_id 的格式是 f"webchat!{username}!{cid}"
+                # TODO: 优化 WebChat 适配器的对话管理
+                if event.session_id:
+                    username, cid = event.session_id.split("!")[1:3]
+                    db_helper = self.ctx.plugin_manager.context._db
+                    db_helper.update_conversation_title(
+                        user_id=username,
+                        cid=cid,
+                        title=title,
+                    )
+                    web_chat_back_queue.put_nowait(
+                        {
+                            "type": "update_title",
+                            "cid": cid,
+                            "data": title,
+                        }
+                    )
 
     async def _save_to_history(
-        self, event: AstrMessageEvent, req: ProviderRequest, llm_response: LLMResponse
+        self,
+        event: AstrMessageEvent,
+        req: ProviderRequest,
+        llm_response: LLMResponse | None,
     ):
-        if not req or not req.conversation or not llm_response:
+        if (
+            not req
+            or not req.conversation
+            or not llm_response
+            or llm_response.role != "assistant"
+        ):
             return
 
-        if llm_response.role == "assistant":
-            # 文本回复
-            contexts = req.contexts.copy()
-            contexts.append(await req.assemble_context())
+        # 历史上下文
+        messages = copy.deepcopy(req.contexts)
+        # 这一轮对话请求的用户输入
+        messages.append(await req.assemble_context())
+        # 这一轮对话的 LLM 响应
+        if req.tool_calls_result:
+            if not isinstance(req.tool_calls_result, list):
+                messages.extend(req.tool_calls_result.to_openai_messages())
+            elif isinstance(req.tool_calls_result, list):
+                for tcr in req.tool_calls_result:
+                    messages.extend(tcr.to_openai_messages())
+        messages.append({"role": "assistant", "content": llm_response.completion_text})
+        messages = list(filter(lambda item: "_no_save" not in item, messages))
+        await self.conv_manager.update_conversation(
+            event.unified_msg_origin, req.conversation.cid, history=messages
+        )
+        logger.debug(f"messages persisted: {messages}")
 
-            # 记录并标记函数调用结果
-            if req.tool_calls_result:
-                tool_calls_messages = req.tool_calls_result.to_openai_messages()
-
-                # 添加标记
-                for message in tool_calls_messages:
-                    message["_tool_call_history"] = True
-
-                processed_tool_messages = self._process_tool_message_pairs(
-                    tool_calls_messages, remove_tags=False
-                )
-
-                contexts.extend(processed_tool_messages)
-
-            contexts.append(
-                {"role": "assistant", "content": llm_response.completion_text}
-            )
-            contexts_to_save = list(
-                filter(lambda item: "_no_save" not in item, contexts)
-            )
-            await self.conv_manager.update_conversation(
-                event.unified_msg_origin, req.conversation.cid, history=contexts_to_save
-            )
-
-    def _process_tool_message_pairs(self, messages, remove_tags=True):
-        """处理工具调用消息，确保assistant和tool消息成对出现
-
-        Args:
-            messages (list): 消息列表
-            remove_tags (bool): 是否移除_tool_call_history标记
-
-        Returns:
-            list: 处理后的消息列表，保证了assistant和对应tool消息的成对出现
-        """
-        result = []
-        i = 0
-
-        while i < len(messages):
-            current_msg = messages[i]
-
-            # 普通消息直接添加
-            if "_tool_call_history" not in current_msg:
-                result.append(current_msg.copy() if remove_tags else current_msg)
-                i += 1
-                continue
-
-            # 工具调用消息成对处理
-            if current_msg.get("role") == "assistant" and "tool_calls" in current_msg:
-                assistant_msg = current_msg.copy()
-
-                if remove_tags and "_tool_call_history" in assistant_msg:
-                    del assistant_msg["_tool_call_history"]
-
-                related_tools = []
-                j = i + 1
-                while (
-                    j < len(messages)
-                    and messages[j].get("role") == "tool"
-                    and "_tool_call_history" in messages[j]
-                ):
-                    tool_msg = messages[j].copy()
-
-                    if remove_tags:
-                        del tool_msg["_tool_call_history"]
-
-                    related_tools.append(tool_msg)
-                    j += 1
-
-                # 成对的时候添加到结果
-                if related_tools:
-                    result.append(assistant_msg)
-                    result.extend(related_tools)
-
-                i = j  # 跳过已处理
+    def fix_messages(self, messages: list[dict]) -> list[dict]:
+        """验证并且修复上下文"""
+        fixed_messages = []
+        for message in messages:
+            if message.get("role") == "tool":
+                # tool block 前面必须要有 user 和 assistant block
+                if len(fixed_messages) < 2:
+                    # 这种情况可能是上下文被截断导致的
+                    # 我们直接将之前的上下文都清空
+                    fixed_messages = []
+                else:
+                    fixed_messages.append(message)
             else:
-                # 单独的tool消息
-                i += 1
-
-        return result
+                fixed_messages.append(message)
+        return fixed_messages

astrbot/core/pipeline/process_stage/method/star_request.py

@@ -50,7 +50,7 @@ class StarRequestSubStage(Stage):
                 logger.debug(
                     f"plugin -> {star_map.get(handler.handler_module_path).name} - {handler.handler_name}"
                 )
-                wrapper = self._call_handler(self.ctx, event, handler.handler, **params)
+                wrapper = self.ctx.call_handler(event, handler.handler, **params)
                 async for ret in wrapper:
                     yield ret
                 event.clear_result()  # 清除上一个 handler 的结果

astrbot/core/pipeline/respond/stage.py

@@ -29,11 +29,10 @@ class RespondStage(Stage):
         Comp.Image: lambda comp: bool(comp.file),  # 图片
         Comp.Reply: lambda comp: bool(comp.id) and comp.sender_id is not None,  # 回复
         Comp.Poke: lambda comp: comp.id != 0 and comp.qq != 0,  # 戳一戳
-        Comp.Node: lambda comp: bool(comp.name)
-        and comp.uin != 0
-        and bool(comp.content),  # 一个转发节点
+        Comp.Node: lambda comp: bool(comp.content),  # 转发节点
         Comp.Nodes: lambda comp: bool(comp.nodes),  # 多个转发节点
         Comp.File: lambda comp: bool(comp.file_ or comp.url),
+        Comp.WechatEmoji: lambda comp: comp.md5 is not None,  # 微信表情
     }
 
     async def initialize(self, ctx: PipelineContext):
@@ -129,9 +128,7 @@ class RespondStage(Stage):
                 "streaming_segmented", False
             )
             logger.info(f"应用流式输出({event.get_platform_name()})")
-            await event._pre_send()
             await event.send_streaming(result.async_stream, use_fallback)
-            await event._post_send()
             return
         elif len(result.chain) > 0:
             # 检查路径映射
@@ -142,8 +139,6 @@ class RespondStage(Stage):
                         component.file = path_Mapping(mappings, component.file)
                         event.get_result().chain[idx] = component
 
-            await event._pre_send()
-
             # 检查消息链是否为空
             try:
                 if await self._is_empty_message_chain(result.chain):
@@ -159,9 +154,14 @@ class RespondStage(Stage):
                 c for c in result.chain if not isinstance(c, Comp.Record)
             ]
 
-            if self.enable_seg and (
-                (self.only_llm_result and result.is_llm_result())
-                or not self.only_llm_result
+            if (
+                self.enable_seg
+                and (
+                    (self.only_llm_result and result.is_llm_result())
+                    or not self.only_llm_result
+                )
+                and event.get_platform_name()
+                not in ["qq_official", "weixin_official_account", "dingtalk"]
             ):
                 decorated_comps = []
                 if self.reply_with_mention:
@@ -192,6 +192,7 @@ class RespondStage(Stage):
                     await asyncio.sleep(i)
                     try:
                         await event.send(MessageChain([*decorated_comps, comp]))
+                        decorated_comps = []  # 清空已发送的装饰组件
                     except Exception as e:
                         logger.error(f"发送消息失败: {e} chain: {result.chain}")
                         break
@@ -208,7 +209,6 @@ class RespondStage(Stage):
                     logger.error(traceback.format_exc())
                     logger.error(f"发送消息失败: {e} chain: {result.chain}")
 
-            await event._post_send()
             logger.info(
                 f"AstrBot -> {event.get_sender_name()}/{event.get_sender_id()}: {event._outline_chain(result.chain)}"
             )

astrbot/core/pipeline/result_decorate/stage.py

@@ -1,17 +1,18 @@
-import time
 import re
+import time
 import traceback
-from typing import Union, AsyncGenerator
-from ..stage import Stage, register_stage, registered_stages
-from ..context import PipelineContext
-from astrbot.core.platform.astr_message_event import AstrMessageEvent
+from typing import AsyncGenerator, Union
+
+from astrbot.core import html_renderer, logger, file_token_service
+from astrbot.core.message.components import At, File, Image, Node, Plain, Record, Reply
 from astrbot.core.message.message_event_result import ResultContentType
+from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.platform.message_type import MessageType
-from astrbot.core import logger
-from astrbot.core.message.components import Plain, Image, At, Reply, Record, File, Node
-from astrbot.core import html_renderer
-from astrbot.core.star.star_handler import star_handlers_registry, EventType
 from astrbot.core.star.star import star_map
+from astrbot.core.star.star_handler import EventType, star_handlers_registry
+
+from ..context import PipelineContext
+from ..stage import Stage, register_stage, registered_stages
 
 
 @register_stage
@@ -140,7 +141,11 @@ class ResultDecorateStage(Stage):
                         break
 
             # 分段回复
-            if self.enable_segmented_reply:
+            if self.enable_segmented_reply and event.get_platform_name() not in [
+                "qq_official",
+                "weixin_official_account",
+                "dingtalk",
+            ]:
                 if (
                     self.only_llm_result and result.is_llm_result()
                 ) or not self.only_llm_result:
@@ -168,30 +173,55 @@ class ResultDecorateStage(Stage):
                     result.chain = new_chain
 
             # TTS
+            tts_provider = self.ctx.plugin_manager.context.get_using_tts_provider(
+                event.unified_msg_origin
+            )
             if (
                 self.ctx.astrbot_config["provider_tts_settings"]["enable"]
                 and result.is_llm_result()
+                and tts_provider
             ):
-                tts_provider = self.ctx.plugin_manager.context.provider_manager.curr_tts_provider_inst
                 new_chain = []
                 for comp in result.chain:
                     if isinstance(comp, Plain) and len(comp.text) > 1:
                         try:
-                            logger.info("TTS 请求: " + comp.text)
+                            logger.info(f"TTS 请求: {comp.text}")
                             audio_path = await tts_provider.get_audio(comp.text)
-                            logger.info("TTS 结果: " + audio_path)
-                            if audio_path:
-                                new_chain.append(
-                                    Record(file=audio_path, url=audio_path)
-                                )
-                                if(self.ctx.astrbot_config["provider_tts_settings"]["dual_output"]):
-                                    new_chain.append(comp)
-                            else:
+                            logger.info(f"TTS 结果: {audio_path}")
+                            if not audio_path:
                                 logger.error(
-                                    f"由于 TTS 音频文件没找到，消息段转语音失败: {comp.text}"
+                                    f"由于 TTS 音频文件未找到，消息段转语音失败: {comp.text}"
                                 )
                                 new_chain.append(comp)
-                        except BaseException:
+                                continue
+
+                            use_file_service = self.ctx.astrbot_config[
+                                "provider_tts_settings"
+                            ]["use_file_service"]
+                            callback_api_base = self.ctx.astrbot_config[
+                                "callback_api_base"
+                            ]
+                            dual_output = self.ctx.astrbot_config[
+                                "provider_tts_settings"
+                            ]["dual_output"]
+
+                            url = None
+                            if use_file_service and callback_api_base:
+                                token = await file_token_service.register_file(
+                                    audio_path
+                                )
+                                url = f"{callback_api_base}/api/file/{token}"
+                                logger.debug(f"已注册：{url}")
+
+                            new_chain.append(
+                                Record(
+                                    file=url or audio_path,
+                                    url=url or audio_path,
+                                )
+                            )
+                            if dual_output:
+                                new_chain.append(comp)
+                        except Exception:
                             logger.error(traceback.format_exc())
                             logger.error("TTS 失败，使用文本发送。")
                             new_chain.append(comp)
@@ -225,6 +255,14 @@ class ResultDecorateStage(Stage):
                     if url:
                         if url.startswith("http"):
                             result.chain = [Image.fromURL(url)]
+                        elif (
+                            self.ctx.astrbot_config["t2i_use_file_service"]
+                            and self.ctx.astrbot_config["callback_api_base"]
+                        ):
+                            token = await file_token_service.register_file(url)
+                            url = f"{self.ctx.astrbot_config['callback_api_base']}/api/file/{token}"
+                            logger.debug(f"已注册：{url}")
+                            result.chain = [Image.fromURL(url)]
                         else:
                             result.chain = [Image.fromFileSystem(url)]
 

astrbot/core/pipeline/stage.py

@@ -1,12 +1,8 @@
 from __future__ import annotations
 import abc
-import inspect
-import traceback
-from astrbot.api import logger
-from typing import List, AsyncGenerator, Union, Awaitable
+from typing import List, AsyncGenerator, Union
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from .context import PipelineContext
-from astrbot.core.message.message_event_result import MessageEventResult, CommandResult
 
 registered_stages: List[Stage] = []  # 维护了所有已注册的 Stage 实现类
 
@@ -41,70 +37,3 @@ class Stage(abc.ABC):
             Union[None, AsyncGenerator[None, None]]: 处理结果，可能是 None 或者异步生成器, 如果为 None 则表示不需要继续处理, 如果为异步生成器则表示需要继续处理(进入下一个阶段)
         """
         raise NotImplementedError
-
-    async def _call_handler(
-        self,
-        ctx: PipelineContext,
-        event: AstrMessageEvent,
-        handler: Awaitable,
-        *args,
-        **kwargs,
-    ) -> AsyncGenerator[None, None]:
-        """执行事件处理函数并处理其返回结果
-
-        该方法负责调用处理函数并处理不同类型的返回值。它支持两种类型的处理函数:
-        1. 异步生成器: 实现洋葱模型，每次yield都会将控制权交回上层
-        2. 协程: 执行一次并处理返回值
-
-        Args:
-            ctx (PipelineContext): 消息管道上下文对象
-            event (AstrMessageEvent): 待处理的事件对象
-            handler (Awaitable): 事件处理函数
-            *args: 传递给handler的位置参数
-            **kwargs: 传递给handler的关键字参数
-
-        Returns:
-            AsyncGenerator[None, None]: 异步生成器，用于在管道中传递控制流
-        """
-        ready_to_call = None  # 一个协程或者异步生成器(async def)
-
-        trace_ = None
-
-        try:
-            ready_to_call = handler(event, *args, **kwargs)
-        except TypeError as _:
-            # 向下兼容
-            trace_ = traceback.format_exc()
-            # 以前的handler会额外传入一个参数, 但是context对象实际上在插件实例中有一份
-            ready_to_call = handler(event, ctx.plugin_manager.context, *args, **kwargs)
-
-        if isinstance(ready_to_call, AsyncGenerator):
-            # 如果是一个异步生成器, 进入洋葱模型
-            _has_yielded = False  # 是否返回过值
-            try:
-                async for ret in ready_to_call:
-                    # 这里逐步执行异步生成器, 对于每个yield返回的ret, 执行下面的代码
-                    # 返回值只能是 MessageEventResult 或者 None（无返回值）
-                    _has_yielded = True
-                    if isinstance(ret, (MessageEventResult, CommandResult)):
-                        # 如果返回值是 MessageEventResult, 设置结果并继续
-                        event.set_result(ret)
-                        yield  # 传递控制权给上一层的process函数
-                    else:
-                        # 如果返回值是 None, 则不设置结果并继续
-                        # 继续执行后续阶段
-                        yield ret  # 传递控制权给上一层的process函数
-                if not _has_yielded:
-                    # 如果这个异步生成器没有执行到yield分支
-                    yield
-            except Exception as e:
-                logger.error(f"Previous Error: {trace_}")
-                raise e
-        elif inspect.iscoroutine(ready_to_call):
-            # 如果只是一个协程, 直接执行
-            ret = await ready_to_call
-            if isinstance(ret, (MessageEventResult, CommandResult)):
-                event.set_result(ret)
-                yield  # 传递控制权给上一层的process函数
-            else:
-                yield ret  # 传递控制权给上一层的process函数

astrbot/core/pipeline/waking_check/stage.py

@@ -4,7 +4,7 @@ from astrbot import logger
 from typing import Union, AsyncGenerator
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.message.message_event_result import MessageEventResult, MessageChain
-from astrbot.core.message.components import At
+from astrbot.core.message.components import At, AtAll, Reply
 from astrbot.core.star.star_handler import star_handlers_registry, EventType
 from astrbot.core.star.star import star_map
 from astrbot.core.star.filter.permission import PermissionTypeFilter
@@ -39,6 +39,9 @@ class WakingCheckStage(Stage):
         self.ignore_bot_self_message = self.ctx.astrbot_config["platform_settings"].get(
             "ignore_bot_self_message", False
         )
+        self.ignore_at_all = self.ctx.astrbot_config["platform_settings"].get(
+            "ignore_at_all", False
+        )
 
     async def process(
         self, event: AstrMessageEvent
@@ -77,11 +80,18 @@ class WakingCheckStage(Stage):
                 event.message_str = event.message_str[len(wake_prefix) :].strip()
                 break
         if not is_wake:
-            # 检查是否有 at 消息
+            # 检查是否有at消息 / at全体成员消息 / 引用了bot的消息
             for message in messages:
-                if isinstance(message, At) and (
-                    str(message.qq) == str(event.get_self_id())
-                    or str(message.qq) == "all"
+                if (
+                    (
+                        isinstance(message, At)
+                        and (str(message.qq) == str(event.get_self_id()))
+                    )
+                    or (isinstance(message, AtAll) and not self.ignore_at_all)
+                    or (
+                        isinstance(message, Reply)
+                        and str(message.sender_id) == str(event.get_self_id())
+                    )
                 ):
                     is_wake = True
                     event.is_wake = True
@@ -125,7 +135,6 @@ class WakingCheckStage(Stage):
                             f"插件 {star_map[handler.handler_module_path].name}: {e}"
                         )
                     )
-                    await event._post_send()
                     event.stop_event()
                     passed = False
                     break
@@ -140,7 +149,6 @@ class WakingCheckStage(Stage):
                                 f"您(ID: {event.get_sender_id()})的权限不足以使用此指令。通过 /sid 获取 ID 并请管理员添加。"
                             )
                         )
-                        await event._post_send()
                     logger.info(
                         f"触发 {star_map[handler.handler_module_path].name} 时, 用户(ID={event.get_sender_id()}) 权限不足。"
                     )

astrbot/core/platform/astr_message_event.py

@@ -235,10 +235,10 @@ class AstrMessageEvent(abc.ABC):
         self._has_send_oper = True
 
     async def _pre_send(self):
-        """调度器会在执行 send() 前调用该方法"""
+        """调度器会在执行 send() 前调用该方法 deprecated in v3.5.18"""
 
     async def _post_send(self):
-        """调度器会在执行 send() 后调用该方法"""
+        """调度器会在执行 send() 后调用该方法 deprecated in v3.5.18"""
 
     def set_result(self, result: Union[MessageEventResult, str]):
         """设置消息事件的结果。

astrbot/core/platform/manager.py

@@ -77,7 +77,15 @@ class PlatformManager:
                 case "wecom":
                     from .sources.wecom.wecom_adapter import WecomPlatformAdapter  # noqa: F401
                 case "weixin_official_account":
-                    from .sources.weixin_official_account.weixin_offacc_adapter import WeixinOfficialAccountPlatformAdapter # noqa
+                    from .sources.weixin_official_account.weixin_offacc_adapter import (
+                        WeixinOfficialAccountPlatformAdapter,  # noqa
+                    )
+                case "discord":
+                    from .sources.discord.discord_platform_adapter import (
+                        DiscordPlatformAdapter,  # noqa: F401
+                    )
+                case "slack":
+                    from .sources.slack.slack_adapter import SlackAdapter  # 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->控制台->安装Pip库 中安装依赖库。"

astrbot/core/platform/sources/aiocqhttp/aiocqhttp_message_event.py

@@ -3,9 +3,17 @@ import re
 from typing import AsyncGenerator, Dict, List
 from aiocqhttp import CQHttp
 from astrbot.api.event import AstrMessageEvent, MessageChain
-from astrbot.api.message_components import At, Image, Node, Nodes, Plain, Record, File
+from astrbot.api.message_components import (
+    Image,
+    Node,
+    Nodes,
+    Plain,
+    Record,
+    Video,
+    File,
+    BaseMessageComponent,
+)
 from astrbot.api.platform import Group, MessageMember
-from astrbot.core import file_token_service, astrbot_config, logger
 
 
 class AiocqhttpMessageEvent(AstrMessageEvent):
@@ -15,28 +23,38 @@ class AiocqhttpMessageEvent(AstrMessageEvent):
         super().__init__(message_str, message_obj, platform_meta, session_id)
         self.bot = bot
 
+    @staticmethod
+    async def _from_segment_to_dict(segment: BaseMessageComponent) -> dict:
+        """修复部分字段"""
+        if isinstance(segment, (Image, Record)):
+            # For Image and Record segments, we convert them to base64
+            bs64 = await segment.convert_to_base64()
+            return {
+                "type": segment.type.lower(),
+                "data": {
+                    "file": f"base64://{bs64}",
+                },
+            }
+        elif isinstance(segment, File):
+            # For File segments, we need to handle the file differently
+            d = await segment.to_dict()
+            return d
+        elif isinstance(segment, Video):
+            d = await segment.to_dict()
+            return d
+        else:
+            # For other segments, we simply convert them to a dict by calling toDict
+            return segment.toDict()
+
     @staticmethod
     async def _parse_onebot_json(message_chain: MessageChain):
         """解析成 OneBot json 格式"""
         ret = []
         for segment in message_chain.chain:
-            d = segment.toDict()
             if isinstance(segment, Plain):
-                d["type"] = "text"
-                d["data"]["text"] = segment.text.strip()
-                # 如果是空文本或者只带换行符的文本，不发送
-                if not d["data"]["text"]:
+                if not segment.text.strip():
                     continue
-            elif isinstance(segment, (Image, Record)):
-                # convert to base64
-                bs64 = await segment.convert_to_base64()
-                d["data"] = {
-                    "file": f"base64://{bs64}",
-                }
-            elif isinstance(segment, At):
-                d["data"] = {
-                    "qq": str(segment.qq),  # 转换为字符串
-                }
+            d = await AiocqhttpMessageEvent._from_segment_to_dict(segment)
             ret.append(d)
         return ret
 
@@ -54,7 +72,8 @@ class AiocqhttpMessageEvent(AstrMessageEvent):
                         nodes = Nodes([seg])
                         seg = nodes
 
-                    payload = seg.toDict()
+                    payload = await seg.to_dict()
+
                     if self.get_group_id():
                         payload["group_id"] = self.get_group_id()
                         await self.bot.call_action("send_group_forward_msg", **payload)
@@ -64,21 +83,7 @@ class AiocqhttpMessageEvent(AstrMessageEvent):
                             "send_private_forward_msg", **payload
                         )
                 elif isinstance(seg, File):
-                    d = seg.toDict()
-                    url_or_path = await seg.get_file(allow_return_url=True)
-                    if url_or_path.startswith("http"):
-                        payload_file = url_or_path
-                    elif callback_host := astrbot_config.get("callback_api_base"):
-                        callback_host = str(callback_host).removesuffix("/")
-                        token = await file_token_service.register_file(url_or_path)
-                        payload_file = f"{callback_host}/api/file/{token}"
-                        logger.debug(f"Generated file callback link: {payload_file}")
-                    else:
-                        payload_file = url_or_path
-                    d["data"] = {
-                        "name": seg.name,
-                        "file": payload_file,
-                    }
+                    d = await AiocqhttpMessageEvent._from_segment_to_dict(seg)
                     await self.bot.send(
                         self.message_obj.raw_message,
                         [d],

astrbot/core/platform/sources/aiocqhttp/aiocqhttp_platform_adapter.py

@@ -168,9 +168,7 @@ class AiocqhttpAdapter(Platform):
 
         if "sub_type" in event:
             if event["sub_type"] == "poke" and "target_id" in event:
-                abm.message.append(
-                    Poke(qq=str(event["target_id"]), type="poke")
-                )  # noqa: F405
+                abm.message.append(Poke(qq=str(event["target_id"]), type="poke"))  # noqa: F405
 
         return abm
 
@@ -221,6 +219,9 @@ class AiocqhttpAdapter(Platform):
             a = None
             if t == "text":
                 current_text = "".join(m["data"]["text"] for m in m_group).strip()
+                if not current_text:
+                    # 如果文本段为空，则跳过
+                    continue
                 message_str += current_text
                 a = ComponentTypes[t](text=current_text)  # noqa: F405
                 abm.message.append(a)
@@ -270,6 +271,8 @@ class AiocqhttpAdapter(Platform):
                                 action="get_msg",
                                 message_id=int(m["data"]["id"]),
                             )
+                            # 添加必要的 post_type 字段，防止 Event.from_payload 报错
+                            reply_event_data["post_type"] = "message"
                             abm_reply = await self._convert_handle_message_event(
                                 Event.from_payload(reply_event_data), get_reply=False
                             )
@@ -304,7 +307,7 @@ class AiocqhttpAdapter(Platform):
                             user_id=int(m["data"]["qq"]),
                         )
                         if at_info:
-                            nickname = at_info.get("nick", "")
+                            nickname = at_info.get("nick", "") or at_info.get("nickname", "")
                             is_at_self = str(m["data"]["qq"]) in {abm.self_id, "all"}
 
                             abm.message.append(
@@ -319,7 +322,7 @@ class AiocqhttpAdapter(Platform):
                                 first_at_self_processed = True
                             else:
                                 # 非第一个@机器人或@其他用户，添加到message_str
-                                message_str += f" @{nickname} "
+                                message_str += f" @{nickname}({m['data']['qq']}) "
                         else:
                             abm.message.append(At(qq=str(m["data"]["qq"]), name=""))
                     except ActionFailed as e:

astrbot/core/platform/sources/dingtalk/dingtalk_event.py

@@ -32,31 +32,31 @@ class DingtalkMessageEvent(AstrMessageEvent):
                 )
             elif isinstance(segment, Comp.Image):
                 markdown_str = ""
-                if segment.file and segment.file.startswith("file:///"):
-                    logger.warning(
-                        "dingtalk only support url image, not: " + segment.file
-                    )
-                    continue
-                elif segment.file and segment.file.startswith("http"):
-                    markdown_str += f"![image]({segment.file})\n\n"
-                elif segment.file and segment.file.startswith("base64://"):
-                    logger.warning("dingtalk only support url image, not base64")
-                    continue
-                else:
-                    logger.warning(
-                        "dingtalk only support url image, not: " + segment.file
-                    )
-                    continue
 
-                ret = await asyncio.get_event_loop().run_in_executor(
-                    None,
-                    client.reply_markdown,
-                    "😄",
-                    markdown_str,
-                    self.message_obj.raw_message,
-                )
-                logger.debug(f"send image: {ret}")
+                try:
+                    if not segment.file:
+                        logger.warning("钉钉图片 segment 缺少 file 字段，跳过")
+                        continue
+                    if segment.file.startswith(("http://", "https://")):
+                        image_url = segment.file
+                    else:
+                        image_url = await segment.register_to_file_service()
 
+                    markdown_str = f"![image]({image_url})\n\n"
+
+                    ret = await asyncio.get_event_loop().run_in_executor(
+                        None,
+                        client.reply_markdown,
+                        "😄",
+                        markdown_str,
+                        self.message_obj.raw_message,
+                    )
+                    logger.debug(f"send image: {ret}")
+
+                except Exception as e:
+                    logger.error(f"钉钉图片处理失败: {e}")
+                    logger.warning(f"跳过图片发送: {image_path}")
+                    continue
     async def send(self, message: MessageChain):
         await self.send_with_client(self.client, message)
         await super().send(message)

astrbot/core/platform/sources/discord/client.py

@@ -0,0 +1,126 @@
+import discord
+from astrbot import logger
+import sys
+
+if sys.version_info >= (3, 12):
+    from typing import override
+else:
+    from typing_extensions import override
+
+
+# Discord Bot客户端
+class DiscordBotClient(discord.Bot):
+    """Discord客户端封装"""
+
+    def __init__(self, token: str, proxy: str = None):
+        self.token = token
+        self.proxy = proxy
+
+        # 设置Intent权限，遵循权限最小化原则
+        intents = discord.Intents.default()
+        intents.message_content = True  # 订阅消息内容事件 (Privileged)
+        intents.members = True  # 订阅成员事件 (Privileged)
+
+        # 初始化Bot
+        super().__init__(intents=intents, proxy=proxy)
+
+        # 回调函数
+        self.on_message_received = None
+        self.on_ready_once_callback = None
+        self._ready_once_fired = False
+
+    @override
+    async def on_ready(self):
+        """当机器人成功连接并准备就绪时触发"""
+        logger.info(f"[Discord] 已作为 {self.user} (ID: {self.user.id}) 登录")
+        logger.info("[Discord] 客户端已准备就绪。")
+
+        if self.on_ready_once_callback and not self._ready_once_fired:
+            self._ready_once_fired = True
+            try:
+                await self.on_ready_once_callback()
+            except Exception as e:
+                logger.error(
+                    f"[Discord] on_ready_once_callback 执行失败: {e}", exc_info=True)
+
+    def _create_message_data(self, message: discord.Message) -> dict:
+        """从 discord.Message 创建数据字典"""
+        is_mentioned = self.user in message.mentions
+        return {
+            "message": message,
+            "bot_id": str(self.user.id),
+            "content": message.content,
+            "username": message.author.display_name,
+            "userid": str(message.author.id),
+            "message_id": str(message.id),
+            "channel_id": str(message.channel.id),
+            "guild_id": str(message.guild.id) if message.guild else None,
+            "type": "message",
+            "is_mentioned": is_mentioned,
+            "clean_content": message.clean_content,
+        }
+
+    def _create_interaction_data(self, interaction: discord.Interaction) -> dict:
+        """从 discord.Interaction 创建数据字典"""
+        return {
+            "interaction": interaction,
+            "bot_id": str(self.user.id),
+            "content": self._extract_interaction_content(interaction),
+            "username": interaction.user.display_name,
+            "userid": str(interaction.user.id),
+            "message_id": str(interaction.id),
+            "channel_id": str(interaction.channel_id)
+            if interaction.channel_id
+            else None,
+            "guild_id": str(interaction.guild_id) if interaction.guild_id else None,
+            "type": "interaction",
+        }
+
+    @override
+    async def on_message(self, message: discord.Message):
+        """当接收到消息时触发"""
+        if message.author.bot:
+            return
+
+        logger.debug(
+            f"[Discord] 收到原始消息 from {message.author.name}: {message.content}"
+        )
+
+        if self.on_message_received:
+            message_data = self._create_message_data(message)
+            await self.on_message_received(message_data)
+
+
+    def _extract_interaction_content(self, interaction: discord.Interaction) -> str:
+        """从交互中提取内容"""
+        interaction_type = interaction.type
+        interaction_data = getattr(interaction, "data", {})
+
+        if not interaction_data:
+            return ""
+
+        if interaction_type == discord.InteractionType.application_command:
+            command_name = interaction_data.get("name", "")
+            if options := interaction_data.get("options", []):
+                params = " ".join(
+                    [f"{opt['name']}:{opt.get('value', '')}" for opt in options]
+                )
+                return f"/{command_name} {params}"
+            return f"/{command_name}"
+
+        elif interaction_type == discord.InteractionType.component:
+            custom_id = interaction_data.get("custom_id", "")
+            component_type = interaction_data.get("component_type", "")
+            return f"component:{custom_id}:{component_type}"
+
+        return str(interaction_data)
+
+    async def start_polling(self):
+        """开始轮询消息，这是个阻塞方法"""
+        await self.start(self.token)
+
+    @override
+    async def close(self):
+        """关闭客户端"""
+        if not self.is_closed():
+            await super().close()

astrbot/core/platform/sources/discord/components.py

@@ -0,0 +1,133 @@
+import discord
+from typing import List
+from astrbot.api.message_components import BaseMessageComponent
+
+
+# Discord专用组件
+class DiscordEmbed(BaseMessageComponent):
+    """Discord Embed消息组件"""
+
+    type: str = "discord_embed"
+
+    def __init__(
+        self,
+        title: str = None,
+        description: str = None,
+        color: int = None,
+        url: str = None,
+        thumbnail: str = None,
+        image: str = None,
+        footer: str = None,
+        fields: List[dict] = None,
+    ):
+        self.title = title
+        self.description = description
+        self.color = color
+        self.url = url
+        self.thumbnail = thumbnail
+        self.image = image
+        self.footer = footer
+        self.fields = fields or []
+
+    def to_discord_embed(self) -> discord.Embed:
+        """转换为Discord Embed对象"""
+        embed = discord.Embed()
+
+        if self.title:
+            embed.title = self.title
+        if self.description:
+            embed.description = self.description
+        if self.color:
+            embed.color = self.color
+        if self.url:
+            embed.url = self.url
+        if self.thumbnail:
+            embed.set_thumbnail(url=self.thumbnail)
+        if self.image:
+            embed.set_image(url=self.image)
+        if self.footer:
+            embed.set_footer(text=self.footer)
+
+        for field in self.fields:
+            embed.add_field(
+                name=field.get("name", ""),
+                value=field.get("value", ""),
+                inline=field.get("inline", False),
+            )
+
+        return embed
+
+
+class DiscordButton(BaseMessageComponent):
+    """Discord按钮组件"""
+
+    type: str = "discord_button"
+
+    def __init__(
+        self,
+        label: str,
+        custom_id: str = None,
+        style: str = "primary",
+        emoji: str = None,
+        url: str = None,
+        disabled: bool = False,
+    ):
+        self.label = label
+        self.custom_id = custom_id
+        self.style = style
+        self.emoji = emoji
+        self.url = url
+        self.disabled = disabled
+
+class DiscordReference(BaseMessageComponent):
+    """Discord引用组件"""
+    type: str = "discord_reference"
+    def __init__(self, message_id: str, channel_id: str):
+        self.message_id = message_id
+        self.channel_id = channel_id
+
+
+class DiscordView(BaseMessageComponent):
+    """Discord视图组件，包含按钮和选择菜单"""
+
+    type: str = "discord_view"
+
+    def __init__(
+        self, components: List[BaseMessageComponent] = None, timeout: float = None
+    ):
+        self.components = components or []
+        self.timeout = timeout
+
+
+    def to_discord_view(self) -> discord.ui.View:
+        """转换为Discord View对象"""
+        view = discord.ui.View(timeout=self.timeout)
+
+        for component in self.components:
+            if isinstance(component, DiscordButton):
+                button_style = getattr(
+                    discord.ButtonStyle, component.style, discord.ButtonStyle.primary
+                )
+
+                if component.url:
+                    # URL按钮
+                    button = discord.ui.Button(
+                        label=component.label,
+                        style=discord.ButtonStyle.link,
+                        url=component.url,
+                        emoji=component.emoji,
+                        disabled=component.disabled,
+                    )
+                else:
+                    # 普通按钮
+                    button = discord.ui.Button(
+                        label=component.label,
+                        style=button_style,
+                        custom_id=component.custom_id,
+                        emoji=component.emoji,
+                        disabled=component.disabled,
+                    )
+
+                view.add_item(button)
+
+        return view

astrbot/core/platform/sources/discord/discord_platform_adapter.py

@@ -0,0 +1,455 @@
+import asyncio
+import discord
+import sys
+import re
+from discord.abc import Messageable
+from discord.channel import DMChannel
+from astrbot.api.platform import (
+    Platform,
+    AstrBotMessage,
+    MessageMember,
+    PlatformMetadata,
+    MessageType,
+)
+from astrbot.api.event import MessageChain
+from astrbot.api.message_components import Plain, Image, File
+from astrbot.core.platform.astr_message_event import MessageSesion
+from astrbot.api.platform import register_platform_adapter
+from astrbot import logger
+from .client import DiscordBotClient
+from .discord_platform_event import DiscordPlatformEvent
+
+from typing import Any, Tuple
+from astrbot.core.star.filter.command import CommandFilter
+from astrbot.core.star.filter.command_group import CommandGroupFilter
+from astrbot.core.star.star import star_map
+from astrbot.core.star.star_handler import StarHandlerMetadata, star_handlers_registry
+
+if sys.version_info >= (3, 12):
+    from typing import override
+else:
+    from typing_extensions import override
+
+
+# 注册平台适配器
+@register_platform_adapter("discord", "Discord 适配器 (基于 Pycord)")
+class DiscordPlatformAdapter(Platform):
+    def __init__(
+        self, platform_config: dict, platform_settings: dict, event_queue: asyncio.Queue
+    ) -> None:
+        super().__init__(event_queue)
+        self.config = platform_config
+        self.settings = platform_settings
+        self.client_self_id = None
+        self.registered_handlers = []
+        # 指令注册相关
+        self.enable_command_register = self.config.get("discord_command_register", True)
+        self.guild_id = self.config.get("discord_guild_id_for_debug", None)
+        self.activity_name = self.config.get("discord_activity_name", None)
+        self.shutdown_event = asyncio.Event()
+        self._polling_task = None
+
+    @override
+    async def send_by_session(
+        self, session: MessageSesion, message_chain: MessageChain
+    ):
+        """通过会话发送消息"""
+        # 创建一个 message_obj 以便在 event 中使用
+        message_obj = AstrBotMessage()
+        if "_" in session.session_id:
+            session.session_id = session.session_id.split("_")[1]
+        channel_id_str = session.session_id
+        channel = None
+        try:
+            channel_id = int(channel_id_str)
+            channel = self.client.get_channel(channel_id)
+        except (ValueError, TypeError):
+            logger.warning(f"[Discord] Invalid channel ID format: {channel_id_str}")
+
+        if channel:
+            message_obj.type = self._get_message_type(channel)
+            message_obj.group_id = self._get_channel_id(channel)
+        else:
+            logger.warning(
+                f"[Discord] Can't get channel info for {channel_id_str}, will guess message type."
+            )
+            message_obj.type = MessageType.GROUP_MESSAGE
+            message_obj.group_id = session.session_id
+
+        message_obj.message_str = message_chain.get_plain_text()
+        message_obj.sender = MessageMember(
+            user_id=str(self.client_self_id), nickname=self.client.user.display_name
+        )
+        message_obj.self_id = self.client_self_id
+        message_obj.session_id = session.session_id
+        message_obj.message = message_chain
+
+        # 创建临时事件对象来发送消息
+        temp_event = DiscordPlatformEvent(
+            message_str=message_chain.get_plain_text(),
+            message_obj=message_obj,
+            platform_meta=self.meta(),
+            session_id=session.session_id,
+            client=self.client,
+        )
+        await temp_event.send(message_chain)
+        await super().send_by_session(session, message_chain)
+
+    @override
+    def meta(self) -> PlatformMetadata:
+        """返回平台元数据"""
+        return PlatformMetadata(
+            "discord",
+            "Discord 适配器",
+            id=self.config.get("id"),
+            default_config_tmpl=self.config,
+        )
+
+    @override
+    async def run(self):
+        """主要运行逻辑"""
+
+        # 初始化回调函数
+        async def on_received(message_data):
+            logger.debug(f"[Discord] 收到消息: {message_data}")
+            if self.client_self_id is None:
+                self.client_self_id = message_data.get("bot_id")
+            abm = await self.convert_message(data=message_data)
+            await self.handle_msg(abm)
+
+        # 初始化 Discord 客户端
+        token = str(self.config.get("discord_token"))
+        if not token:
+            logger.error("[Discord] Bot Token 未配置。请在配置文件中正确设置 token。")
+            return
+
+        proxy = self.config.get("discord_proxy") or None
+        self.client = DiscordBotClient(token, proxy)
+        self.client.on_message_received = on_received
+
+        async def callback():
+            if self.enable_command_register:
+                await self._collect_and_register_commands()
+            if self.activity_name:
+                await self.client.change_presence(
+                    status=discord.Status.online,
+                    activity=discord.CustomActivity(name=self.activity_name),
+                )
+
+        self.client.on_ready_once_callback = callback
+
+        try:
+            self._polling_task = asyncio.create_task(self.client.start_polling())
+            await self.shutdown_event.wait()
+        except discord.errors.LoginFailure:
+            logger.error("[Discord] 登录失败。请检查你的 Bot Token 是否正确。")
+        except discord.errors.ConnectionClosed:
+            logger.warning("[Discord] 与 Discord 的连接已关闭。")
+        except Exception as e:
+            logger.error(f"[Discord] 适配器运行时发生意外错误: {e}", exc_info=True)
+
+    def _get_message_type(
+        self, channel: Messageable, guild_id: int | None = None
+    ) -> MessageType:
+        """根据 channel 对象和 guild_id 判断消息类型"""
+        if guild_id is not None:
+            return MessageType.GROUP_MESSAGE
+        if isinstance(channel, DMChannel) or getattr(channel, "guild", None) is None:
+            return MessageType.FRIEND_MESSAGE
+        return MessageType.GROUP_MESSAGE
+
+    def _get_channel_id(self, channel: Messageable) -> str:
+        """根据 channel 对象获取ID"""
+        return str(getattr(channel, "id", None))
+
+    def _convert_message_to_abm(self, data: dict) -> AstrBotMessage:
+        """将普通消息转换为 AstrBotMessage"""
+        message: discord.Message = data["message"]
+
+        content = message.content
+
+        # 如果机器人被@，移除@部分
+        # 剥离 User Mention (<@id>, <@!id>)
+        if self.client and self.client.user:
+            mention_str = f"<@{self.client.user.id}>"
+            mention_str_nickname = f"<@!{self.client.user.id}>"
+            if content.startswith(mention_str):
+                content = content[len(mention_str) :].lstrip()
+            elif content.startswith(mention_str_nickname):
+                content = content[len(mention_str_nickname) :].lstrip()
+
+        # 剥离 Role Mention（bot 拥有的任一角色被提及，<@&role_id>）
+        if (
+            hasattr(message, "role_mentions")
+            and hasattr(message, "guild")
+            and message.guild
+        ):
+            bot_member = (
+                message.guild.get_member(self.client.user.id)
+                if self.client and self.client.user
+                else None
+            )
+            if bot_member and hasattr(bot_member, "roles"):
+                for role in bot_member.roles:
+                    role_mention_str = f"<@&{role.id}>"
+                    if content.startswith(role_mention_str):
+                        content = content[len(role_mention_str) :].lstrip()
+                        break  # 只剥离第一个匹配的角色 mention
+
+        abm = AstrBotMessage()
+        abm.type = self._get_message_type(message.channel)
+        abm.group_id = self._get_channel_id(message.channel)
+        abm.message_str = content
+        abm.sender = MessageMember(
+            user_id=str(message.author.id), nickname=message.author.display_name
+        )
+        message_chain = []
+        if abm.message_str:
+            message_chain.append(Plain(text=abm.message_str))
+        if message.attachments:
+            for attachment in message.attachments:
+                if attachment.content_type and attachment.content_type.startswith(
+                    "image/"
+                ):
+                    message_chain.append(
+                        Image(file=attachment.url, filename=attachment.filename)
+                    )
+                else:
+                    message_chain.append(
+                        File(name=attachment.filename, url=attachment.url)
+                    )
+        abm.message = message_chain
+        abm.raw_message = message
+        abm.self_id = self.client_self_id
+        abm.session_id = str(message.channel.id)
+        abm.message_id = str(message.id)
+        return abm
+
+    async def convert_message(self, data: dict) -> AstrBotMessage:
+        """将平台消息转换成 AstrBotMessage"""
+        # 由于 on_interaction 已被禁用，我们只处理普通消息
+        return self._convert_message_to_abm(data)
+
+    async def handle_msg(self, message: AstrBotMessage, followup_webhook=None):
+        """处理消息"""
+        message_event = DiscordPlatformEvent(
+            message_str=message.message_str,
+            message_obj=message,
+            platform_meta=self.meta(),
+            session_id=message.session_id,
+            client=self.client,
+            interaction_followup_webhook=followup_webhook,
+        )
+
+        # 检查是否为斜杠指令
+        is_slash_command = message_event.interaction_followup_webhook is not None
+
+        # 检查是否被@（User Mention 或 Bot 拥有的 Role Mention）
+        is_mention = False
+        # User Mention
+        if (
+            self.client
+            and self.client.user
+            and hasattr(message.raw_message, "mentions")
+        ):
+            if self.client.user in message.raw_message.mentions:
+                is_mention = True
+        # Role Mention（Bot 拥有的角色被提及）
+        if not is_mention and hasattr(message.raw_message, "role_mentions"):
+            bot_member = None
+            if hasattr(message.raw_message, "guild") and message.raw_message.guild:
+                try:
+                    bot_member = message.raw_message.guild.get_member(
+                        self.client.user.id
+                    )
+                except Exception:
+                    bot_member = None
+            if bot_member and hasattr(bot_member, "roles"):
+                bot_roles = set(bot_member.roles)
+                mentioned_roles = set(message.raw_message.role_mentions)
+                if (
+                    bot_roles
+                    and mentioned_roles
+                    and bot_roles.intersection(mentioned_roles)
+                ):
+                    is_mention = True
+
+        # 如果是斜杠指令或被@的消息，设置为唤醒状态
+        if is_slash_command or is_mention:
+            message_event.is_wake = True
+            message_event.is_at_or_wake_command = True
+
+        self.commit_event(message_event)
+
+    @override
+    async def terminate(self):
+        """终止适配器"""
+        logger.info("[Discord] 正在终止适配器... (step 1: cancel polling task)")
+        self.shutdown_event.set()
+        # 优先 cancel polling_task
+        if self._polling_task:
+            self._polling_task.cancel()
+            try:
+                await asyncio.wait_for(self._polling_task, timeout=10)
+            except asyncio.CancelledError:
+                logger.info("[Discord] polling_task 已取消。")
+            except Exception as e:
+                logger.warning(f"[Discord] polling_task 取消异常: {e}")
+        logger.info("[Discord] 正在清理已注册的斜杠指令... (step 2)")
+        # 清理指令
+        if self.enable_command_register and self.client:
+            try:
+                await asyncio.wait_for(
+                    self.client.sync_commands(
+                        commands=[],
+                        guild_ids=[self.guild_id] if self.guild_id else None,
+                    ),
+                    timeout=10,
+                )
+                logger.info("[Discord] 指令清理完成。")
+            except Exception as e:
+                logger.error(f"[Discord] 清理指令时发生错误: {e}", exc_info=True)
+        logger.info("[Discord] 正在关闭 Discord 客户端... (step 3)")
+        if self.client and hasattr(self.client, "close"):
+            try:
+                await asyncio.wait_for(self.client.close(), timeout=10)
+            except Exception as e:
+                logger.warning(f"[Discord] 客户端关闭异常: {e}")
+        logger.info("[Discord] 适配器已终止。")
+
+    def register_handler(self, handler_info):
+        """注册处理器信息"""
+        self.registered_handlers.append(handler_info)
+
+    async def _collect_and_register_commands(self):
+        """收集所有指令并注册到Discord"""
+        logger.info("[Discord] 开始收集并注册斜杠指令...")
+        registered_commands = []
+
+        for handler_md in star_handlers_registry:
+            if not star_map[handler_md.handler_module_path].activated:
+                continue
+            for event_filter in handler_md.event_filters:
+                cmd_info = self._extract_command_info(event_filter, handler_md)
+                if not cmd_info:
+                    continue
+
+                cmd_name, description, cmd_filter_instance = cmd_info
+
+                # 创建动态回调
+                callback = self._create_dynamic_callback(cmd_name)
+
+                # 创建一个通用的参数选项来接收所有文本输入
+                options = [
+                    discord.Option(
+                        name="params",
+                        description="指令的所有参数",
+                        type=discord.SlashCommandOptionType.string,
+                        required=False,
+                    )
+                ]
+
+                # 创建SlashCommand
+                slash_command = discord.SlashCommand(
+                    name=cmd_name,
+                    description=description,
+                    func=callback,
+                    options=options,
+                    guild_ids=[self.guild_id] if self.guild_id else None,
+                )
+                self.client.add_application_command(slash_command)
+                registered_commands.append(cmd_name)
+
+        if registered_commands:
+            logger.info(
+                f"[Discord] 准备同步 {len(registered_commands)} 个指令: {', '.join(registered_commands)}"
+            )
+        else:
+            logger.info("[Discord] 没有发现可注册的指令。")
+
+        # 使用 Pycord 的方法同步指令
+        # 注意：这可能需要一些时间，并且有频率限制
+        await self.client.sync_commands()
+        logger.info("[Discord] 指令同步完成。")
+
+    def _create_dynamic_callback(self, cmd_name: str):
+        """为每个指令动态创建一个异步回调函数"""
+
+        async def dynamic_callback(ctx: discord.ApplicationContext, params: str = None):
+            # 将平台特定的前缀'/'剥离，以适配通用的CommandFilter
+            logger.debug(f"[Discord] 回调函数触发: {cmd_name}")
+            logger.debug(f"[Discord] 回调函数参数: {ctx}")
+            logger.debug(f"[Discord] 回调函数参数: {params}")
+            message_str_for_filter = cmd_name
+            if params:
+                message_str_for_filter += f" {params}"
+
+            logger.debug(
+                f"[Discord] 斜杠指令 '{cmd_name}' 被触发。 "
+                f"原始参数: '{params}'. "
+                f"构建的指令字符串: '{message_str_for_filter}'"
+            )
+
+            # 尝试立即响应，防止超时
+            followup_webhook = None
+            try:
+                await ctx.defer()
+                followup_webhook = ctx.followup
+            except Exception as e:
+                logger.warning(f"[Discord] 指令 '{cmd_name}' defer 失败: {e}")
+
+            # 2. 构建 AstrBotMessage
+            abm = AstrBotMessage()
+            abm.type = self._get_message_type(ctx.channel, ctx.guild_id)
+            abm.group_id = self._get_channel_id(ctx.channel)
+            abm.message_str = message_str_for_filter
+            abm.sender = MessageMember(
+                user_id=str(ctx.author.id), nickname=ctx.author.display_name
+            )
+            abm.message = [Plain(text=message_str_for_filter)]
+            abm.raw_message = ctx.interaction
+            abm.self_id = self.client_self_id
+            abm.session_id = str(ctx.channel_id)
+            abm.message_id = str(ctx.interaction.id)
+
+            # 3. 将消息和 webhook 分别交给 handle_msg 处理
+            await self.handle_msg(abm, followup_webhook)
+
+        return dynamic_callback
+
+    @staticmethod
+    def _extract_command_info(
+        event_filter: Any, handler_metadata: StarHandlerMetadata
+    ) -> Tuple[str, str, CommandFilter] | None:
+        """从事件过滤器中提取指令信息"""
+        cmd_name = None
+        # is_group = False
+        cmd_filter_instance = None
+
+        if isinstance(event_filter, CommandFilter):
+            # 暂不支持子指令注册为斜杠指令
+            if (
+                event_filter.parent_command_names
+                and event_filter.parent_command_names != [""]
+            ):
+                return None
+            cmd_name = event_filter.command_name
+            cmd_filter_instance = event_filter
+
+        elif isinstance(event_filter, CommandGroupFilter):
+            # 暂不支持指令组直接注册为斜杠指令，因为它们没有 handle 方法
+            return None
+
+        if not cmd_name:
+            return None
+
+        # Discord 斜杠指令名称规范
+        if not re.match(r"^[a-z0-9_-]{1,32}$", cmd_name):
+            logger.debug(f"[Discord] 跳过不符合规范的指令: {cmd_name}")
+            return None
+
+        description = handler_metadata.desc or f"指令: {cmd_name}"
+        if len(description) > 100:
+            description = f"{description[:97]}..."
+
+        return cmd_name, description, cmd_filter_instance

astrbot/core/platform/sources/discord/discord_platform_event.py

@@ -0,0 +1,291 @@
+import asyncio
+import discord
+import base64
+from io import BytesIO
+from pathlib import Path
+from typing import Optional
+import sys
+
+from astrbot.api.event import AstrMessageEvent, MessageChain
+from astrbot.api.platform import AstrBotMessage, PlatformMetadata, At
+from astrbot.api.message_components import (
+    Plain,
+    Image,
+    File,
+    BaseMessageComponent,
+    Reply,
+)
+from astrbot import logger
+from .client import DiscordBotClient
+from .components import DiscordEmbed, DiscordView
+
+if sys.version_info >= (3, 12):
+    from typing import override
+else:
+    from typing_extensions import override
+
+
+# 自定义Discord视图组件（兼容旧版本）
+class DiscordViewComponent(BaseMessageComponent):
+    type: str = "discord_view"
+
+    def __init__(self, view: discord.ui.View):
+        self.view = view
+
+
+class DiscordPlatformEvent(AstrMessageEvent):
+    def __init__(
+        self,
+        message_str: str,
+        message_obj: AstrBotMessage,
+        platform_meta: PlatformMetadata,
+        session_id: str,
+        client: DiscordBotClient,
+        interaction_followup_webhook: Optional[discord.Webhook] = None,
+    ):
+        super().__init__(message_str, message_obj, platform_meta, session_id)
+        self.client = client
+        self.interaction_followup_webhook = interaction_followup_webhook
+
+    @override
+    async def send(self, message: MessageChain):
+        """发送消息到Discord平台"""
+
+        # 解析消息链为 Discord 所需的对象
+        try:
+            content, files, view, embeds, reference_message_id = await self._parse_to_discord(message)
+        except Exception as e:
+            logger.error(f"[Discord] 解析消息链时失败: {e}", exc_info=True)
+            return
+
+        kwargs = {}
+        if content:
+            kwargs["content"] = content
+        if files:
+            kwargs["files"] = files
+        if view:
+            kwargs["view"] = view
+        if embeds:
+            kwargs["embeds"] = embeds
+        if reference_message_id and not self.interaction_followup_webhook:
+            kwargs["reference"] = self.client.get_message(int(reference_message_id))
+        if not kwargs:
+            logger.debug("[Discord] 尝试发送空消息，已忽略。")
+            return
+
+        # 根据上下文执行发送/回复操作
+        try:
+            # -- 斜杠指令/交互上下文 --
+            if self.interaction_followup_webhook:
+                await self.interaction_followup_webhook.send(**kwargs)
+
+            # -- 常规消息上下文 --
+            else:
+                channel = await self._get_channel()
+                if not channel:
+                    return
+                else:
+                    await channel.send(**kwargs)
+
+        except Exception as e:
+            logger.error(f"[Discord] 发送消息时发生未知错误: {e}", exc_info=True)
+
+        await super().send(message)
+
+    async def _get_channel(self) -> Optional[discord.abc.Messageable]:
+        """获取当前事件对应的频道对象"""
+        try:
+            channel_id = int(self.session_id)
+            return self.client.get_channel(
+                channel_id
+            ) or await self.client.fetch_channel(channel_id)
+        except (ValueError, discord.errors.NotFound, discord.errors.Forbidden):
+            logger.error(f"[Discord] 无法获取频道 {self.session_id}")
+            return None
+
+    async def _parse_to_discord(
+        self,
+        message: MessageChain,
+    ) -> tuple[str, list[discord.File], Optional[discord.ui.View], list[discord.Embed]]:
+        """将 MessageChain 解析为 Discord 发送所需的内容"""
+        content = ""
+        files = []
+        view = None
+        embeds = []
+        reference_message_id = None
+        for i in message.chain:  # 遍历消息链
+            if isinstance(i, Plain):  # 如果是文字类型的
+                content += i.text
+            elif isinstance(i, Reply):
+                reference_message_id = i.id
+            elif isinstance(i, At):
+                content += f"<@{i.qq}>"
+            elif isinstance(i, Image):
+                logger.debug(f"[Discord] 开始处理 Image 组件: {i}")
+                try:
+                    filename = getattr(i, "filename", None)
+                    file_content = getattr(i, "file", None)
+
+                    if not file_content:
+                        logger.warning(f"[Discord] Image 组件没有 file 属性: {i}")
+                        continue
+
+                    discord_file = None
+
+                    # 1. URL
+                    if file_content.startswith("http"):
+                        logger.debug(f"[Discord] 处理 URL 图片: {file_content}")
+                        embed = discord.Embed().set_image(url=file_content)
+                        embeds.append(embed)
+                        continue
+
+                    # 2. File URI
+                    elif file_content.startswith("file:///"):
+                        logger.debug(f"[Discord] 处理 File URI: {file_content}")
+                        path = Path(file_content[8:])
+                        if await asyncio.to_thread(path.exists):
+                            file_bytes = await asyncio.to_thread(path.read_bytes)
+                            discord_file = discord.File(
+                                BytesIO(file_bytes), filename=filename or path.name
+                            )
+                        else:
+                            logger.warning(f"[Discord] 图片文件不存在: {path}")
+
+                    # 3. Base64 URI
+                    elif file_content.startswith("base64://"):
+                        logger.debug("[Discord] 处理 Base64 URI")
+                        b64_data = file_content.split("base64://", 1)[1]
+                        missing_padding = len(b64_data) % 4
+                        if missing_padding:
+                            b64_data += "=" * (4 - missing_padding)
+                        img_bytes = base64.b64decode(b64_data)
+                        discord_file = discord.File(
+                            BytesIO(img_bytes), filename=filename or "image.png"
+                        )
+
+                    # 4. 裸 Base64 或本地路径
+                    else:
+                        try:
+                            logger.debug("[Discord] 尝试作为裸 Base64 处理")
+                            b64_data = file_content
+                            missing_padding = len(b64_data) % 4
+                            if missing_padding:
+                                b64_data += "=" * (4 - missing_padding)
+                            img_bytes = base64.b64decode(b64_data)
+                            discord_file = discord.File(
+                                BytesIO(img_bytes), filename=filename or "image.png"
+                            )
+                        except (ValueError, TypeError, base64.binascii.Error):
+                            logger.debug(
+                                f"[Discord] 裸 Base64 解码失败，作为本地路径处理: {file_content}"
+                            )
+                            path = Path(file_content)
+                            if await asyncio.to_thread(path.exists):
+                                file_bytes = await asyncio.to_thread(path.read_bytes)
+                                discord_file = discord.File(
+                                    BytesIO(file_bytes), filename=filename or path.name
+                                )
+                            else:
+                                logger.warning(f"[Discord] 图片文件不存在: {path}")
+
+                    if discord_file:
+                        files.append(discord_file)
+
+                except Exception:
+                    # 使用 getattr 来安全地访问 i.file，以防 i 本身就是问题
+                    file_info = getattr(i, "file", "未知")
+                    logger.error(
+                        f"[Discord] 处理图片时发生未知严重错误: {file_info}",
+                        exc_info=True,
+                    )
+            elif isinstance(i, File):
+                try:
+                    file_path_str = await i.get_file()
+                    if file_path_str:
+                        path = Path(file_path_str)
+                        if await asyncio.to_thread(path.exists):
+                            file_bytes = await asyncio.to_thread(path.read_bytes)
+                            files.append(
+                                discord.File(BytesIO(file_bytes),
+                                             filename=i.name)
+                            )
+                        else:
+                            logger.warning(
+                                f"[Discord] 获取文件失败，路径不存在: {file_path_str}"
+                            )
+                    else:
+                        logger.warning(f"[Discord] 获取文件失败: {i.name}")
+                except Exception as e:
+                    logger.warning(f"[Discord] 处理文件失败: {i.name}, 错误: {e}")
+            elif isinstance(i, DiscordEmbed):
+                # Discord Embed消息
+                embeds.append(i.to_discord_embed())
+            elif isinstance(i, DiscordView):
+                # Discord视图组件（按钮、选择菜单等）
+                view = i.to_discord_view()
+            elif isinstance(i, DiscordViewComponent):
+                # 如果消息链中包含Discord视图组件（兼容旧版本）
+                if isinstance(i.view, discord.ui.View):
+                    view = i.view
+            else:
+                logger.debug(f"[Discord] 忽略了不支持的消息组件: {i.type}")
+
+        if len(content) > 2000:
+            logger.warning("[Discord] 消息内容超过2000字符，将被截断。")
+            content = content[:2000]
+        return content, files, view, embeds, reference_message_id
+
+    async def react(self, emoji: str):
+        """对原消息添加反应"""
+        try:
+            if hasattr(self.message_obj, "raw_message") and hasattr(
+                self.message_obj.raw_message, "add_reaction"
+            ):
+                await self.message_obj.raw_message.add_reaction(emoji)
+        except Exception as e:
+            logger.error(f"[Discord] 添加反应失败: {e}")
+
+    def is_slash_command(self) -> bool:
+        """判断是否为斜杠命令"""
+        return (
+            hasattr(self.message_obj, "raw_message")
+            and hasattr(self.message_obj.raw_message, "type")
+            and self.message_obj.raw_message.type
+            == discord.InteractionType.application_command
+        )
+
+    def is_button_interaction(self) -> bool:
+        """判断是否为按钮交互"""
+        return (
+            hasattr(self.message_obj, "raw_message")
+            and hasattr(self.message_obj.raw_message, "type")
+            and self.message_obj.raw_message.type == discord.InteractionType.component
+        )
+
+    def get_interaction_custom_id(self) -> str:
+        """获取交互组件的custom_id"""
+        if self.is_button_interaction():
+            try:
+                return self.message_obj.raw_message.data.get("custom_id", "")
+            except Exception:
+                pass
+        return ""
+
+    def is_mentioned(self) -> bool:
+        """判断机器人是否被@"""
+        if hasattr(self.message_obj, "raw_message") and hasattr(
+            self.message_obj.raw_message, "mentions"
+        ):
+            return any(
+                mention.id == int(self.message_obj.self_id)
+                for mention in self.message_obj.raw_message.mentions
+            )
+        return False
+
+    def get_mention_clean_content(self) -> str:
+        """获取去除@后的清洁内容"""
+        if hasattr(self.message_obj, "raw_message") and hasattr(
+            self.message_obj.raw_message, "clean_content"
+        ):
+            return self.message_obj.raw_message.clean_content
+        return self.message_str

astrbot/core/platform/sources/qqofficial/qqofficial_message_event.py

@@ -28,10 +28,8 @@ class QQOfficialMessageEvent(AstrMessageEvent):
         self.send_buffer = None
 
     async def send(self, message: MessageChain):
-        if not self.send_buffer:
-            self.send_buffer = message
-        else:
-            self.send_buffer.chain.extend(message.chain)
+        self.send_buffer = message
+        await self._post_send()
 
     async def send_streaming(self, generator, use_fallback: bool = False):
         """流式输出仅支持消息列表私聊"""

astrbot/core/platform/sources/slack/client.py

@@ -0,0 +1,162 @@
+import json
+import hmac
+import hashlib
+import asyncio
+import logging
+from typing import Callable, Optional
+from quart import Quart, request, Response
+from slack_sdk.web.async_client import AsyncWebClient
+from slack_sdk.socket_mode.aiohttp import SocketModeClient
+from slack_sdk.socket_mode.request import SocketModeRequest
+from slack_sdk.socket_mode.response import SocketModeResponse
+from astrbot.api import logger
+
+
+class SlackWebhookClient:
+    """Slack Webhook 模式客户端，使用 Quart 作为 Web 服务器"""
+
+    def __init__(
+        self,
+        web_client: AsyncWebClient,
+        signing_secret: str,
+        host: str = "0.0.0.0",
+        port: int = 3000,
+        path: str = "/slack/events",
+        event_handler: Optional[Callable] = None,
+    ):
+        self.web_client = web_client
+        self.signing_secret = signing_secret
+        self.host = host
+        self.port = port
+        self.path = path
+        self.event_handler = event_handler
+
+        self.app = Quart(__name__)
+        self._setup_routes()
+
+        # 禁用 Quart 的默认日志输出
+        logging.getLogger("quart.app").setLevel(logging.WARNING)
+        logging.getLogger("quart.serving").setLevel(logging.WARNING)
+
+        self.shutdown_event = asyncio.Event()
+
+    def _setup_routes(self):
+        """设置路由"""
+
+        @self.app.route(self.path, methods=["POST"])
+        async def slack_events():
+            """处理 Slack 事件"""
+            try:
+                # 获取请求体和头部
+                body = await request.get_data()
+                event_data = json.loads(body.decode("utf-8"))
+
+                # Verify Slack request signature
+                timestamp = request.headers.get("X-Slack-Request-Timestamp")
+                signature = request.headers.get("X-Slack-Signature")
+                if not timestamp or not signature:
+                    return Response("Missing headers", status=400)
+                # Calculate the HMAC signature
+                sig_basestring = f"v0:{timestamp}:{body.decode('utf-8')}"
+                my_signature = (
+                    "v0="
+                    + hmac.new(
+                        self.signing_secret.encode("utf-8"),
+                        sig_basestring.encode("utf-8"),
+                        hashlib.sha256,
+                    ).hexdigest()
+                )
+                # Verify the signature
+                if not hmac.compare_digest(my_signature, signature):
+                    logger.warning("Slack request signature verification failed")
+                    return Response("Invalid signature", status=400)
+                logger.info(f"Received Slack event: {event_data}")
+
+                # 处理 URL 验证事件
+                if event_data.get("type") == "url_verification":
+                    return {"challenge": event_data.get("challenge")}
+                # 处理事件
+                if self.event_handler and event_data.get("type") == "event_callback":
+                    await self.event_handler(event_data)
+
+                return Response("", status=200)
+
+            except Exception as e:
+                logger.error(f"处理 Slack 事件时出错: {e}")
+                return Response("Internal Server Error", status=500)
+
+        @self.app.route("/health", methods=["GET"])
+        async def health_check():
+            """健康检查端点"""
+            return {"status": "ok", "service": "slack-webhook"}
+
+    async def start(self):
+        """启动 Webhook 服务器"""
+        logger.info(
+            f"Slack Webhook 服务器启动中，监听 {self.host}:{self.port}{self.path}..."
+        )
+
+        await self.app.run_task(
+            host=self.host,
+            port=self.port,
+            debug=False,
+            shutdown_trigger=self.shutdown_trigger,
+        )
+
+    async def shutdown_trigger(self):
+        await self.shutdown_event.wait()
+
+    async def stop(self):
+        """停止 Webhook 服务器"""
+        self.shutdown_event.set()
+        logger.info("Slack Webhook 服务器已停止")
+
+
+class SlackSocketClient:
+    """Slack Socket 模式客户端"""
+
+    def __init__(
+        self,
+        web_client: AsyncWebClient,
+        app_token: str,
+        event_handler: Optional[Callable] = None,
+    ):
+        self.web_client = web_client
+        self.app_token = app_token
+        self.event_handler = event_handler
+        self.socket_client = None
+
+    async def _handle_events(self, _: SocketModeClient, req: SocketModeRequest):
+        """处理 Socket Mode 事件"""
+        try:
+            # 确认收到事件
+            response = SocketModeResponse(envelope_id=req.envelope_id)
+            await self.socket_client.send_socket_mode_response(response)
+
+            # 处理事件
+            if self.event_handler:
+                await self.event_handler(req)
+
+        except Exception as e:
+            logger.error(f"处理 Socket Mode 事件时出错: {e}")
+
+    async def start(self):
+        """启动 Socket Mode 连接"""
+        self.socket_client = SocketModeClient(
+            app_token=self.app_token,
+            logger=logger,
+            web_client=self.web_client,
+        )
+
+        # 注册事件处理器
+        self.socket_client.socket_mode_request_listeners.append(self._handle_events)
+
+        logger.info("Slack Socket Mode 客户端启动中...")
+        await self.socket_client.connect()
+
+    async def stop(self):
+        """停止 Socket Mode 连接"""
+        if self.socket_client:
+            await self.socket_client.disconnect()
+            await self.socket_client.close()
+        logger.info("Slack Socket Mode 客户端已停止")

astrbot/core/platform/sources/slack/slack_adapter.py

@@ -0,0 +1,396 @@
+import time
+import asyncio
+import uuid
+import aiohttp
+import re
+import base64
+from typing import Awaitable, Any
+from slack_sdk.web.async_client import AsyncWebClient
+from slack_sdk.socket_mode.request import SocketModeRequest
+from astrbot.api.platform import (
+    Platform,
+    AstrBotMessage,
+    MessageMember,
+    MessageType,
+    PlatformMetadata,
+)
+from astrbot.api.event import MessageChain
+from .slack_event import SlackMessageEvent
+from .client import SlackWebhookClient, SlackSocketClient
+from astrbot.api.message_components import *  # noqa: F403
+from astrbot.api import logger
+from astrbot.core.platform.astr_message_event import MessageSesion
+from ...register import register_platform_adapter
+
+
+@register_platform_adapter(
+    "slack", "适用于 Slack 的消息平台适配器，支持 Socket Mode 和 Webhook Mode。"
+)
+class SlackAdapter(Platform):
+    def __init__(
+        self, platform_config: dict, platform_settings: dict, event_queue: asyncio.Queue
+    ) -> None:
+        super().__init__(event_queue)
+
+        self.config = platform_config
+        self.settings = platform_settings
+        self.unique_session = platform_settings.get("unique_session", False)
+
+        self.bot_token = platform_config.get("bot_token")
+        self.app_token = platform_config.get("app_token")
+        self.signing_secret = platform_config.get("signing_secret")
+        self.connection_mode = platform_config.get("slack_connection_mode", "socket")
+        self.webhook_host = platform_config.get("slack_webhook_host", "0.0.0.0")
+        self.webhook_port = platform_config.get("slack_webhook_port", 3000)
+        self.webhook_path = platform_config.get(
+            "slack_webhook_path", "/astrbot-slack-webhook/callback"
+        )
+
+        if not self.bot_token:
+            raise ValueError("Slack bot_token 是必需的")
+
+        if self.connection_mode == "socket" and not self.app_token:
+            raise ValueError("Socket Mode 需要 app_token")
+
+        if self.connection_mode == "webhook" and not self.signing_secret:
+            raise ValueError("Webhook Mode 需要 signing_secret")
+
+        self.metadata = PlatformMetadata(
+            name="slack",
+            description="适用于 Slack 的消息平台适配器，支持 Socket Mode 和 Webhook Mode。",
+            id=self.config.get("id"),
+        )
+
+        # 初始化 Slack Web Client
+        self.web_client = AsyncWebClient(token=self.bot_token, logger=logger)
+        self.socket_client = None
+        self.webhook_client = None
+
+        self.bot_self_id = None
+
+    async def send_by_session(
+        self, session: MessageSesion, message_chain: MessageChain
+    ):
+        blocks, text = SlackMessageEvent._parse_slack_blocks(
+            message_chain=message_chain, web_client=self.web_client
+        )
+
+        try:
+            if session.message_type == MessageType.GROUP_MESSAGE:
+                # 发送到频道
+                channel_id = (
+                    session.session_id.split("_")[-1]
+                    if "_" in session.session_id
+                    else session.session_id
+                )
+                await self.web_client.chat_postMessage(
+                    channel=channel_id,
+                    text=text,
+                    blocks=blocks if blocks else None,
+                )
+            else:
+                # 发送私信
+                await self.web_client.chat_postMessage(
+                    channel=session.session_id,
+                    text=text,
+                    blocks=blocks if blocks else None,
+                )
+        except Exception as e:
+            logger.error(f"Slack 发送消息失败: {e}")
+
+        await super().send_by_session(session, message_chain)
+
+    async def convert_message(self, event: dict) -> AstrBotMessage:
+        logger.debug(f"[slack] RawMessage {event}")
+
+        abm = AstrBotMessage()
+        abm.self_id = self.bot_self_id
+
+        # 获取用户信息
+        user_id = event.get("user", "")
+        try:
+            user_info = await self.web_client.users_info(user=user_id)
+            user_data = user_info["user"]
+            user_name = user_data.get("real_name") or user_data.get("name", user_id)
+        except Exception:
+            user_name = user_id
+
+        abm.sender = MessageMember(user_id=user_id, nickname=user_name)
+
+        # 判断消息类型
+        channel_id = event.get("channel", "")
+        try:
+            channel_info = await self.web_client.conversations_info(channel=channel_id)
+            is_im = channel_info["channel"]["is_im"]
+
+            if is_im:
+                abm.type = MessageType.FRIEND_MESSAGE
+            else:
+                abm.type = MessageType.GROUP_MESSAGE
+                abm.group_id = channel_id
+        except Exception:
+            # 默认作为群组消息处理
+            abm.type = MessageType.GROUP_MESSAGE
+            abm.group_id = channel_id
+
+        # 设置会话ID
+        if self.unique_session and abm.type == MessageType.GROUP_MESSAGE:
+            abm.session_id = f"{user_id}_{channel_id}"
+        else:
+            abm.session_id = (
+                channel_id if abm.type == MessageType.GROUP_MESSAGE else user_id
+            )
+
+        abm.message_id = event.get("client_msg_id", uuid.uuid4().hex)
+        abm.timestamp = int(float(event.get("ts", time.time())))
+
+        # 处理消息内容
+        message_text = event.get("text", "")
+        abm.message_str = message_text
+        abm.message = []
+
+        # 优先使用 blocks 字段解析消息
+        if "blocks" in event and event["blocks"]:
+            abm.message = self._parse_blocks(event["blocks"])
+            # 更新 message_str
+            abm.message_str = ""
+            for component in abm.message:
+                if isinstance(component, Plain):
+                    abm.message_str += component.text
+        elif message_text:
+            # 处理传统的文本消息
+            if "<@" in message_text:
+                mentions = re.findall(r"<@([^>]+)>", message_text)
+                for mention in mentions:
+                    try:
+                        mentioned_user = await self.web_client.users_info(user=mention)
+                        user_data = mentioned_user["user"]
+                        user_name = user_data.get("real_name") or user_data.get(
+                            "name", mention
+                        )
+                        abm.message.append(At(qq=mention, name=user_name))
+                    except Exception:
+                        abm.message.append(At(qq=mention, name=""))
+
+                # 清理消息文本中的@标记
+                if clean_text := re.sub(r"<@[^>]+>", "", message_text).strip():
+                    abm.message.append(Plain(text=clean_text))
+            else:
+                abm.message.append(Plain(text=message_text))
+
+        # 处理文件附件
+        if "files" in event:
+            for file_info in event["files"]:
+                file_name = file_info.get("name", "unknown")
+                file_url = file_info.get("url_private", "")
+                if file_info.get("mimetype", "").startswith("image/"):
+                    file_url = await self.get_file_base64(file_url)
+                    abm.message.append(Image.fromBase64(base64=file_url))
+                else:
+                    # TODO: 下载鉴权
+                    abm.message.append(
+                        File(name=file_name, file=file_url, url=file_url)
+                    )
+
+        abm.raw_message = event
+        return abm
+
+    def _parse_blocks(self, blocks: list) -> list:
+        """解析 Slack blocks 格式的消息内容"""
+        message_components = []
+
+        for block in blocks:
+            block_type = block.get("type", "")
+
+            if block_type == "rich_text":
+                # 处理富文本块
+                elements = block.get("elements", [])
+                for element in elements:
+                    if element.get("type") == "rich_text_section":
+                        # 处理富文本段落
+                        section_elements = element.get("elements", [])
+                        text_content = ""
+
+                        for section_element in section_elements:
+                            element_type = section_element.get("type", "")
+
+                            if element_type == "text":
+                                # 普通文本
+                                text_content += section_element.get("text", "")
+                            elif element_type == "user":
+                                # @用户提及
+                                user_id = section_element.get("user_id", "")
+                                if user_id:
+                                    # 将之前的文本内容先添加到组件中
+                                    if text_content.strip():
+                                        message_components.append(
+                                            Plain(text=text_content)
+                                        )
+                                        text_content = ""
+                                    # 添加@提及组件
+                                    message_components.append(At(qq=user_id, name=""))
+                            elif element_type == "channel":
+                                # #频道提及
+                                channel_id = section_element.get("channel_id", "")
+                                text_content += f"#{channel_id}"
+                            elif element_type == "link":
+                                # 链接
+                                url = section_element.get("url", "")
+                                link_text = section_element.get("text", url)
+                                text_content += f"[{link_text}]({url})"
+                            elif element_type == "emoji":
+                                # 表情符号
+                                emoji_name = section_element.get("name", "")
+                                text_content += f":{emoji_name}:"
+
+                        if text_content.strip():
+                            message_components.append(Plain(text=text_content))
+
+                    elif element.get("type") == "rich_text_list":
+                        # 处理列表
+                        list_items = element.get("elements", [])
+                        list_text = ""
+                        for item in list_items:
+                            if item.get("type") == "rich_text_section":
+                                item_elements = item.get("elements", [])
+                                item_text = ""
+                                for item_element in item_elements:
+                                    if item_element.get("type") == "text":
+                                        item_text += item_element.get("text", "")
+                                list_text += f"• {item_text}\n"
+
+                        if list_text.strip():
+                            message_components.append(Plain(text=list_text.strip()))
+
+            elif block_type == "section":
+                # 处理段落块
+                if "text" in block:
+                    text_obj = block["text"]
+                    if text_obj.get("type") == "mrkdwn":
+                        text_content = text_obj.get("text", "")
+                        message_components.append(Plain(text=text_content))
+
+        return message_components
+
+    async def _handle_socket_event(self, req: SocketModeRequest):
+        """处理 Socket Mode 事件"""
+        if req.type == "events_api":
+            # 事件 API
+            event = req.payload.get("event", {})
+
+            # 忽略机器人自己的消息和消息编辑
+            if event.get("subtype") in [
+                "bot_message",
+                "message_changed",
+                "message_deleted",
+            ]:
+                return
+
+            if event.get("bot_id"):
+                return
+
+            if event.get("type") in ["message", "app_mention"]:
+                abm = await self.convert_message(event)
+                if abm:
+                    await self.handle_msg(abm)
+
+    async def get_bot_user_id(self):
+        auth_info = await self.web_client.auth_test()
+        return auth_info.get("user_id")
+
+    async def get_file_base64(self, url: str) -> str:
+        """下载 Slack 文件并返回 Base64 编码的内容"""
+        headers = {"Authorization": f"Bearer {self.bot_token}"}
+        async with aiohttp.ClientSession() as session:
+            async with session.get(url, headers=headers) as resp:
+                if resp.status == 200:
+                    content = await resp.read()
+                    base64_content = base64.b64encode(content).decode("utf-8")
+                    return base64_content
+                else:
+                    logger.error(f"Failed to download slack file: {resp.status} {await resp.text()}")
+                    raise Exception(f"下载文件失败: {resp.status}")
+
+    async def run(self) -> Awaitable[Any]:
+        self.bot_self_id = await self.get_bot_user_id()
+        logger.info(f"Slack auth test OK. Bot ID: {self.bot_self_id}")
+
+        if self.connection_mode == "socket":
+            if not self.app_token:
+                raise ValueError("Socket Mode 需要 app_token")
+
+            # 创建 Socket 客户端
+            self.socket_client = SlackSocketClient(
+                self.web_client, self.app_token, self._handle_socket_event
+            )
+
+            logger.info("Slack 适配器 (Socket Mode) 启动中...")
+            await self.socket_client.start()
+
+        elif self.connection_mode == "webhook":
+            if not self.signing_secret:
+                raise ValueError("Webhook Mode 需要 signing_secret")
+
+            # 创建 Webhook 客户端
+            self.webhook_client = SlackWebhookClient(
+                self.web_client,
+                self.signing_secret,
+                self.webhook_host,
+                self.webhook_port,
+                self.webhook_path,
+                self._handle_webhook_event,
+            )
+
+            logger.info(
+                f"Slack 适配器 (Webhook Mode) 启动中，监听 {self.webhook_host}:{self.webhook_port}{self.webhook_path}..."
+            )
+            await self.webhook_client.start()
+
+        else:
+            raise ValueError(
+                f"不支持的连接模式: {self.connection_mode}，请使用 'socket' 或 'webhook'"
+            )
+
+    async def _handle_webhook_event(self, event_data: dict):
+        """处理 Webhook 事件"""
+        event = event_data.get("event", {})
+
+        # 忽略机器人自己的消息和消息编辑
+        if event.get("subtype") in [
+            "bot_message",
+            "message_changed",
+            "message_deleted",
+        ]:
+            return
+
+        if event.get("bot_id"):
+            return
+
+        if event.get("type") in ["message", "app_mention"]:
+            abm = await self.convert_message(event)
+            if abm:
+                await self.handle_msg(abm)
+
+    async def terminate(self):
+        if self.socket_client:
+            await self.socket_client.stop()
+        if self.webhook_client:
+            await self.webhook_client.stop()
+        logger.info("Slack 适配器已被优雅地关闭")
+
+    def meta(self) -> PlatformMetadata:
+        return self.metadata
+
+    async def handle_msg(self, message: AstrBotMessage):
+        message_event = SlackMessageEvent(
+            message_str=message.message_str,
+            message_obj=message,
+            platform_meta=self.meta(),
+            session_id=message.session_id,
+            web_client=self.web_client,
+        )
+
+        self.commit_event(message_event)
+
+    def get_client(self):
+        return self.web_client

astrbot/core/platform/sources/slack/slack_event.py

@@ -0,0 +1,237 @@
+import asyncio
+import re
+from typing import AsyncGenerator
+from slack_sdk.web.async_client import AsyncWebClient
+from astrbot.api.event import AstrMessageEvent, MessageChain
+from astrbot.api.message_components import (
+    Image,
+    Plain,
+    File,
+    BaseMessageComponent,
+)
+from astrbot.api.platform import Group, MessageMember
+from astrbot.api import logger
+
+
+class SlackMessageEvent(AstrMessageEvent):
+    def __init__(
+        self,
+        message_str,
+        message_obj,
+        platform_meta,
+        session_id,
+        web_client: AsyncWebClient,
+    ):
+        super().__init__(message_str, message_obj, platform_meta, session_id)
+        self.web_client = web_client
+
+    @staticmethod
+    async def _from_segment_to_slack_block(
+        segment: BaseMessageComponent, web_client: AsyncWebClient
+    ) -> dict:
+        """将消息段转换为 Slack 块格式"""
+        if isinstance(segment, Plain):
+            return {"type": "section", "text": {"type": "mrkdwn", "text": segment.text}}
+        elif isinstance(segment, Image):
+            # upload file
+            url = segment.url or segment.file
+            if url.startswith("http"):
+                return {
+                    "type": "image",
+                    "image_url": url,
+                    "alt_text": "图片",
+                }
+            path = await segment.convert_to_file_path()
+            response = await web_client.files_upload_v2(
+                file=path,
+                filename="image.jpg",
+            )
+            if not response["ok"]:
+                logger.error(f"Slack file upload failed: {response['error']}")
+                return {
+                    "type": "section",
+                    "text": {"type": "mrkdwn", "text": "图片上传失败"},
+                }
+            image_url = response["files"][0]["url_private"]
+            logger.debug(f"Slack file upload response: {response}")
+            return {
+                "type": "image",
+                "slack_file": {
+                    "url": image_url,
+                },
+                "alt_text": "图片",
+            }
+        elif isinstance(segment, File):
+            # upload file
+            url = segment.url or segment.file
+            response = await web_client.files_upload_v2(
+                file=url,
+                filename=segment.name or "file",
+            )
+            if not response["ok"]:
+                logger.error(f"Slack file upload failed: {response['error']}")
+                return {
+                    "type": "section",
+                    "text": {"type": "mrkdwn", "text": "文件上传失败"},
+                }
+            file_url = response["files"][0]["permalink"]
+            return {"type": "section", "text": {"type": "mrkdwn", "text": f"文件: <{file_url}|{segment.name or '文件'}>"}}
+        else:
+            return {"type": "section", "text": {"type": "mrkdwn", "text": str(segment)}}
+
+    @staticmethod
+    async def _parse_slack_blocks(
+        message_chain: MessageChain, web_client: AsyncWebClient
+    ):
+        """解析成 Slack 块格式"""
+        blocks = []
+        text_content = ""
+
+        for segment in message_chain.chain:
+            if isinstance(segment, Plain):
+                text_content += segment.text
+            else:
+                # 如果有文本内容，先添加文本块
+                if text_content.strip():
+                    blocks.append(
+                        {
+                            "type": "section",
+                            "text": {"type": "mrkdwn", "text": text_content},
+                        }
+                    )
+                    text_content = ""
+
+                # 添加其他类型的块
+                block = await SlackMessageEvent._from_segment_to_slack_block(
+                    segment, web_client
+                )
+                blocks.append(block)
+
+        # 如果最后还有文本内容
+        if text_content.strip():
+            blocks.append(
+                {"type": "section", "text": {"type": "mrkdwn", "text": text_content}}
+            )
+
+        return blocks, "" if blocks else text_content
+
+    async def send(self, message: MessageChain):
+        blocks, text = await SlackMessageEvent._parse_slack_blocks(
+            message, self.web_client
+        )
+
+        try:
+            if self.get_group_id():
+                # 发送到频道
+                await self.web_client.chat_postMessage(
+                    channel=self.get_group_id(),
+                    text=text,
+                    blocks=blocks or None,
+                )
+            else:
+                # 发送私信
+                await self.web_client.chat_postMessage(
+                    channel=self.get_sender_id(),
+                    text=text,
+                    blocks=blocks or None,
+                )
+        except Exception:
+            # 如果块发送失败，尝试只发送文本
+            fallback_text = ""
+            for segment in message.chain:
+                if isinstance(segment, Plain):
+                    fallback_text += segment.text
+                elif isinstance(segment, File):
+                    fallback_text += f" [文件: {segment.name}] "
+                elif isinstance(segment, Image):
+                    fallback_text += " [图片] "
+
+            if self.get_group_id():
+                await self.web_client.chat_postMessage(
+                    channel=self.get_group_id(), text=fallback_text
+                )
+            else:
+                await self.web_client.chat_postMessage(
+                    channel=self.get_sender_id(), text=fallback_text
+                )
+
+        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
+            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)  # 限速
+
+        if buffer.strip():
+            await self.send(MessageChain([Plain(buffer)]))
+        return await super().send_streaming(generator, use_fallback)
+
+    async def get_group(self, group_id=None, **kwargs):
+        if group_id:
+            channel_id = group_id
+        elif self.get_group_id():
+            channel_id = self.get_group_id()
+        else:
+            return None
+
+        try:
+            # 获取频道信息
+            channel_info = await self.web_client.conversations_info(channel=channel_id)
+
+            # 获取频道成员
+            members_response = await self.web_client.conversations_members(
+                channel=channel_id
+            )
+
+            members = []
+            for member_id in members_response["members"]:
+                try:
+                    user_info = await self.web_client.users_info(user=member_id)
+                    user_data = user_info["user"]
+                    members.append(
+                        MessageMember(
+                            user_id=member_id,
+                            nickname=user_data.get("real_name")
+                            or user_data.get("name", member_id),
+                        )
+                    )
+                except Exception:
+                    # 如果获取用户信息失败，使用默认信息
+                    members.append(MessageMember(user_id=member_id, nickname=member_id))
+
+            channel_data = channel_info["channel"]
+            return Group(
+                group_id=channel_id,
+                group_name=channel_data.get("name", ""),
+                group_avatar="",
+                group_admins=[],  # Slack 的管理员信息需要特殊权限获取
+                group_owner=channel_data.get("creator", ""),
+                members=members,
+            )
+        except Exception:
+            return None

astrbot/core/platform/sources/telegram/tg_adapter.py

@@ -144,8 +144,8 @@ class TelegramPlatformAdapter(Platform):
         command_dict = {}
         skip_commands = {"start"}
 
-        for handler_md in star_handlers_registry._handlers:
-            handler_metadata = handler_md[1]
+        for handler_md in star_handlers_registry:
+            handler_metadata = handler_md
             if not star_map[handler_metadata.handler_module_path].activated:
                 continue
             for event_filter in handler_metadata.event_filters:

astrbot/core/platform/sources/telegram/tg_event.py

@@ -1,4 +1,5 @@
 import os
+import re
 import asyncio
 import telegramify_markdown
 from astrbot.api.event import AstrMessageEvent, MessageChain
@@ -18,6 +19,16 @@ from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 
 
 class TelegramPlatformEvent(AstrMessageEvent):
+    # Telegram 的最大消息长度限制
+    MAX_MESSAGE_LENGTH = 4096
+
+    SPLIT_PATTERNS = {
+        "paragraph": re.compile(r"\n\n"),
+        "line": re.compile(r"\n"),
+        "sentence": re.compile(r"[.!?。！？]"),
+        "word": re.compile(r"\s"),
+    }
+
     def __init__(
         self,
         message_str: str,
@@ -29,8 +40,33 @@ class TelegramPlatformEvent(AstrMessageEvent):
         super().__init__(message_str, message_obj, platform_meta, session_id)
         self.client = client
 
-    @staticmethod
-    async def send_with_client(client: ExtBot, message: MessageChain, user_name: str):
+    @classmethod
+    def _split_message(cls, text: str) -> list[str]:
+        if len(text) <= cls.MAX_MESSAGE_LENGTH:
+            return [text]
+
+        chunks = []
+        while text:
+            if len(text) <= cls.MAX_MESSAGE_LENGTH:
+                chunks.append(text)
+                break
+
+            split_point = cls.MAX_MESSAGE_LENGTH
+            segment = text[: cls.MAX_MESSAGE_LENGTH]
+
+            for _, pattern in cls.SPLIT_PATTERNS.items():
+                if matches := list(pattern.finditer(segment)):
+                    last_match = matches[-1]
+                    split_point = last_match.end()
+                    break
+
+            chunks.append(text[:split_point])
+            text = text[split_point:].lstrip()
+
+        return chunks
+
+    @classmethod
+    async def send_with_client(cls, client: ExtBot, message: MessageChain, user_name: str):
         image_path = None
 
         has_reply = False
@@ -59,19 +95,22 @@ class TelegramPlatformEvent(AstrMessageEvent):
 
             if isinstance(i, Plain):
                 if at_user_id and not at_flag:
-                    i.text = f"@{at_user_id} " + i.text
+                    i.text = f"@{at_user_id} {i.text}"
                     at_flag = True
-                text = i.text
-                try:
-                    text = telegramify_markdown.markdownify(
-                        i.text, max_line_length=None, normalize_whitespace=False
-                    )
-                except Exception as e:
-                    logger.warning(
-                        f"MarkdownV2 conversion failed: {e}. Using plain text instead."
-                    )
-                    return
-                await client.send_message(text=text, parse_mode="MarkdownV2", **payload)
+                chunks = cls._split_message(i.text)
+                for chunk in chunks:
+                    try:
+                        md_text = telegramify_markdown.markdownify(
+                            chunk, max_line_length=None, normalize_whitespace=False
+                        )
+                        await client.send_message(
+                            text=md_text, parse_mode="MarkdownV2", **payload
+                        )
+                    except Exception as e:
+                        logger.warning(
+                            f"MarkdownV2 send failed: {e}. Using plain text instead."
+                        )
+                        await client.send_message(text=chunk, **payload)
             elif isinstance(i, Image):
                 image_path = await i.convert_to_file_path()
                 await client.send_photo(photo=image_path, **payload)
@@ -119,6 +158,12 @@ class TelegramPlatformEvent(AstrMessageEvent):
 
         async for chain in generator:
             if isinstance(chain, MessageChain):
+                if chain.type == "break":
+                    # 分割符
+                    message_id = None  # 重置消息 ID
+                    delta = ""  # 重置 delta
+                    continue
+
                 # 处理消息链中的每个组件
                 for i in chain.chain:
                     if isinstance(i, Plain):
@@ -147,17 +192,7 @@ class TelegramPlatformEvent(AstrMessageEvent):
                         continue
 
                 # Plain
-                if not message_id:
-                    try:
-                        msg = await self.client.send_message(text=delta, **payload)
-                        current_content = delta
-                    except Exception as e:
-                        logger.warning(f"发送消息失败(streaming): {e!s}")
-                    message_id = msg.message_id
-                    last_edit_time = (
-                        asyncio.get_event_loop().time()
-                    )  # 记录初始消息发送时间
-                else:
+                if message_id and len(delta) <= self.MAX_MESSAGE_LENGTH:
                     current_time = asyncio.get_event_loop().time()
                     time_since_last_edit = current_time - last_edit_time
 
@@ -176,6 +211,18 @@ class TelegramPlatformEvent(AstrMessageEvent):
                         last_edit_time = (
                             asyncio.get_event_loop().time()
                         )  # 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
+                    try:
+                        msg = await self.client.send_message(text=delta, **payload)
+                        current_content = delta
+                        delta = ""
+                    except Exception as e:
+                        logger.warning(f"发送消息失败(streaming): {e!s}")
+                    message_id = msg.message_id
+                    last_edit_time = (
+                        asyncio.get_event_loop().time()
+                    )  # 记录初始消息发送时间
 
         try:
             if delta and current_content != delta:

astrbot/core/platform/sources/webchat/webchat_event.py

@@ -35,6 +35,7 @@ class WebChatMessageEvent(AstrMessageEvent):
                         "cid": cid,
                         "data": data,
                         "streaming": streaming,
+                        "chain_type": message.type,
                     }
                 )
             elif isinstance(comp, Image):
@@ -110,6 +111,18 @@ class WebChatMessageEvent(AstrMessageEvent):
     async def send_streaming(self, generator, use_fallback: bool = False):
         final_data = ""
         async for chain in generator:
+            if chain.type == "break" and final_data:
+                # 分割符
+                await web_chat_back_queue.put(
+                    {
+                        "type": "end",
+                        "data": final_data,
+                        "streaming": True,
+                        "cid": self.session_id.split("!")[-1],
+                    }
+                )
+                final_data = ""
+                continue
             final_data += await WebChatMessageEvent._send(
                 chain, session_id=self.session_id, streaming=True
             )

astrbot/core/platform/sources/wechatpadpro/wechatpadpro_adapter.py

@@ -1,14 +1,16 @@
 import asyncio
+import base64
 import json
 import os
+import traceback
 import time
 from typing import Optional
 
 import aiohttp
+import anyio
 import websockets
-
 from astrbot import logger
-from astrbot.api.message_components import Plain, Image
+from astrbot.api.message_components import Plain, Image, At, Record
 from astrbot.api.platform import Platform, PlatformMetadata
 from astrbot.core.message.message_event_result import MessageChain
 from astrbot.core.platform.astrbot_message import (
@@ -22,6 +24,13 @@ from astrbot.core.platform.astr_message_event import MessageSesion
 from ...register import register_platform_adapter
 from .wechatpadpro_message_event import WeChatPadProMessageEvent
 
+try:
+    from .xml_data_parser import GeweDataParser
+except ImportError as e:
+    logger.warning(
+        f"警告: 可能未安装 defusedxml 依赖库，将导致无法解析微信的 表情包、引用 类型的消息: {str(e)}"
+    )
+
 
 @register_platform_adapter("wechatpadpro", "WeChatPadPro 消息平台适配器")
 class WeChatPadProAdapter(Platform):
@@ -59,6 +68,18 @@ class WeChatPadProAdapter(Platform):
         )  # 持久化文件路径
         self.ws_handle_task = None
 
+        # 添加图片消息缓存，用于引用消息处理
+        self.cached_images = {}
+        """缓存图片消息。key是NewMsgId (对应引用消息的svrid)，value是图片的base64数据"""
+        # 设置缓存大小限制，避免内存占用过大
+        self.max_image_cache = 50
+
+        # 添加文本消息缓存，用于引用消息处理
+        self.cached_texts = {}
+        """缓存文本消息。key是NewMsgId (对应引用消息的svrid)，value是消息文本内容"""
+        # 设置文本缓存大小限制
+        self.max_text_cache = 100
+
     async def run(self) -> None:
         """
         启动平台适配器的运行实例。
@@ -69,39 +90,42 @@ class WeChatPadProAdapter(Platform):
             self.auth_key = loaded_credentials.get("auth_key")
             self.wxid = loaded_credentials.get("wxid")
 
+        isLoginIn = await self.check_online_status()
+
         # 检查在线状态
-        if self.auth_key and await self.check_online_status():
-            logger.info("WeChatPadPro 设备已在线，跳过扫码登录。")
+        if self.auth_key and isLoginIn:
+            logger.info("WeChatPadPro 设备已在线，凭据存在，跳过扫码登录。")
             # 如果在线，连接 WebSocket 接收消息
             self.ws_handle_task = asyncio.create_task(self.connect_websocket())
         else:
-            logger.info("WeChatPadPro 设备不在线或无可用凭据，开始扫码登录流程。")
             # 1. 生成授权码
-            await self.generate_auth_key()
-
             if not self.auth_key:
-                logger.error("无法获取授权码，WeChatPadPro 适配器启动失败。")
-                return
+                logger.info("WeChatPadPro 无可用凭据，将生成新的授权码。")
+                await self.generate_auth_key()
 
             # 2. 获取登录二维码
-            qr_code_url = await self.get_login_qr_code()
+            if not isLoginIn:
+                logger.info("WeChatPadPro 设备已离线，开始扫码登录。")
+                qr_code_url = await self.get_login_qr_code()
 
-            if qr_code_url:
-                logger.info(f"请扫描以下二维码登录: {qr_code_url}")
-            else:
-                logger.error("无法获取登录二维码。")
-                return
+                if qr_code_url:
+                    logger.info(f"请扫描以下二维码登录: {qr_code_url}")
+                else:
+                    logger.error("无法获取登录二维码。")
+                    return
 
-            # 3. 检测扫码状态
-            login_successful = await self.check_login_status()
+                # 3. 检测扫码状态
+                login_successful = await self.check_login_status()
 
-            if login_successful:
-                # 登录成功后，连接 WebSocket 接收消息
-                self.ws_handle_task = asyncio.create_task(self.connect_websocket())
-            else:
-                logger.warning("登录失败或超时，WeChatPadPro 适配器将关闭。")
-                await self.terminate()
-                return
+                if login_successful:
+                    logger.info("登录成功，WeChatPadPro适配器已连接。")
+                else:
+                    logger.warning("登录失败或超时，WeChatPadPro 适配器将关闭。")
+                    await self.terminate()
+                    return
+
+            # 登录成功后，连接 WebSocket 接收消息
+            self.ws_handle_task = asyncio.create_task(self.connect_websocket())
 
         self._shutdown_event = asyncio.Event()
         await self._shutdown_event.wait()
@@ -135,7 +159,6 @@ class WeChatPadProAdapter(Platform):
             os.makedirs(data_dir, exist_ok=True)
             with open(self.credentials_file, "w") as f:
                 json.dump(credentials, f)
-            logger.info("成功保存 WeChatPadPro 凭据。")
         except Exception as e:
             logger.error(f"保存 WeChatPadPro 凭据失败: {e}")
 
@@ -143,6 +166,8 @@ class WeChatPadProAdapter(Platform):
         """
         检查 WeChatPadPro 设备是否在线。
         """
+        if not self.auth_key:
+            return False
         url = f"{self.base_url}/login/GetLoginStatus"
         params = {"key": self.auth_key}
 
@@ -156,21 +181,33 @@ class WeChatPadProAdapter(Platform):
                         if login_state == 1:
                             logger.info("WeChatPadPro 设备当前在线。")
                             return True
-                        else:
-                            logger.info(
-                                f"WeChatPadPro 设备不在线，登录状态: {login_state}"
-                            )
+                        # login_state == 3 为离线状态
+                        elif login_state == 3:
+                            logger.info("WeChatPadPro 设备不在线。")
                             return False
+                        else:
+                            logger.error(f"未知的在线状态: {response_data}")
+                            return False
+                    # Code == 300 为微信退出状态。
+                    elif response.status == 200 and response_data.get("Code") == 300:
+                        logger.info("WeChatPadPro 设备已退出。")
+                        return False
+                    elif response.status == 200 and response_data.get("Code") == -2:
+                        # 该链接不存在
+                        self.auth_key = None
+                        return False
                     else:
                         logger.error(
                             f"检查在线状态失败: {response.status}, {response_data}"
                         )
                         return False
+
             except aiohttp.ClientConnectorError as e:
                 logger.error(f"连接到 WeChatPadPro 服务失败: {e}")
                 return False
             except Exception as e:
                 logger.error(f"检查在线状态时发生错误: {e}")
+                logger.error(traceback.format_exc())
                 return False
 
     async def generate_auth_key(self):
@@ -179,7 +216,7 @@ class WeChatPadProAdapter(Platform):
         """
         url = f"{self.base_url}/admin/GenAuthKey1"
         params = {"key": self.admin_key}
-        payload = {"Count": 1, "Days": 30}  # 生成一个有效期30天的授权码
+        payload = {"Count": 1, "Days": 365}  # 生成一个有效期365天的授权码
 
         async with aiohttp.ClientSession() as session:
             try:
@@ -194,7 +231,7 @@ class WeChatPadProAdapter(Platform):
                             and len(response_data["Data"]) > 0
                         ):
                             self.auth_key = response_data["Data"][0]
-                            logger.info("成功获取授权码")
+                            logger.info(f"成功获取授权码 {self.auth_key[:8]}...")
                         else:
                             logger.error(
                                 f"生成授权码成功但未找到授权码: {response_data}"
@@ -220,7 +257,6 @@ class WeChatPadProAdapter(Platform):
             try:
                 async with session.post(url, params=params, json=payload) as response:
                     response_data = await response.json()
-                    # 修正成功判断条件和数据提取路径
                     if response.status == 200 and response_data.get("Code") == 200:
                         # 二维码地址在 Data.QrCodeUrl 字段中
                         if response_data.get("Data") and response_data["Data"].get(
@@ -232,6 +268,13 @@ class WeChatPadProAdapter(Platform):
                                 f"获取登录二维码成功但未找到二维码地址: {response_data}"
                             )
                             return None
+                    elif "该 key 无效" in response_data.get("Text"):
+                        logger.error(
+                            "授权码无效，已经清除。请重新启动 AstrBot 或者本消息适配器。原因也可能是 WeChatPadPro 的 MySQL 服务没有启动成功，请检查 WeChatPadPro 服务的日志。"
+                        )
+                        self.auth_key = None
+                        self.save_credentials()
+                        return None
                     else:
                         logger.error(
                             f"获取登录二维码失败: {response.status}, {response_data}"
@@ -324,7 +367,7 @@ class WeChatPadProAdapter(Platform):
         while True:
             try:
                 async with websockets.connect(ws_url) as websocket:
-                    logger.info("WebSocket 连接成功。")
+                    logger.debug("WebSocket 连接成功。")
                     # 设置空闲超时重连
                     wait_time = (
                         self.active_message_poll_interval
@@ -336,12 +379,10 @@ class WeChatPadProAdapter(Platform):
                             message = await asyncio.wait_for(
                                 websocket.recv(), timeout=wait_time
                             )
-                            logger.info(message)
+                            # logger.debug(message) # 不显示原始消息内容
                             asyncio.create_task(self.handle_websocket_message(message))
                         except asyncio.TimeoutError:
-                            logger.warning(
-                                f"WebSocket 连接空闲超过 {wait_time} s"
-                            )
+                            logger.debug(f"WebSocket 连接空闲超过 {wait_time} s")
                             break
                         except websockets.exceptions.ConnectionClosedOK:
                             logger.info("WebSocket 连接正常关闭。")
@@ -350,7 +391,9 @@ class WeChatPadProAdapter(Platform):
                             logger.error(f"处理 WebSocket 消息时发生错误: {e}")
                             break
             except Exception as e:
-                logger.error(f"WebSocket 连接失败: {e}")
+                logger.error(
+                    f"WebSocket 连接失败: {e}, 请检查WeChatPadPro服务状态，或尝试重启WeChatPadPro适配器。"
+                )
                 await asyncio.sleep(5)
 
     async def handle_websocket_message(self, message: str):
@@ -443,6 +486,7 @@ class WeChatPadProAdapter(Platform):
         """
         if from_user_name == "weixin":
             return False
+        at_me = False
         if "@chatroom" in from_user_name:
             abm.type = MessageType.GROUP_MESSAGE
             abm.group_id = from_user_name
@@ -461,9 +505,17 @@ class WeChatPadProAdapter(Platform):
 
             # 对于群聊，session_id 可以是群聊 ID 或发送者 ID + 群聊 ID (如果 unique_session 为 True)
             if self.unique_session:
-                abm.session_id = f"{from_user_name}_{to_user_name}"
+                abm.session_id = f"{from_user_name}#{abm.sender.user_id}"
             else:
                 abm.session_id = from_user_name
+
+            msg_source = raw_message.get("msg_source", "")
+            if self.wxid in msg_source:
+                at_me = True
+            if "在群聊中@了你" in raw_message.get("push_content", ""):
+                at_me = True
+            if at_me:
+                abm.message.insert(0, At(qq=abm.self_id, name=""))
         else:
             abm.type = MessageType.FRIEND_MESSAGE
             abm.group_id = ""
@@ -544,6 +596,32 @@ class WeChatPadProAdapter(Platform):
                 logger.error(f"下载图片时发生错误: {e}")
                 return None
 
+    async def download_voice(
+        self, to_user_name: str, new_msg_id: str, bufid: str, length: int
+    ):
+        """下载原始音频。"""
+        url = f"{self.base_url}/message/GetMsgVoice"
+        params = {"key": self.auth_key}
+        payload = {
+            "Bufid": bufid,
+            "ToUserName": to_user_name,
+            "NewMsgId": new_msg_id,
+            "Length": length,
+        }
+        async with aiohttp.ClientSession() as session:
+            try:
+                async with session.post(url, params=params, json=payload) as response:
+                    if response.status == 200:
+                        return await response.json()
+                    logger.error(f"下载音频失败: {response.status}")
+                    return None
+            except aiohttp.ClientConnectorError as e:
+                logger.error(f"连接到 WeChatPadPro 服务失败: {e}")
+                return None
+            except Exception as e:
+                logger.error(f"下载音频时发生错误: {e}")
+                return None
+
     async def _process_message_content(
         self, abm: AstrBotMessage, raw_message: dict, msg_type: int, content: str
     ):
@@ -555,12 +633,82 @@ class WeChatPadProAdapter(Platform):
             if abm.type == MessageType.GROUP_MESSAGE:
                 parts = content.split(":\n", 1)
                 if len(parts) == 2:
-                    abm.message_str = parts[1]
-                    abm.message.append(Plain(abm.message_str))
+                    message_content = parts[1]
+                    abm.message_str = message_content
+
+                    # 检查是否@了机器人，参考 gewechat 的实现方式
+                    # 微信大部分客户端在@用户昵称后面，紧接着是一个\u2005字符（四分之一空格）
+                    at_me = False
+
+                    # 检查 msg_source 中是否包含机器人的 wxid
+                    # wechatpadpro 的格式: <atuserlist>wxid</atuserlist>
+                    # gewechat 的格式: <atuserlist><![CDATA[wxid]]></atuserlist>
+                    msg_source = raw_message.get("msg_source", "")
+                    if (
+                        f"<atuserlist>{abm.self_id}</atuserlist>" in msg_source
+                        or f"<atuserlist>{abm.self_id}," in msg_source
+                        or f",{abm.self_id}</atuserlist>" in msg_source
+                    ):
+                        at_me = True
+
+                    # 也检查 push_content 中是否有@提示
+                    push_content = raw_message.get("push_content", "")
+                    if "在群聊中@了你" in push_content:
+                        at_me = True
+
+                    if at_me:
+                        # 被@了，在消息开头插入At组件（参考gewechat的做法）
+                        bot_nickname = await self._get_group_member_nickname(
+                            abm.group_id, abm.self_id
+                        )
+                        abm.message.insert(
+                            0, At(qq=abm.self_id, name=bot_nickname or abm.self_id)
+                        )
+
+                        # 只有当消息内容不仅仅是@时才添加Plain组件
+                        if "\u2005" in message_content:
+                            # 检查@之后是否还有其他内容
+                            parts = message_content.split("\u2005")
+                            if len(parts) > 1 and any(
+                                part.strip() for part in parts[1:]
+                            ):
+                                abm.message.append(Plain(message_content))
+                        else:
+                            # 检查是否只包含@机器人
+                            is_pure_at = False
+                            if (
+                                bot_nickname
+                                and message_content.strip() == f"@{bot_nickname}"
+                            ):
+                                is_pure_at = True
+                            if not is_pure_at:
+                                abm.message.append(Plain(message_content))
+                    else:
+                        # 没有@机器人，作为普通文本处理
+                        abm.message.append(Plain(message_content))
                 else:
                     abm.message.append(Plain(abm.message_str))
             else:  # 私聊消息
                 abm.message.append(Plain(abm.message_str))
+
+            # 缓存文本消息，以便引用消息可以查找
+            try:
+                # 获取msg_id作为缓存的key
+                new_msg_id = raw_message.get("new_msg_id")
+                if new_msg_id:
+                    # 限制缓存大小
+                    if (
+                        len(self.cached_texts) >= self.max_text_cache
+                        and self.cached_texts
+                    ):
+                        # 删除最早的一条缓存
+                        oldest_key = next(iter(self.cached_texts))
+                        self.cached_texts.pop(oldest_key)
+
+                    logger.debug(f"缓存文本消息，new_msg_id={new_msg_id}")
+                    self.cached_texts[str(new_msg_id)] = content
+            except Exception as e:
+                logger.error(f"缓存文本消息失败: {e}")
         elif msg_type == 3:
             # 图片消息
             from_user_name = raw_message.get("from_user_name", {}).get("str", "")
@@ -574,15 +722,87 @@ class WeChatPadProAdapter(Platform):
             )
             if image_bs64_data:
                 abm.message.append(Image.fromBase64(image_bs64_data))
+                # 缓存图片，以便引用消息可以查找
+                try:
+                    # 获取msg_id作为缓存的key
+                    new_msg_id = raw_message.get("new_msg_id")
+                    if new_msg_id:
+                        # 限制缓存大小
+                        if (
+                            len(self.cached_images) >= self.max_image_cache
+                            and self.cached_images
+                        ):
+                            # 删除最早的一条缓存
+                            oldest_key = next(iter(self.cached_images))
+                            self.cached_images.pop(oldest_key)
+
+                        logger.debug(f"缓存图片消息，new_msg_id={new_msg_id}")
+                        self.cached_images[str(new_msg_id)] = image_bs64_data
+                except Exception as e:
+                    logger.error(f"缓存图片消息失败: {e}")
         elif msg_type == 47:
             # 视频消息 (注意：表情消息也是 47，需要区分)
-            logger.warning("收到视频消息，待实现。")
+            data_parser = GeweDataParser(
+                content=content,
+                is_private_chat=(abm.type != MessageType.GROUP_MESSAGE),
+                raw_message=raw_message,
+            )
+            emoji_message = data_parser.parse_emoji()
+            if emoji_message is not None:
+                abm.message.append(emoji_message)
         elif msg_type == 50:
-            # 语音/视频
             logger.warning("收到语音/视频消息，待实现。")
+        elif msg_type == 34:
+            # 语音消息
+            bufid = 0
+            to_user_name = raw_message.get("to_user_name", {}).get("str", "")
+            new_msg_id = raw_message.get("new_msg_id")
+            data_parser = GeweDataParser(
+                content=content,
+                is_private_chat=(abm.type != MessageType.GROUP_MESSAGE),
+                raw_message=raw_message,
+            )
+
+            voicemsg = data_parser._format_to_xml().find("voicemsg")
+            bufid = voicemsg.get("bufid") or "0"
+            length = int(voicemsg.get("length") or 0)
+            voice_resp = await self.download_voice(
+                to_user_name=to_user_name,
+                new_msg_id=new_msg_id,
+                bufid=bufid,
+                length=length,
+            )
+            voice_bs64_data = voice_resp.get("Data", {}).get("Base64", None)
+            if voice_bs64_data:
+                voice_bs64_data = base64.b64decode(voice_bs64_data)
+                temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+                file_path = os.path.join(
+                    temp_dir, f"wechatpadpro_voice_{abm.message_id}.silk"
+                )
+
+                async with await anyio.open_file(file_path, "wb") as f:
+                    await f.write(voice_bs64_data)
+                abm.message.append(Record(file=file_path, url=file_path))
         elif msg_type == 49:
-            # 引用消息
-            logger.warning("收到引用消息，待实现。")
+            try:
+                parser = GeweDataParser(
+                    content=content,
+                    is_private_chat=(abm.type != MessageType.GROUP_MESSAGE),
+                    cached_texts=self.cached_texts,
+                    cached_images=self.cached_images,
+                    raw_message=raw_message,
+                    downloader=self._download_raw_image,
+                )
+                components = await parser.parse_mutil_49()
+                if components:
+                    abm.message.extend(components)
+                    abm.message_str = "\n".join(
+                        c.text for c in components if isinstance(c, Plain)
+                    )
+            except Exception as e:
+                logger.warning(f"msg_type 49 处理失败: {e}")
+                abm.message.append(Plain("[XML 消息处理失败]"))
+                abm.message_str = "[XML 消息处理失败]"
         else:
             logger.warning(f"收到未处理的消息类型: {msg_type}。")
 
@@ -612,7 +832,10 @@ class WeChatPadProAdapter(Platform):
         # 根据 session_id 判断消息类型
         if "@chatroom" in session.session_id:
             dummy_message_obj.type = MessageType.GROUP_MESSAGE
-            dummy_message_obj.group_id = session.session_id
+            if "#" in session.session_id:
+                dummy_message_obj.group_id = session.session_id.split("#")[0]
+            else:
+                dummy_message_obj.group_id = session.session_id
             dummy_message_obj.sender = MessageMember(user_id="", nickname="")
         else:
             dummy_message_obj.type = MessageType.FRIEND_MESSAGE
@@ -627,3 +850,67 @@ class WeChatPadProAdapter(Platform):
         )
         # 调用实例方法 send
         await sending_event.send(message_chain)
+
+    async def get_contact_list(self):
+        """
+        获取联系人列表。
+        """
+        url = f"{self.base_url}/friend/GetContactList"
+        params = {"key": self.auth_key}
+        payload = {"CurrentChatRoomContactSeq": 0, "CurrentWxcontactSeq": 0}
+        async with aiohttp.ClientSession() as session:
+            try:
+                async with session.post(url, params=params, json=payload) as response:
+                    if response.status != 200:
+                        logger.error(f"获取联系人列表失败: {response.status}")
+                        return None
+                    result = await response.json()
+                    if result.get("Code") == 200 and result.get("Data"):
+                        contact_list = (
+                            result.get("Data", {})
+                            .get("ContactList", {})
+                            .get("contactUsernameList", [])
+                        )
+                        return contact_list
+                    else:
+                        logger.error(f"获取联系人列表失败: {result}")
+                        return None
+            except aiohttp.ClientConnectorError as e:
+                logger.error(f"连接到 WeChatPadPro 服务失败: {e}")
+                return None
+            except Exception as e:
+                logger.error(f"获取联系人列表时发生错误: {e}")
+                return None
+
+    async def get_contact_details_list(
+        self, room_wx_id_list: list[str] = None, user_names: list[str] = None
+    ) -> Optional[dict]:
+        """
+        获取联系人详情列表。
+        """
+        if room_wx_id_list is None:
+            room_wx_id_list = []
+        if user_names is None:
+            user_names = []
+        url = f"{self.base_url}/friend/GetContactDetailsList"
+        params = {"key": self.auth_key}
+        payload = {"RoomWxIDList": room_wx_id_list, "UserNames": user_names}
+        async with aiohttp.ClientSession() as session:
+            try:
+                async with session.post(url, params=params, json=payload) as response:
+                    if response.status != 200:
+                        logger.error(f"获取联系人详情列表失败: {response.status}")
+                        return None
+                    result = await response.json()
+                    if result.get("Code") == 200 and result.get("Data"):
+                        contact_list = result.get("Data", {}).get("contactList", {})
+                        return contact_list
+                    else:
+                        logger.error(f"获取联系人详情列表失败: {result}")
+                        return None
+            except aiohttp.ClientConnectorError as e:
+                logger.error(f"连接到 WeChatPadPro 服务失败: {e}")
+                return None
+            except Exception as e:
+                logger.error(f"获取联系人详情列表时发生错误: {e}")
+                return None

astrbot/core/platform/sources/wechatpadpro/wechatpadpro_message_event.py

@@ -7,11 +7,17 @@ import aiohttp
 from PIL import Image as PILImage  # 使用别名避免冲突
 
 from astrbot import logger
-from astrbot.core.message.components import Image, Plain  # Import Image
+from astrbot.core.message.components import (
+    Image,
+    Plain,
+    WechatEmoji,
+    Record,
+)  # Import Image
 from astrbot.core.message.message_event_result import MessageChain
 from astrbot.core.platform.astr_message_event import AstrMessageEvent
 from astrbot.core.platform.astrbot_message import AstrBotMessage, MessageType
 from astrbot.core.platform.platform_metadata import PlatformMetadata
+from astrbot.core.utils.tencent_record_helper import audio_to_tencent_silk_base64
 
 if TYPE_CHECKING:
     from .wechatpadpro_adapter import WeChatPadProAdapter
@@ -38,6 +44,10 @@ class WeChatPadProMessageEvent(AstrMessageEvent):
                     await self._send_text(session, comp.text)
                 elif isinstance(comp, Image):
                     await self._send_image(session, comp)
+                elif isinstance(comp, WechatEmoji):
+                    await self._send_emoji(session, comp)
+                elif isinstance(comp, Record):
+                    await self._send_voice(session, comp)
         await super().send(message)
 
     async def _send_image(self, session: aiohttp.ClientSession, comp: Image):
@@ -71,14 +81,48 @@ class WeChatPadProMessageEvent(AstrMessageEvent):
             # logger.info(f"已添加 @ 信息: {message_text}")
         else:
             message_text = text
+        if self.get_group_id() and "#" in self.session_id:
+            session_id = self.session_id.split("#")[0]
+        else:
+            session_id = self.session_id
         payload = {
             "MsgItem": [
-                {"MsgType": 1, "TextContent": message_text, "ToUserName": self.session_id}
+                {
+                    "MsgType": 1,
+                    "TextContent": message_text,
+                    "ToUserName": session_id,
+                }
             ]
         }
         url = f"{self.adapter.base_url}/message/SendTextMessage"
         await self._post(session, url, payload)
 
+    async def _send_emoji(self, session: aiohttp.ClientSession, comp: WechatEmoji):
+        payload = {
+            "EmojiList": [
+                {
+                    "EmojiMd5": comp.md5,
+                    "EmojiSize": comp.md5_len,
+                    "ToUserName": self.session_id,
+                }
+            ]
+        }
+        url = f"{self.adapter.base_url}/message/SendEmojiMessage"
+        await self._post(session, url, payload)
+
+    async def _send_voice(self, session: aiohttp.ClientSession, comp: Record):
+        record_path = await comp.convert_to_file_path()
+        # 默认已经存在 data/temp 中
+        b64, duration = await audio_to_tencent_silk_base64(record_path)
+        payload = {
+            "ToUserName": self.session_id,
+            "VoiceData": b64,
+            "VoiceFormat": 4,
+            "VoiceSecond": duration,
+        }
+        url = f"{self.adapter.base_url}/message/SendVoice"
+        await self._post(session, url, payload)
+
     @staticmethod
     def _validate_base64(b64: str) -> bytes:
         return base64.b64decode(b64, validate=True)

astrbot/core/platform/sources/wechatpadpro/xml_data_parser.py

@@ -0,0 +1,160 @@
+from defusedxml import ElementTree as eT
+from astrbot.api import logger
+from astrbot.api.message_components import (
+    WechatEmoji as Emoji,
+    Plain,
+    Image,
+    BaseMessageComponent,
+)
+
+
+class GeweDataParser:
+    def __init__(
+        self,
+        content: str,
+        is_private_chat: bool = False,
+        cached_texts=None,
+        cached_images=None,
+        raw_message: dict = None,
+        downloader=None,
+    ):
+        self._xml = None
+        self.content = content
+        self.is_private_chat = is_private_chat
+        self.cached_texts = cached_texts or {}
+        self.cached_images = cached_images or {}
+        self.downloader = downloader
+
+        raw_message = raw_message or {}
+        self.from_user_name = raw_message.get("from_user_name", {}).get("str", "")
+        self.to_user_name = raw_message.get("to_user_name", {}).get("str", "")
+        self.msg_id = raw_message.get("msg_id", "")
+
+    def _format_to_xml(self):
+        if self._xml:
+            return self._xml
+
+        try:
+            msg_str = self.content
+            if not self.is_private_chat:
+                parts = self.content.split(":\n", 1)
+                msg_str = parts[1] if len(parts) == 2 else self.content
+
+            self._xml = eT.fromstring(msg_str)
+            return self._xml
+        except Exception as e:
+            logger.error(f"[XML解析失败] {e}")
+            raise
+
+    async def parse_mutil_49(self) -> list[BaseMessageComponent] | None:
+        """
+        处理 msg_type == 49 的多种 appmsg 类型（目前支持 type==57）
+        """
+        try:
+            appmsg_type = self._format_to_xml().findtext(".//appmsg/type")
+            if appmsg_type == "57":
+                return await self.parse_reply()
+        except Exception as e:
+            logger.warning(f"[parse_mutil_49] 解析失败: {e}")
+        return None
+
+    async def parse_reply(self) -> list[BaseMessageComponent]:
+        """
+        处理 type == 57 的引用消息：支持文本（1）、图片（3）、嵌套49（49）
+        """
+        components = []
+
+        try:
+            appmsg = self._format_to_xml().find("appmsg")
+            if appmsg is None:
+                return [Plain("[引用消息解析失败]")]
+
+            refermsg = appmsg.find("refermsg")
+            if refermsg is None:
+                return [Plain("[引用消息解析失败]")]
+
+            quote_type = int(refermsg.findtext("type", "0"))
+            nickname = refermsg.findtext("displayname", "未知发送者")
+            quote_content = refermsg.findtext("content", "")
+            svrid = refermsg.findtext("svrid")
+
+            match quote_type:
+                case 1:  # 文本引用
+                    quoted_text = self.cached_texts.get(str(svrid), quote_content)
+                    components.append(Plain(f"[引用] {nickname}: {quoted_text}"))
+
+                case 3:  # 图片引用
+                    quoted_image_b64 = self.cached_images.get(str(svrid))
+                    if not quoted_image_b64:
+                        try:
+                            quote_xml = eT.fromstring(quote_content)
+                            img = quote_xml.find("img")
+                            cdn_url = (
+                                img.get("cdnbigimgurl") or img.get("cdnmidimgurl")
+                                if img is not None
+                                else None
+                            )
+                            if cdn_url and self.downloader:
+                                image_resp = await self.downloader(
+                                    self.from_user_name, self.to_user_name, self.msg_id
+                                )
+                                quoted_image_b64 = (
+                                    image_resp.get("Data", {})
+                                    .get("Data", {})
+                                    .get("Buffer")
+                                )
+                        except Exception as e:
+                            logger.warning(f"[引用图片解析失败] svrid={svrid} err={e}")
+
+                    if quoted_image_b64:
+                        components.extend(
+                            [
+                                Image.fromBase64(quoted_image_b64),
+                                Plain(f"[引用] {nickname}: [引用的图片]"),
+                            ]
+                        )
+                    else:
+                        components.append(
+                            Plain(f"[引用] {nickname}: [引用的图片 - 未能获取]")
+                        )
+
+                case 49:  # 嵌套引用
+                    try:
+                        nested_root = eT.fromstring(quote_content)
+                        nested_title = nested_root.findtext(".//appmsg/title", "")
+                        components.append(Plain(f"[引用] {nickname}: {nested_title}"))
+                    except Exception as e:
+                        logger.warning(f"[嵌套引用解析失败] err={e}")
+                        components.append(Plain(f"[引用] {nickname}: [嵌套引用消息]"))
+
+                case _:  # 其他未识别类型
+                    logger.info(f"[未知引用类型] quote_type={quote_type}")
+                    components.append(Plain(f"[引用] {nickname}: [不支持的引用类型]"))
+
+            # 主消息标题
+            title = appmsg.findtext("title", "")
+            if title:
+                components.append(Plain(title))
+
+        except Exception as e:
+            logger.error(f"[parse_reply] 总体解析失败: {e}")
+            return [Plain("[引用消息解析失败]")]
+
+        return components
+
+    def parse_emoji(self) -> Emoji | None:
+        """
+        处理 msg_type == 47 的表情消息（emoji）
+        """
+        try:
+            emoji_element = self._format_to_xml().find(".//emoji")
+            if emoji_element is not None:
+                return Emoji(
+                    md5=emoji_element.get("md5"),
+                    md5_len=emoji_element.get("len"),
+                    cdnurl=emoji_element.get("cdnurl"),
+                )
+        except Exception as e:
+            logger.error(f"[parse_emoji] 解析失败: {e}")
+
+        return None

astrbot/core/platform/sources/wecom/wecom_adapter.py

@@ -303,6 +303,7 @@ class WecomPlatformAdapter(Platform):
         abm.session_id = external_userid
         abm.type = MessageType.FRIEND_MESSAGE
         abm.message_id = msg.get("msgid", uuid.uuid4().hex[:8])
+        abm.message_str = ""
         if msgtype == "text":
             text = msg.get("text", {}).get("content", "").strip()
             abm.message = [Plain(text=text)]
@@ -316,7 +317,29 @@ class WecomPlatformAdapter(Platform):
             with open(path, "wb") as f:
                 f.write(resp.content)
             abm.message = [Image(file=path, url=path)]
-            abm.message_str = "[图片]"
+        elif msgtype == "voice":
+            media_id = msg.get("voice", {}).get("media_id", "")
+            resp: Response = await asyncio.get_event_loop().run_in_executor(
+                None, self.client.media.download, media_id
+            )
+
+            temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+            path = os.path.join(temp_dir, f"weixinkefu_{media_id}.amr")
+            with open(path, "wb") as f:
+                f.write(resp.content)
+
+            try:
+                from pydub import AudioSegment
+
+                path_wav = os.path.join(temp_dir, f"weixinkefu_{media_id}.wav")
+                audio = AudioSegment.from_file(path)
+                audio.export(path_wav, format="wav")
+            except Exception as e:
+                logger.error(f"转换音频失败: {e}。如果没有安装 ffmpeg 请先安装。")
+                path_wav = path
+                return
+
+            abm.message = [Record(file=path_wav, url=path_wav)]
         else:
             logger.warning(f"未实现的微信客服消息事件: {msg}")
             return

astrbot/core/platform/sources/wecom/wecom_event.py

@@ -120,6 +120,30 @@ class WecomPlatformEvent(AstrMessageEvent):
                             self.get_self_id(),
                             response["media_id"],
                         )
+                elif isinstance(comp, Record):
+                    record_path = await comp.convert_to_file_path()
+                    # 转成amr
+                    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+                    record_path_amr = os.path.join(temp_dir, f"{uuid.uuid4()}.amr")
+                    pydub.AudioSegment.from_wav(record_path).export(
+                        record_path_amr, format="amr"
+                    )
+
+                    with open(record_path_amr, "rb") as f:
+                        try:
+                            response = self.client.media.upload("voice", f)
+                        except Exception as e:
+                            logger.error(f"微信客服上传语音失败: {e}")
+                            await self.send(
+                                MessageChain().message(f"微信客服上传语音失败: {e}")
+                            )
+                            return
+                        logger.info(f"微信客服上传语音返回: {response}")
+                        kf_message_api.send_voice(
+                            user_id,
+                            self.get_self_id(),
+                            response["media_id"],
+                        )
                 else:
                     logger.warning(f"还没实现这个消息类型的发送逻辑: {comp.type}。")
         else:

astrbot/core/platform/sources/weixin_official_account/weixin_offacc_adapter.py

@@ -20,7 +20,7 @@ from requests import Response
 from wechatpy.utils import check_signature
 from wechatpy.crypto import WeChatCrypto
 from wechatpy import WeChatClient
-from wechatpy.messages import TextMessage, ImageMessage, VoiceMessage
+from wechatpy.messages import TextMessage, ImageMessage, VoiceMessage, BaseMessage
 from wechatpy.exceptions import InvalidSignatureException
 from wechatpy import parse_message
 from .weixin_offacc_event import WeixinOfficialAccountPlatformEvent
@@ -87,7 +87,11 @@ class WecomServer:
             logger.info(f"解析成功: {msg}")
 
             if self.callback:
-                await self.callback(msg)
+                result_xml = await self.callback(msg)
+                if not result_xml:
+                    return "success"
+                if isinstance(result_xml, str):
+                    return result_xml
 
         return "success"
 
@@ -117,6 +121,7 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
         self.api_base_url = platform_config.get(
             "api_base_url", "https://api.weixin.qq.com/cgi-bin/"
         )
+        self.active_send_mode = self.config.get("active_send_mode", False)
 
         if not self.api_base_url:
             self.api_base_url = "https://api.weixin.qq.com/cgi-bin/"
@@ -138,9 +143,29 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
 
         self.client.API_BASE_URL = self.api_base_url
 
-        async def callback(msg):
+        # 微信公众号必须 5 秒内进行回复，否则会重试 3 次，我们需要对其进行消息排重
+        # msgid -> Future
+        self.wexin_event_workers: dict[str, asyncio.Future] = {}
+
+        async def callback(msg: BaseMessage):
             try:
-                await self.convert_message(msg)
+                if self.active_send_mode:
+                    await self.convert_message(msg, None)
+                else:
+                    if msg.id in self.wexin_event_workers:
+                        future = self.wexin_event_workers[msg.id]
+                        logger.debug(f"duplicate message id checked: {msg.id}")
+                    else:
+                        future = asyncio.get_event_loop().create_future()
+                        self.wexin_event_workers[msg.id] = future
+                        await self.convert_message(msg, future)
+                    # I love shield so much!
+                    result = await asyncio.wait_for(asyncio.shield(future), 60)  # wait for 60s
+                    logger.debug(f"Got future result: {result}")
+                    self.wexin_event_workers.pop(msg.id, None)
+                    return result  # xml. see weixin_offacc_event.py
+            except asyncio.TimeoutError:
+                pass
             except Exception as e:
                 logger.error(f"转换消息时出现异常: {e}")
 
@@ -163,7 +188,9 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
     async def run(self):
         await self.server.start_polling()
 
-    async def convert_message(self, msg) -> AstrBotMessage | None:
+    async def convert_message(
+        self, msg, future: asyncio.Future = None
+    ) -> AstrBotMessage | None:
         abm = AstrBotMessage()
         if isinstance(msg, TextMessage):
             abm.message_str = msg.content
@@ -177,7 +204,6 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
             abm.message_id = msg.id
             abm.timestamp = msg.time
             abm.session_id = abm.sender.user_id
-            abm.raw_message = msg
         elif msg.type == "image":
             assert isinstance(msg, ImageMessage)
             abm.message_str = "[图片]"
@@ -191,7 +217,6 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
             abm.message_id = msg.id
             abm.timestamp = msg.time
             abm.session_id = abm.sender.user_id
-            abm.raw_message = msg
         elif msg.type == "voice":
             assert isinstance(msg, VoiceMessage)
 
@@ -209,7 +234,9 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
                 audio = AudioSegment.from_file(path)
                 audio.export(path_wav, format="wav")
             except Exception as e:
-                logger.error(f"转换音频失败: {e}。如果没有安装 pydub 和 ffmpeg 请先安装。")
+                logger.error(
+                    f"转换音频失败: {e}。如果没有安装 pydub 和 ffmpeg 请先安装。"
+                )
                 path_wav = path
                 return
 
@@ -224,11 +251,16 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
             abm.message_id = msg.id
             abm.timestamp = msg.time
             abm.session_id = abm.sender.user_id
-            abm.raw_message = msg
         else:
             logger.warning(f"暂未实现的事件: {msg.type}")
+            future.set_result(None)
             return
-
+        # 很不优雅 :(
+        abm.raw_message = {
+            "message": msg,
+            "future": future,
+            "active_send_mode": self.active_send_mode,
+        }
         logger.info(f"abm: {abm}")
         await self.handle_msg(abm)
 

astrbot/core/platform/sources/weixin_official_account/weixin_offacc_event.py

@@ -4,6 +4,8 @@ from astrbot.api.event import AstrMessageEvent, MessageChain
 from astrbot.api.platform import AstrBotMessage, PlatformMetadata
 from astrbot.api.message_components import Plain, Image, Record
 from wechatpy import WeChatClient
+from wechatpy.replies import TextReply, ImageReply, VoiceReply
+
 
 from astrbot.api import logger
 
@@ -82,12 +84,23 @@ class WeixinOfficialAccountPlatformEvent(AstrMessageEvent):
 
     async def send(self, message: MessageChain):
         message_obj = self.message_obj
+        active_send_mode = message_obj.raw_message.get("active_send_mode", False)
         for comp in message.chain:
             if isinstance(comp, Plain):
                 # Split long text messages if needed
                 plain_chunks = await self.split_plain(comp.text)
                 for chunk in plain_chunks:
-                    self.client.message.send_text(message_obj.sender.user_id, chunk)
+                    if active_send_mode:
+                        self.client.message.send_text(message_obj.sender.user_id, chunk)
+                    else:
+                        reply = TextReply(
+                            content=chunk,
+                            message=self.message_obj.raw_message["message"],
+                        )
+                        xml = reply.render()
+                        future = self.message_obj.raw_message["future"]
+                        assert isinstance(future, asyncio.Future)
+                        future.set_result(xml)
                     await asyncio.sleep(0.5)  # Avoid sending too fast
             elif isinstance(comp, Image):
                 img_path = await comp.convert_to_file_path()
@@ -102,10 +115,22 @@ class WeixinOfficialAccountPlatformEvent(AstrMessageEvent):
                         )
                         return
                     logger.debug(f"微信公众平台上传图片返回: {response}")
-                    self.client.message.send_image(
-                        message_obj.sender.user_id,
-                        response["media_id"],
-                    )
+
+                    if active_send_mode:
+                        self.client.message.send_image(
+                            message_obj.sender.user_id,
+                            response["media_id"],
+                        )
+                    else:
+                        reply = ImageReply(
+                            media_id=response["media_id"],
+                            message=self.message_obj.raw_message["message"],
+                        )
+                        xml = reply.render()
+                        future = self.message_obj.raw_message["future"]
+                        assert isinstance(future, asyncio.Future)
+                        future.set_result(xml)
+
             elif isinstance(comp, Record):
                 record_path = await comp.convert_to_file_path()
                 # 转成amr
@@ -124,10 +149,23 @@ class WeixinOfficialAccountPlatformEvent(AstrMessageEvent):
                         )
                         return
                     logger.info(f"微信公众平台上传语音返回: {response}")
-                    self.client.message.send_voice(
-                        message_obj.sender.user_id,
-                        response["media_id"],
-                    )
+
+
+                    if active_send_mode:
+                        self.client.message.send_voice(
+                            message_obj.sender.user_id,
+                            response["media_id"],
+                        )
+                    else:
+                        reply = VoiceReply(
+                            media_id=response["media_id"],
+                            message=self.message_obj.raw_message["message"],
+                        )
+                        xml = reply.render()
+                        future = self.message_obj.raw_message["future"]
+                        assert isinstance(future, asyncio.Future)
+                        future.set_result(xml)
+
             else:
                 logger.warning(f"还没实现这个消息类型的发送逻辑: {comp.type}。")
 

astrbot/core/provider/entities.py

@@ -19,6 +19,7 @@ class ProviderType(enum.Enum):
     CHAT_COMPLETION = "chat_completion"
     SPEECH_TO_TEXT = "speech_to_text"
     TEXT_TO_SPEECH = "text_to_speech"
+    EMBEDDING = "embedding"
 
 
 @dataclass
@@ -57,7 +58,7 @@ class AssistantMessageSegment:
     """OpenAI 格式的上下文中 role 为 assistant 的消息段。参考: https://platform.openai.com/docs/guides/function-calling"""
 
     content: str = None
-    tool_calls: List[ChatCompletionMessageToolCall | Dict] = None
+    tool_calls: List[ChatCompletionMessageToolCall | Dict] = field(default_factory=list)
     role: str = "assistant"
 
     def to_dict(self):
@@ -66,7 +67,7 @@ class AssistantMessageSegment:
         }
         if self.content:
             ret["content"] = self.content
-        elif self.tool_calls:
+        if self.tool_calls:
             ret["tool_calls"] = self.tool_calls
         return ret
 
@@ -94,19 +95,19 @@ class ProviderRequest:
     """提示词"""
     session_id: str = ""
     """会话 ID"""
-    image_urls: List[str] = None
+    image_urls: list[str] = field(default_factory=list)
     """图片 URL 列表"""
-    func_tool: FuncCall = None
+    func_tool: FuncCall | None = None
     """可用的函数工具"""
-    contexts: List = None
+    contexts: list[dict] = field(default_factory=list)
     """上下文。格式与 openai 的上下文格式一致：
     参考 https://platform.openai.com/docs/api-reference/chat/create#chat-create-messages
     """
     system_prompt: str = ""
     """系统提示词"""
-    conversation: Conversation = None
+    conversation: Conversation | None = None
 
-    tool_calls_result: ToolCallsResult = None
+    tool_calls_result: list[ToolCallsResult] | ToolCallsResult | None = None
     """附加的上次请求后工具调用的结果。参考: https://platform.openai.com/docs/guides/function-calling#handling-function-calls"""
 
     def __repr__(self):
@@ -115,6 +116,14 @@ class ProviderRequest:
     def __str__(self):
         return self.__repr__()
 
+    def append_tool_calls_result(self, tool_calls_result: ToolCallsResult):
+        """添加工具调用结果到请求中"""
+        if not self.tool_calls_result:
+            self.tool_calls_result = []
+        if isinstance(self.tool_calls_result, ToolCallsResult):
+            self.tool_calls_result = [self.tool_calls_result]
+        self.tool_calls_result.append(tool_calls_result)
+
     def _print_friendly_context(self):
         """打印友好的消息上下文。将 image_url 的值替换为 <Image>"""
         if not self.contexts:
@@ -155,7 +164,9 @@ class ProviderRequest:
         if self.image_urls:
             user_content = {
                 "role": "user",
-                "content": [{"type": "text", "text": self.prompt if self.prompt else "[图片]"}],
+                "content": [
+                    {"type": "text", "text": self.prompt if self.prompt else "[图片]"}
+                ],
             }
             for image_url in self.image_urls:
                 if image_url.startswith("http"):

astrbot/core/provider/func_tool_manager.py

@@ -4,6 +4,7 @@ import textwrap
 import os
 import asyncio
 import logging
+from datetime import timedelta
 
 from typing import Dict, List, Awaitable, Literal, Any
 from dataclasses import dataclass
@@ -20,6 +21,13 @@ try:
 except (ModuleNotFoundError, ImportError):
     logger.warning("警告: 缺少依赖库 'mcp'，将无法使用 MCP 服务。")
 
+try:
+    from mcp.client.streamable_http import streamablehttp_client
+except (ModuleNotFoundError, ImportError):
+    logger.warning(
+        "警告: 缺少依赖库 'mcp' 或者 mcp 库版本过低，无法使用 Streamable HTTP 连接方式。"
+    )
+
 DEFAULT_MCP_CONFIG = {"mcpServers": {}}
 
 SUPPORTED_TYPES = [
@@ -96,7 +104,10 @@ class MCPClient:
     async def connect_to_server(self, mcp_server_config: dict, name: str):
         """连接到 MCP 服务器
 
-        如果 `url` 参数存在，则使用 SSE 的方式连接到 MCP 服务。
+        如果 `url` 参数存在：
+            1. 当 transport 指定为 `streamable_http` 时，使用 Streamable HTTP 连接方式。
+            1. 当 transport 指定为 `sse` 时，使用 SSE 连接方式。
+            2. 如果没有指定，默认使用 SSE 的方式连接到 MCP 服务。
 
         Args:
             mcp_server_config (dict): Configuration for the MCP server. See https://modelcontextprotocol.io/quickstart/server
@@ -108,15 +119,41 @@ class MCPClient:
         cfg.pop("active", None)  # Remove active flag from config
 
         if "url" in cfg:
-            # SSE transport method
-            self._streams_context = sse_client(url=cfg["url"])
-            streams = await self._streams_context.__aenter__()
+            is_sse = True
+            if cfg.get("transport") == "streamable_http":
+                is_sse = False
+            if is_sse:
+                # SSE transport method
+                self._streams_context = sse_client(
+                    url=cfg["url"],
+                    headers=cfg.get("headers", {}),
+                    timeout=cfg.get("timeout", 5),
+                    sse_read_timeout=cfg.get("sse_read_timeout", 60 * 5),
+                )
+                streams = await self._streams_context.__aenter__()
 
-            # Create a new client session
-            # self.session = await self._session_context.__aenter__()
-            self.session = await self.exit_stack.enter_async_context(
-                mcp.ClientSession(*streams)
-            )
+                # Create a new client session
+                self.session = await self.exit_stack.enter_async_context(
+                    mcp.ClientSession(*streams)
+                )
+            else:
+                timeout = timedelta(seconds=cfg.get("timeout", 30))
+                sse_read_timeout = timedelta(
+                    seconds=cfg.get("sse_read_timeout", 60 * 5)
+                )
+                self._streams_context = streamablehttp_client(
+                    url=cfg["url"],
+                    headers=cfg.get("headers", {}),
+                    timeout=timeout,
+                    sse_read_timeout=sse_read_timeout,
+                    terminate_on_close=cfg.get("terminate_on_close", True),
+                )
+                read_s, write_s, _ = await self._streams_context.__aenter__()
+
+                # Create a new client session
+                self.session = await self.exit_stack.enter_async_context(
+                    mcp.ClientSession(read_stream=read_s, write_stream=write_s)
+                )
 
         else:
             server_params = mcp.StdioServerParameters(

astrbot/core/provider/manager.py

@@ -1,12 +1,14 @@
-import traceback
 import asyncio
-from astrbot.core.config.astrbot_config import AstrBotConfig
-from .provider import Provider, STTProvider, TTSProvider, Personality
-from .entities import ProviderType
+import traceback
 from typing import List
-from astrbot.core.db import BaseDatabase
-from .register import provider_cls_map, llm_tools
+
 from astrbot.core import logger, sp
+from astrbot.core.config.astrbot_config import AstrBotConfig
+from astrbot.core.db import BaseDatabase
+
+from .entities import ProviderType
+from .provider import Personality, Provider, STTProvider, TTSProvider
+from .register import llm_tools, provider_cls_map
 
 
 class ProviderManager:
@@ -18,13 +20,6 @@ class ProviderManager:
         self.persona_configs: list = config.get("persona", [])
         self.astrbot_config = config
 
-        self.selected_provider_id = sp.get("curr_provider")
-        self.selected_stt_provider_id = self.provider_stt_settings.get("provider_id")
-        self.selected_tts_provider_id = self.provider_settings.get("provider_id")
-        self.provider_enabled = self.provider_settings.get("enable", False)
-        self.stt_enabled = self.provider_stt_settings.get("enable", False)
-        self.tts_enabled = self.provider_tts_settings.get("enable", False)
-
         # 人格情景管理
         # 目前没有拆成独立的模块
         self.default_persona_name = self.provider_settings.get(
@@ -45,13 +40,11 @@ class ProviderManager:
                     begin_dialogs = []
                 user_turn = True
                 for dialog in begin_dialogs:
-                    bd_processed.append(
-                        {
-                            "role": "user" if user_turn else "assistant",
-                            "content": dialog,
-                            "_no_save": None,  # 不持久化到 db
-                        }
-                    )
+                    bd_processed.append({
+                        "role": "user" if user_turn else "assistant",
+                        "content": dialog,
+                        "_no_save": None,  # 不持久化到 db
+                    })
                     user_turn = not user_turn
             if mood_imitation_dialogs:
                 if len(mood_imitation_dialogs) % 2 != 0:
@@ -98,15 +91,18 @@ class ProviderManager:
         """加载的 Speech To Text Provider 的实例"""
         self.tts_provider_insts: List[TTSProvider] = []
         """加载的 Text To Speech Provider 的实例"""
+        self.embedding_provider_insts: List[Provider] = []
+        """加载的 Embedding Provider 的实例"""
         self.inst_map = {}
         """Provider 实例映射. key: provider_id, value: Provider 实例"""
         self.llm_tools = llm_tools
+
         self.curr_provider_inst: Provider = None
-        """当前使用的 Provider 实例"""
+        """默认的 Provider 实例"""
         self.curr_stt_provider_inst: STTProvider = None
-        """当前使用的 Speech To Text Provider 实例"""
+        """默认的 Speech To Text Provider 实例"""
         self.curr_tts_provider_inst: TTSProvider = None
-        """当前使用的 Text To Speech Provider 实例"""
+        """默认的 Text To Speech Provider 实例"""
         self.db_helper = db_helper
 
         # kdb(experimental)
@@ -115,18 +111,57 @@ class ProviderManager:
         if kdb_cfg and len(kdb_cfg):
             self.curr_kdb_name = list(kdb_cfg.keys())[0]
 
+    async def set_provider(
+        self, provider_id: str, provider_type: ProviderType, umo: str = None
+    ):
+        """设置提供商。
+
+        Args:
+            provider_id (str): 提供商 ID。
+            provider_type (ProviderType): 提供商类型。
+            umo (str, optional): 用户会话 ID，用于提供商会话隔离。当用户启用了提供商会话隔离时此参数才生效。
+        """
+        if provider_id not in self.inst_map:
+            raise ValueError(f"提供商 {provider_id} 不存在，无法设置。")
+        if umo and self.provider_settings["separate_provider"]:
+            perf = sp.get("session_provider_perf", {})
+            session_perf = perf.get(umo, {})
+            session_perf[provider_type.value] = provider_id
+            perf[umo] = session_perf
+            sp.put("session_provider_perf", perf)
+            return
+        # 不启用提供商会话隔离模式的情况
+        self.curr_provider_inst = self.inst_map[provider_id]
+        if provider_type == ProviderType.TEXT_TO_SPEECH:
+            sp.put("curr_provider_tts", provider_id)
+        elif provider_type == ProviderType.SPEECH_TO_TEXT:
+            sp.put("curr_provider_stt", provider_id)
+        elif provider_type == ProviderType.CHAT_COMPLETION:
+            sp.put("curr_provider", provider_id)
+
     async def initialize(self):
+        # 逐个初始化提供商
         for provider_config in self.providers_config:
             await self.load_provider(provider_config)
 
-        if not self.curr_provider_inst:
-            logger.warning("未启用任何用于 文本生成 的提供商适配器。")
+        # 设置默认提供商
+        self.curr_provider_inst = self.inst_map.get(
+            self.provider_settings.get("default_provider_id")
+        )
+        if not self.curr_provider_inst and self.provider_insts:
+            self.curr_provider_inst = self.provider_insts[0]
 
-        if self.stt_enabled and not self.curr_stt_provider_inst:
-            logger.warning("未启用任何用于 语音转文本 的提供商适配器。")
+        self.curr_stt_provider_inst = self.inst_map.get(
+            self.provider_stt_settings.get("provider_id")
+        )
+        if not self.curr_stt_provider_inst and self.stt_provider_insts:
+            self.curr_stt_provider_inst = self.stt_provider_insts[0]
 
-        if self.tts_enabled and not self.curr_tts_provider_inst:
-            logger.warning("未启用任何用于 文本转语音 的提供商适配器。")
+        self.curr_tts_provider_inst = self.inst_map.get(
+            self.provider_tts_settings.get("provider_id")
+        )
+        if not self.curr_tts_provider_inst and self.tts_provider_insts:
+            self.curr_tts_provider_inst = self.tts_provider_insts[0]
 
         # 初始化 MCP Client 连接
         asyncio.create_task(
@@ -155,11 +190,6 @@ class ProviderManager:
                     from .sources.anthropic_source import (
                         ProviderAnthropic as ProviderAnthropic,
                     )
-                case "llm_tuner":
-                    logger.info("加载 LLM Tuner 工具 ...")
-                    from .sources.llmtuner_source import (
-                        LLMTunerModelLoader as LLMTunerModelLoader,
-                    )
                 case "dify":
                     from .sources.dify_source import ProviderDify as ProviderDify
                 case "dashscope":
@@ -190,6 +220,10 @@ class ProviderManager:
                     from .sources.edge_tts_source import (
                         ProviderEdgeTTS as ProviderEdgeTTS,
                     )
+                case "gsv_tts_selfhost":
+                    from .sources.gsv_selfhosted_source import (
+                        ProviderGSVTTS as ProviderGSVTTS,
+                    )
                 case "gsvi_tts_api":
                     from .sources.gsvi_tts_source import (
                         ProviderGSVITTS as ProviderGSVITTS,
@@ -210,6 +244,22 @@ class ProviderManager:
                     from .sources.minimax_tts_api_source import (
                         ProviderMiniMaxTTSAPI as ProviderMiniMaxTTSAPI,
                     )
+                case "volcengine_tts":
+                    from .sources.volcengine_tts import (
+                        ProviderVolcengineTTS as ProviderVolcengineTTS,
+                    )
+                case "gemini_tts":
+                    from .sources.gemini_tts_source import (
+                        ProviderGeminiTTSAPI as ProviderGeminiTTSAPI,
+                    )
+                case "openai_embedding":
+                    from .sources.openai_embedding_source import (
+                        OpenAIEmbeddingProvider as OpenAIEmbeddingProvider,
+                    )
+                case "gemini_embedding":
+                    from .sources.gemini_embedding_source import (
+                        GeminiEmbeddingProvider as GeminiEmbeddingProvider,
+                    )
         except (ImportError, ModuleNotFoundError) as e:
             logger.critical(
                 f"加载 {provider_config['type']}({provider_config['id']}) 提供商适配器失败：{e}。可能是因为有未安装的依赖。"
@@ -242,14 +292,14 @@ class ProviderManager:
 
                 self.stt_provider_insts.append(inst)
                 if (
-                    self.selected_stt_provider_id == provider_config["id"]
-                    and self.stt_enabled
+                    self.provider_stt_settings.get("provider_id")
+                    == provider_config["id"]
                 ):
                     self.curr_stt_provider_inst = inst
                     logger.info(
                         f"已选择 {provider_config['type']}({provider_config['id']}) 作为当前语音转文本提供商适配器。"
                     )
-                if not self.curr_stt_provider_inst and self.stt_enabled:
+                if not self.curr_stt_provider_inst:
                     self.curr_stt_provider_inst = inst
 
             elif provider_metadata.provider_type == ProviderType.TEXT_TO_SPEECH:
@@ -262,15 +312,12 @@ class ProviderManager:
                     await inst.initialize()
 
                 self.tts_provider_insts.append(inst)
-                if (
-                    self.selected_tts_provider_id == provider_config["id"]
-                    and self.tts_enabled
-                ):
+                if self.provider_settings.get("provider_id") == provider_config["id"]:
                     self.curr_tts_provider_inst = inst
                     logger.info(
                         f"已选择 {provider_config['type']}({provider_config['id']}) 作为当前文本转语音提供商适配器。"
                     )
-                if not self.curr_tts_provider_inst and self.tts_enabled:
+                if not self.curr_tts_provider_inst:
                     self.curr_tts_provider_inst = inst
 
             elif provider_metadata.provider_type == ProviderType.CHAT_COMPLETION:
@@ -278,8 +325,6 @@ class ProviderManager:
                 inst = provider_metadata.cls_type(
                     provider_config,
                     self.provider_settings,
-                    self.db_helper,
-                    self.provider_settings.get("persistant_history", True),
                     self.selected_default_persona,
                 )
 
@@ -288,16 +333,24 @@ class ProviderManager:
 
                 self.provider_insts.append(inst)
                 if (
-                    self.selected_provider_id == provider_config["id"]
-                    and self.provider_enabled
+                    self.provider_settings.get("default_provider_id")
+                    == provider_config["id"]
                 ):
                     self.curr_provider_inst = inst
                     logger.info(
                         f"已选择 {provider_config['type']}({provider_config['id']}) 作为当前提供商适配器。"
                     )
-                if not self.curr_provider_inst and self.provider_enabled:
+                if not self.curr_provider_inst:
                     self.curr_provider_inst = inst
 
+            elif provider_metadata.provider_type == ProviderType.EMBEDDING:
+                inst = provider_metadata.cls_type(
+                    provider_config, self.provider_settings
+                )
+                if getattr(inst, "initialize", None):
+                    await inst.initialize()
+                self.embedding_provider_insts.append(inst)
+
             self.inst_map[provider_config["id"]] = inst
         except Exception as e:
             logger.error(traceback.format_exc())
@@ -318,39 +371,24 @@ class ProviderManager:
 
         if len(self.provider_insts) == 0:
             self.curr_provider_inst = None
-        elif (
-            self.curr_provider_inst is None
-            and len(self.provider_insts) > 0
-            and self.provider_enabled
-        ):
+        elif self.curr_provider_inst is None and len(self.provider_insts) > 0:
             self.curr_provider_inst = self.provider_insts[0]
-            self.selected_provider_id = self.curr_provider_inst.meta().id
             logger.info(
                 f"自动选择 {self.curr_provider_inst.meta().id} 作为当前提供商适配器。"
             )
 
         if len(self.stt_provider_insts) == 0:
             self.curr_stt_provider_inst = None
-        elif (
-            self.curr_stt_provider_inst is None
-            and len(self.stt_provider_insts) > 0
-            and self.stt_enabled
-        ):
+        elif self.curr_stt_provider_inst is None and len(self.stt_provider_insts) > 0:
             self.curr_stt_provider_inst = self.stt_provider_insts[0]
-            self.selected_stt_provider_id = self.curr_stt_provider_inst.meta().id
             logger.info(
                 f"自动选择 {self.curr_stt_provider_inst.meta().id} 作为当前语音转文本提供商适配器。"
             )
 
         if len(self.tts_provider_insts) == 0:
             self.curr_tts_provider_inst = None
-        elif (
-            self.curr_tts_provider_inst is None
-            and len(self.tts_provider_insts) > 0
-            and self.tts_enabled
-        ):
+        elif self.curr_tts_provider_inst is None and len(self.tts_provider_insts) > 0:
             self.curr_tts_provider_inst = self.tts_provider_insts[0]
-            self.selected_tts_provider_id = self.curr_tts_provider_inst.meta().id
             logger.info(
                 f"自动选择 {self.curr_tts_provider_inst.meta().id} 作为当前文本转语音提供商适配器。"
             )

astrbot/core/provider/provider.py

@@ -1,6 +1,5 @@
 import abc
 from typing import List
-from astrbot.core.db import BaseDatabase
 from typing import TypedDict, AsyncGenerator
 from astrbot.core.provider.func_tool_manager import FuncCall
 from astrbot.core.provider.entities import LLMResponse, ToolCallsResult
@@ -53,15 +52,13 @@ class Provider(AbstractProvider):
         self,
         provider_config: dict,
         provider_settings: dict,
-        persistant_history: bool = True,
-        db_helper: BaseDatabase = None,
-        default_persona: Personality = None,
+        default_persona: Personality | None = None,
     ) -> None:
         super().__init__(provider_config)
 
         self.provider_settings = provider_settings
 
-        self.curr_personality: Personality = default_persona
+        self.curr_personality = default_persona
         """维护了当前的使用的 persona，即人格。可能为 None"""
 
     @abc.abstractmethod
@@ -86,11 +83,11 @@ class Provider(AbstractProvider):
         self,
         prompt: str,
         session_id: str = None,
-        image_urls: List[str] = None,
+        image_urls: list[str] = None,
         func_tool: FuncCall = None,
-        contexts: List = None,
+        contexts: list = None,
         system_prompt: str = None,
-        tool_calls_result: ToolCallsResult = None,
+        tool_calls_result: ToolCallsResult | list[ToolCallsResult] = None,
         **kwargs,
     ) -> LLMResponse:
         """获得 LLM 的文本对话结果。会使用当前的模型进行对话。
@@ -114,11 +111,11 @@ class Provider(AbstractProvider):
         self,
         prompt: str,
         session_id: str = None,
-        image_urls: List[str] = None,
+        image_urls: list[str] = None,
         func_tool: FuncCall = None,
-        contexts: List = None,
+        contexts: list = None,
         system_prompt: str = None,
-        tool_calls_result: ToolCallsResult = None,
+        tool_calls_result: ToolCallsResult | list[ToolCallsResult] = None,
         **kwargs,
     ) -> AsyncGenerator[LLMResponse, None]:
         """获得 LLM 的流式文本对话结果。会使用当前的模型进行对话。在生成的最后会返回一次完整的结果。
@@ -179,3 +176,25 @@ class TTSProvider(AbstractProvider):
     async def get_audio(self, text: str) -> str:
         """获取文本的音频，返回音频文件路径"""
         raise NotImplementedError()
+
+
+class EmbeddingProvider(AbstractProvider):
+    def __init__(self, provider_config: dict, provider_settings: dict) -> None:
+        super().__init__(provider_config)
+        self.provider_config = provider_config
+        self.provider_settings = provider_settings
+
+    @abc.abstractmethod
+    async def get_embedding(self, text: str) -> list[float]:
+        """获取文本的向量"""
+        ...
+
+    @abc.abstractmethod
+    async def get_embeddings(self, text: list[str]) -> list[list[float]]:
+        """批量获取文本的向量"""
+        ...
+
+    @abc.abstractmethod
+    def get_dim(self) -> int:
+        """获取向量的维度"""
+        ...

astrbot/core/provider/sources/anthropic_source.py

@@ -1,3 +1,6 @@
+import json
+import anthropic
+import base64
 from typing import List
 from mimetypes import guess_type
 
@@ -5,41 +8,33 @@ from anthropic import AsyncAnthropic
 from anthropic.types import Message
 
 from astrbot.core.utils.io import download_image_by_url
-from astrbot.core.db import BaseDatabase
-from astrbot.api.provider import Provider, Personality
+from astrbot.api.provider import Provider
 from astrbot import logger
 from astrbot.core.provider.func_tool_manager import FuncCall
 from ..register import register_provider_adapter
-from astrbot.core.message.message_event_result import MessageChain
-from astrbot.core.provider.entities import LLMResponse, ToolCallsResult
-from .openai_source import ProviderOpenAIOfficial
+from astrbot.core.provider.entities import LLMResponse
+from typing import AsyncGenerator
 
 
 @register_provider_adapter(
     "anthropic_chat_completion", "Anthropic Claude API 提供商适配器"
 )
-class ProviderAnthropic(ProviderOpenAIOfficial):
+class ProviderAnthropic(Provider):
     def __init__(
         self,
-        provider_config: dict,
-        provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=True,
-        default_persona: Personality = None,
+        provider_config,
+        provider_settings,
+        default_persona=None,
     ) -> None:
-        # Skip OpenAI's __init__ and call Provider's __init__ directly
-        Provider.__init__(
-            self,
+        super().__init__(
             provider_config,
             provider_settings,
-            persistant_history,
-            db_helper,
             default_persona,
         )
 
-        self.chosen_api_key = None
+        self.chosen_api_key: str = ""
         self.api_keys: List = provider_config.get("key", [])
-        self.chosen_api_key = self.api_keys[0] if len(self.api_keys) > 0 else None
+        self.chosen_api_key = self.api_keys[0] if len(self.api_keys) > 0 else ""
         self.base_url = provider_config.get("api_base", "https://api.anthropic.com")
         self.timeout = provider_config.get("timeout", 120)
         if isinstance(self.timeout, str):
@@ -51,10 +46,63 @@ class ProviderAnthropic(ProviderOpenAIOfficial):
 
         self.set_model(provider_config["model_config"]["model"])
 
+    def _prepare_payload(self, messages: list[dict]):
+        """准备 Anthropic API 的请求 payload
+
+        Args:
+            messages: OpenAI 格式的消息列表，包含用户输入和系统提示等信息
+        Returns:
+            system_prompt: 系统提示内容
+            new_messages: 处理后的消息列表，去除系统提示
+        """
+        system_prompt = ""
+        new_messages = []
+        for message in messages:
+            if message["role"] == "system":
+                system_prompt = message["content"]
+            elif message["role"] == "assistant":
+                blocks = []
+                if isinstance(message["content"], str):
+                    blocks.append({"type": "text", "text": message["content"]})
+                if "tool_calls" in message:
+                    for tool_call in message["tool_calls"]:
+                        blocks.append(  # noqa: PERF401
+                            {
+                                "type": "tool_use",
+                                "name": tool_call["function"]["name"],
+                                "input": json.loads(tool_call["function"]["arguments"])
+                                if isinstance(tool_call["function"]["arguments"], str)
+                                else tool_call["function"]["arguments"],
+                                "id": tool_call["id"],
+                            }
+                        )
+                new_messages.append(
+                    {
+                        "role": "assistant",
+                        "content": blocks,
+                    }
+                )
+            elif message["role"] == "tool":
+                new_messages.append(
+                    {
+                        "role": "user",
+                        "content": [
+                            {
+                                "type": "tool_result",
+                                "tool_use_id": message["tool_call_id"],
+                                "content": message["content"],
+                            }
+                        ],
+                    }
+                )
+            else:
+                new_messages.append(message)
+
+        return system_prompt, new_messages
+
     async def _query(self, payloads: dict, tools: FuncCall) -> LLMResponse:
         if tools:
-            tool_list = tools.get_func_desc_anthropic_style()
-            if tool_list:
+            if tool_list := tools.get_func_desc_anthropic_style():
                 payloads["tools"] = tool_list
 
         completion = await self.client.messages.create(**payloads, stream=False)
@@ -64,68 +112,157 @@ class ProviderAnthropic(ProviderOpenAIOfficial):
 
         if len(completion.content) == 0:
             raise Exception("API 返回的 completion 为空。")
-        # TODO: 如果进行函数调用，思维链被截断，用户可能需要思维链的内容
-        # 选最后一条消息，如果要进行函数调用，anthropic会先返回文本消息的思维链，然后再返回函数调用请求
-        content = completion.content[-1]
 
-        llm_response = LLMResponse("assistant")
+        llm_response = LLMResponse(role="assistant")
 
-        if content.type == "text":
-            # text completion
-            completion_text = str(content.text).strip()
-            # llm_response.completion_text = completion_text
-            llm_response.result_chain = MessageChain().message(completion_text)
-
-        # Anthropic每次只返回一个函数调用
-        if completion.stop_reason == "tool_use":
-            # tools call (function calling)
-            args_ls = []
-            func_name_ls = []
-            tool_use_ids = []
-            func_name_ls.append(content.name)
-            args_ls.append(content.input)
-            tool_use_ids.append(content.id)
-            llm_response.role = "tool"
-            llm_response.tools_call_args = args_ls
-            llm_response.tools_call_name = func_name_ls
-            llm_response.tools_call_ids = tool_use_ids
+        for content_block in completion.content:
+            if content_block.type == "text":
+                completion_text = str(content_block.text).strip()
+                llm_response.completion_text = completion_text
 
+            if content_block.type == "tool_use":
+                llm_response.tools_call_args.append(content_block.input)
+                llm_response.tools_call_name.append(content_block.name)
+                llm_response.tools_call_ids.append(content_block.id)
+        # TODO(Soulter): 处理 end_turn 情况
         if not llm_response.completion_text and not llm_response.tools_call_args:
-            logger.error(f"API 返回的 completion 无法解析：{completion}。")
-            raise Exception(f"API 返回的 completion 无法解析：{completion}。")
-
-        llm_response.raw_completion = completion
+            raise Exception(f"Anthropic API 返回的 completion 无法解析：{completion}。")
 
         return llm_response
 
+    async def _query_stream(
+        self, payloads: dict, tools: FuncCall
+    ) -> AsyncGenerator[LLMResponse, None]:
+        if tools:
+            if tool_list := tools.get_func_desc_anthropic_style():
+                payloads["tools"] = tool_list
+
+        # 用于累积工具调用信息
+        tool_use_buffer = {}
+        # 用于累积最终结果
+        final_text = ""
+        final_tool_calls = []
+
+        async with self.client.messages.stream(**payloads) as stream:
+            assert isinstance(stream, anthropic.AsyncMessageStream)
+            async for event in stream:
+                if event.type == "content_block_start":
+                    if event.content_block.type == "text":
+                        # 文本块开始
+                        yield LLMResponse(
+                            role="assistant", completion_text="", is_chunk=True
+                        )
+                    elif event.content_block.type == "tool_use":
+                        # 工具使用块开始，初始化缓冲区
+                        tool_use_buffer[event.index] = {
+                            "id": event.content_block.id,
+                            "name": event.content_block.name,
+                            "input": {},
+                        }
+
+                elif event.type == "content_block_delta":
+                    if event.delta.type == "text_delta":
+                        # 文本增量
+                        final_text += event.delta.text
+                        yield LLMResponse(
+                            role="assistant",
+                            completion_text=event.delta.text,
+                            is_chunk=True,
+                        )
+                    elif event.delta.type == "input_json_delta":
+                        # 工具调用参数增量
+                        if event.index in tool_use_buffer:
+                            # 累积 JSON 输入
+                            if "input_json" not in tool_use_buffer[event.index]:
+                                tool_use_buffer[event.index]["input_json"] = ""
+                            tool_use_buffer[event.index]["input_json"] += (
+                                event.delta.partial_json
+                            )
+
+                elif event.type == "content_block_stop":
+                    # 内容块结束
+                    if event.index in tool_use_buffer:
+                        # 解析完整的工具调用
+                        tool_info = tool_use_buffer[event.index]
+                        try:
+                            if "input_json" in tool_info:
+                                tool_info["input"] = json.loads(tool_info["input_json"])
+
+                            # 添加到最终结果
+                            final_tool_calls.append(
+                                {
+                                    "id": tool_info["id"],
+                                    "name": tool_info["name"],
+                                    "input": tool_info["input"],
+                                }
+                            )
+
+                            yield LLMResponse(
+                                role="tool",
+                                completion_text="",
+                                tools_call_args=[tool_info["input"]],
+                                tools_call_name=[tool_info["name"]],
+                                tools_call_ids=[tool_info["id"]],
+                                is_chunk=True,
+                            )
+                        except json.JSONDecodeError:
+                            # JSON 解析失败，跳过这个工具调用
+                            logger.warning(f"工具调用参数 JSON 解析失败: {tool_info}")
+
+                        # 清理缓冲区
+                        del tool_use_buffer[event.index]
+
+        # 返回最终的完整结果
+        final_response = LLMResponse(
+            role="assistant", completion_text=final_text, is_chunk=False
+        )
+
+        if final_tool_calls:
+            final_response.tools_call_args = [
+                call["input"] for call in final_tool_calls
+            ]
+            final_response.tools_call_name = [call["name"] for call in final_tool_calls]
+            final_response.tools_call_ids = [call["id"] for call in final_tool_calls]
+
+        yield final_response
+
     async def text_chat(
         self,
-        prompt: str,
-        session_id: str = None,
-        image_urls: List[str] = [],
-        func_tool: FuncCall = None,
-        contexts=[],
+        prompt,
+        session_id=None,
+        image_urls=None,
+        func_tool=None,
+        contexts=None,
         system_prompt=None,
-        tool_calls_result: ToolCallsResult = None,
+        tool_calls_result=None,
         **kwargs,
     ) -> LLMResponse:
-        if not prompt:
-            prompt = "<image>"
-
+        if contexts is None:
+            contexts = []
         new_record = await self.assemble_context(prompt, image_urls)
         context_query = [*contexts, new_record]
+        if system_prompt:
+            context_query.insert(0, {"role": "system", "content": system_prompt})
 
         for part in context_query:
             if "_no_save" in part:
                 del part["_no_save"]
 
+        # tool calls result
         if tool_calls_result:
-            # 暂时这样写。
-            prompt += f"Here are the related results via using tools: {str(tool_calls_result.tool_calls_result)}"
+            if not isinstance(tool_calls_result, list):
+                context_query.extend(tool_calls_result.to_openai_messages())
+            else:
+                for tcr in tool_calls_result:
+                    context_query.extend(tcr.to_openai_messages())
+
+        system_prompt, new_messages = self._prepare_payload(context_query)
 
         model_config = self.provider_config.get("model_config", {})
+        model_config["model"] = self.get_model()
+
+        payloads = {"messages": new_messages, **model_config}
 
-        payloads = {"messages": context_query, **model_config}
         # Anthropic has a different way of handling system prompts
         if system_prompt:
             payloads["system"] = system_prompt
@@ -133,32 +270,9 @@ class ProviderAnthropic(ProviderOpenAIOfficial):
         llm_response = None
         try:
             llm_response = await self._query(payloads, func_tool)
-
         except Exception as e:
-            if "maximum context length" in str(e):
-                retry_cnt = 20
-                while retry_cnt > 0:
-                    logger.warning(
-                        f"上下文长度超过限制。尝试弹出最早的记录然后重试。当前记录条数: {len(context_query)}"
-                    )
-                    try:
-                        await self.pop_record(context_query)
-                        response = await self.client.messages.create(
-                            messages=context_query, **model_config
-                        )
-                        llm_response = LLMResponse("assistant")
-                        llm_response.result_chain = MessageChain().message(response.content[0].text)
-                        llm_response.raw_completion = response
-                        return llm_response
-                    except Exception as e:
-                        if "maximum context length" in str(e):
-                            retry_cnt -= 1
-                        else:
-                            raise e
-                return LLMResponse("err", "err: 请尝试 /reset 清除会话记录。")
-            else:
-                logger.error(f"发生了错误。Provider 配置如下: {model_config}")
-                raise e
+            logger.error(f"发生了错误。Provider 配置如下: {model_config}")
+            raise e
 
         return llm_response
 
@@ -173,21 +287,34 @@ class ProviderAnthropic(ProviderOpenAIOfficial):
         tool_calls_result=None,
         **kwargs,
     ):
-        # raise NotImplementedError("This method is not implemented yet.")
-        # 调用 text_chat 模拟流式
-        llm_response = await self.text_chat(
-            prompt=prompt,
-            session_id=session_id,
-            image_urls=image_urls,
-            func_tool=func_tool,
-            contexts=contexts,
-            system_prompt=system_prompt,
-            tool_calls_result=tool_calls_result,
-        )
-        llm_response.is_chunk = True
-        yield llm_response
-        llm_response.is_chunk = False
-        yield llm_response
+        if contexts is None:
+            contexts = []
+        new_record = await self.assemble_context(prompt, image_urls)
+        context_query = [*contexts, new_record]
+        if system_prompt:
+            context_query.insert(0, {"role": "system", "content": system_prompt})
+
+        for part in context_query:
+            if "_no_save" in part:
+                del part["_no_save"]
+
+        # tool calls result
+        if tool_calls_result:
+            context_query.extend(tool_calls_result.to_openai_messages())
+
+        system_prompt, new_messages = self._prepare_payload(context_query)
+
+        model_config = self.provider_config.get("model_config", {})
+        model_config["model"] = self.get_model()
+
+        payloads = {"messages": new_messages, **model_config}
+
+        # Anthropic has a different way of handling system prompts
+        if system_prompt:
+            payloads["system"] = system_prompt
+
+        async for llm_response in self._query_stream(payloads, func_tool):
+            yield llm_response
 
     async def assemble_context(self, text: str, image_urls: List[str] = None):
         """组装上下文，支持文本和图片"""
@@ -230,3 +357,28 @@ class ProviderAnthropic(ProviderOpenAIOfficial):
             )
 
         return {"role": "user", "content": content}
+
+    async def encode_image_bs64(self, image_url: str) -> str:
+        """
+        将图片转换为 base64
+        """
+        if image_url.startswith("base64://"):
+            return image_url.replace("base64://", "data:image/jpeg;base64,")
+        with open(image_url, "rb") as f:
+            image_bs64 = base64.b64encode(f.read()).decode("utf-8")
+            return "data:image/jpeg;base64," + image_bs64
+        return ""
+
+    def get_current_key(self) -> str:
+        return self.chosen_api_key
+
+    async def get_models(self) -> List[str]:
+        models_str = []
+        models = await self.client.models.list()
+        models = sorted(models.data, key=lambda x: x.id)
+        for model in models:
+            models_str.append(model.id)
+        return models_str
+
+    def set_key(self, key: str):
+        self.chosen_api_key = key

astrbot/core/provider/sources/azure_tts_source.py

@@ -53,8 +53,8 @@ class OTTSProvider:
     async def _generate_signature(self) -> str:
         await self._sync_time()
         timestamp = int(time.time()) + self.time_offset
-        nonce = ''.join(random.choices('abcdefghijklmnopqrstuvwxyz0123456789', k=10))
-        path = re.sub(r'^https?://[^/]+', '', self.api_url) or '/'
+        nonce = "".join(random.choices("abcdefghijklmnopqrstuvwxyz0123456789", k=10))
+        path = re.sub(r"^https?://[^/]+", "", self.api_url) or "/"
         return f"{timestamp}-{nonce}-0-{hashlib.md5(f'{path}-{timestamp}-{nonce}-0-{self.skey}'.encode()).hexdigest()}"
 
     async def get_audio(self, text: str, voice_params: Dict) -> str:
@@ -92,7 +92,7 @@ class AzureNativeProvider(TTSProvider):
     def __init__(self, provider_config: dict, provider_settings: dict):
         super().__init__(provider_config, provider_settings)
         self.subscription_key = provider_config.get("azure_tts_subscription_key", "").strip()
-        if not re.fullmatch(r'^[a-zA-Z0-9]{32}$', self.subscription_key):
+        if not re.fullmatch(r"^[a-zA-Z0-9]{32}$", self.subscription_key):
             raise ValueError("无效的Azure订阅密钥")
         self.region = provider_config.get("azure_tts_region", "eastus").strip()
         self.endpoint = f"https://{self.region}.tts.speech.microsoft.com/cognitiveservices/v1"
@@ -188,7 +188,7 @@ class AzureTTSProvider(TTSProvider):
                 raise ValueError(error_msg) from e
             except KeyError as e:
                 raise ValueError(f"配置错误: 缺少必要参数 {e}") from e
-        if re.fullmatch(r'^[a-zA-Z0-9]{32}$', key_value):
+        if re.fullmatch(r"^[a-zA-Z0-9]{32}$", key_value):
             return AzureNativeProvider(config, self.provider_settings)
         raise ValueError("订阅密钥格式无效，应为32位字母数字或other[...]格式")
 

astrbot/core/provider/sources/dashscope_source.py

@@ -5,7 +5,6 @@ from typing import List
 from .. import Provider, Personality
 from ..entities import LLMResponse
 from ..func_tool_manager import FuncCall
-from astrbot.core.db import BaseDatabase
 from ..register import register_provider_adapter
 from astrbot.core.message.message_event_result import MessageChain
 from .openai_source import ProviderOpenAIOfficial
@@ -19,16 +18,12 @@ class ProviderDashscope(ProviderOpenAIOfficial):
         self,
         provider_config: dict,
         provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=False,
-        default_persona: Personality = None,
+        default_persona: Personality | None = None,
     ) -> None:
         Provider.__init__(
             self,
             provider_config,
             provider_settings,
-            persistant_history,
-            db_helper,
             default_persona,
         )
         self.api_key = provider_config.get("dashscope_api_key", "")
@@ -74,6 +69,8 @@ class ProviderDashscope(ProviderOpenAIOfficial):
         system_prompt: str = None,
         **kwargs,
     ) -> LLMResponse:
+        if contexts is None:
+            contexts = []
         # 获得会话变量
         payload_vars = self.variables.copy()
         # 动态变量

astrbot/core/provider/sources/dify_source.py

@@ -1,10 +1,9 @@
 import astrbot.core.message.components as Comp
 import os
 from typing import List
-from .. import Provider, Personality
+from .. import Provider
 from ..entities import LLMResponse
 from ..func_tool_manager import FuncCall
-from astrbot.core.db import BaseDatabase
 from ..register import register_provider_adapter
 from astrbot.core.utils.dify_api_client import DifyAPIClient
 from astrbot.core.utils.io import download_image_by_url, download_file
@@ -17,17 +16,13 @@ from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 class ProviderDify(Provider):
     def __init__(
         self,
-        provider_config: dict,
-        provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=False,
-        default_persona: Personality = None,
+        provider_config,
+        provider_settings,
+        default_persona = None,
     ) -> None:
         super().__init__(
             provider_config,
             provider_settings,
-            persistant_history,
-            db_helper,
             default_persona,
         )
         self.api_key = provider_config.get("dify_api_key", "")
@@ -61,13 +56,16 @@ class ProviderDify(Provider):
         self,
         prompt: str,
         session_id: str = None,
-        image_urls: List[str] = [],
+        image_urls: List[str] = None,
         func_tool: FuncCall = None,
         contexts: List = None,
         system_prompt: str = None,
         **kwargs,
     ) -> LLMResponse:
+        if image_urls is None:
+            image_urls = []
         result = ""
+        session_id = session_id or kwargs.get("user") # 1734
         conversation_id = self.conversation_ids.get(session_id, "")
 
         files_payload = []

astrbot/core/provider/sources/gemini_embedding_source.py

@@ -0,0 +1,63 @@
+from google import genai
+from google.genai import types
+from google.genai.errors import APIError
+from ..provider import EmbeddingProvider
+from ..register import register_provider_adapter
+from ..entities import ProviderType
+
+
+@register_provider_adapter(
+    "gemini_embedding",
+    "Google Gemini Embedding 提供商适配器",
+    provider_type=ProviderType.EMBEDDING,
+)
+class GeminiEmbeddingProvider(EmbeddingProvider):
+    def __init__(self, provider_config: dict, provider_settings: dict) -> None:
+        super().__init__(provider_config, provider_settings)
+        self.provider_config = provider_config
+        self.provider_settings = provider_settings
+
+        api_key: str = provider_config.get("embedding_api_key")
+        api_base: str = provider_config.get("embedding_api_base", None)
+        timeout: int = int(provider_config.get("timeout", 20))
+
+        http_options = types.HttpOptions(timeout=timeout * 1000)
+        if api_base:
+            if api_base.endswith("/"):
+                api_base = api_base[:-1]
+            http_options.base_url = api_base
+
+        self.client = genai.Client(api_key=api_key, http_options=http_options).aio
+
+        self.model = provider_config.get(
+            "embedding_model", "gemini-embedding-exp-03-07"
+        )
+        self.dimension = provider_config.get("embedding_dimensions", 768)
+
+    async def get_embedding(self, text: str) -> list[float]:
+        """
+        获取文本的嵌入
+        """
+        try:
+            result = await self.client.models.embed_content(
+                model=self.model, contents=text
+            )
+            return result.embeddings[0].values
+        except APIError as e:
+            raise Exception(f"Gemini Embedding API请求失败: {e.message}")
+
+    async def get_embeddings(self, texts: list[str]) -> list[list[float]]:
+        """
+        批量获取文本的嵌入
+        """
+        try:
+            result = await self.client.models.embed_content(
+                model=self.model, contents=texts
+            )
+            return [embedding.values for embedding in result.embeddings]
+        except APIError as e:
+            raise Exception(f"Gemini Embedding API批量请求失败: {e.message}")
+
+    def get_dim(self) -> int:
+        """获取向量的维度"""
+        return self.dimension

astrbot/core/provider/sources/gemini_source.py

@@ -12,8 +12,7 @@ from google.genai.errors import APIError
 
 import astrbot.core.message.components as Comp
 from astrbot import logger
-from astrbot.api.provider import Personality, Provider
-from astrbot.core.db import BaseDatabase
+from astrbot.api.provider import Provider
 from astrbot.core.message.message_event_result import MessageChain
 from astrbot.core.provider.entities import LLMResponse, ToolCallsResult
 from astrbot.core.provider.func_tool_manager import FuncCall
@@ -52,17 +51,13 @@ class ProviderGoogleGenAI(Provider):
 
     def __init__(
         self,
-        provider_config: dict,
-        provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=True,
-        default_persona: Personality = None,
+        provider_config,
+        provider_settings,
+        default_persona=None,
     ) -> None:
         super().__init__(
             provider_config,
             provider_settings,
-            persistant_history,
-            db_helper,
             default_persona,
         )
         self.api_keys: list = provider_config.get("key", [])
@@ -141,24 +136,66 @@ class ProviderGoogleGenAI(Provider):
             logger.warning("流式输出不支持图片模态，已自动降级为文本模态")
             modalities = ["Text"]
 
-        tool_list = None
+        tool_list = []
+        model_name = self.get_model()
         native_coderunner = self.provider_config.get("gm_native_coderunner", False)
         native_search = self.provider_config.get("gm_native_search", False)
+        url_context = self.provider_config.get("gm_url_context", False)
 
-        if native_coderunner:
-            tool_list = [types.Tool(code_execution=types.ToolCodeExecution())]
-            if native_search:
-                logger.warning("已启用代码执行工具，搜索工具将被忽略")
-            if tools:
-                logger.warning("已启用代码执行工具，函数工具将被忽略")
-        elif native_search:
-            tool_list = [types.Tool(google_search=types.GoogleSearch())]
-            if tools:
-                logger.warning("已启用搜索工具，函数工具将被忽略")
+        if "gemini-2.5" in model_name:
+            if native_coderunner:
+                tool_list.append(types.Tool(code_execution=types.ToolCodeExecution()))
+                if native_search:
+                    logger.warning("代码执行工具与搜索工具互斥，已忽略搜索工具")
+                if url_context:
+                    logger.warning(
+                        "代码执行工具与URL上下文工具互斥，已忽略URL上下文工具"
+                    )
+            else:
+                if native_search:
+                    tool_list.append(types.Tool(google_search=types.GoogleSearch()))
+
+                if url_context:
+                    if hasattr(types, "UrlContext"):
+                        tool_list.append(types.Tool(url_context=types.UrlContext()))
+                    else:
+                        logger.warning(
+                            "当前 SDK 版本不支持 URL 上下文工具，已忽略该设置，请升级 google-genai 包"
+                        )
+
+        elif "gemini-2.0-lite" in model_name:
+            if native_coderunner or native_search or url_context:
+                logger.warning(
+                    "gemini-2.0-lite 不支持代码执行、搜索工具和URL上下文，将忽略这些设置"
+                )
+            tool_list = None
+
+        else:
+            if native_coderunner:
+                tool_list.append(types.Tool(code_execution=types.ToolCodeExecution()))
+                if native_search:
+                    logger.warning("代码执行工具与搜索工具互斥，已忽略搜索工具")
+            elif native_search:
+                tool_list.append(types.Tool(google_search=types.GoogleSearch()))
+
+            if url_context and not native_coderunner:
+                if hasattr(types, "UrlContext"):
+                    tool_list.append(types.Tool(url_context=types.UrlContext()))
+                else:
+                    logger.warning(
+                        "当前 SDK 版本不支持 URL 上下文工具，已忽略该设置，请升级 google-genai 包"
+                    )
+
+        if not tool_list:
+            tool_list = None
+
+        if tools and tool_list:
+            logger.warning("已启用原生工具，函数工具将被忽略")
         elif tools and (func_desc := tools.get_func_desc_google_genai_style()):
             tool_list = [
                 types.Tool(function_declarations=func_desc["function_declarations"])
             ]
+
         return types.GenerateContentConfig(
             system_instruction=system_instruction,
             temperature=temperature,
@@ -222,12 +259,10 @@ class ProviderGoogleGenAI(Provider):
                 contents.append(content_cls(parts=part))
 
         gemini_contents: list[types.Content] = []
-        native_tool_enabled = any(
-            [
-                self.provider_config.get("gm_native_coderunner", False),
-                self.provider_config.get("gm_native_search", False),
-            ]
-        )
+        native_tool_enabled = any([
+            self.provider_config.get("gm_native_coderunner", False),
+            self.provider_config.get("gm_native_search", False),
+        ])
         for message in payloads["messages"]:
             role, content = message["role"], message.get("content")
 
@@ -291,19 +326,19 @@ class ProviderGoogleGenAI(Provider):
         result_parts: Optional[types.Part] = result.candidates[0].content.parts
 
         if finish_reason == types.FinishReason.SAFETY:
-            raise Exception("模型生成内容未通过用户定义的内容安全检查")
+            raise Exception("模型生成内容未通过 Gemini 平台的安全检查")
 
         if finish_reason in {
             types.FinishReason.PROHIBITED_CONTENT,
             types.FinishReason.SPII,
             types.FinishReason.BLOCKLIST,
         }:
-            raise Exception("模型生成内容违反Gemini平台政策")
+            raise Exception("模型生成内容违反 Gemini 平台政策")
 
         # 防止旧版本SDK不存在IMAGE_SAFETY
         if hasattr(types.FinishReason, "IMAGE_SAFETY"):
             if finish_reason == types.FinishReason.IMAGE_SAFETY:
-                raise Exception("模型生成内容违反Gemini平台政策")
+                raise Exception("模型生成内容违反 Gemini 平台政策")
 
         if not result_parts:
             logger.debug(result.candidates)
@@ -464,12 +499,12 @@ class ProviderGoogleGenAI(Provider):
     async def text_chat(
         self,
         prompt: str,
-        session_id: str = None,
-        image_urls: list[str] = None,
-        func_tool: FuncCall = None,
-        contexts: list = None,
-        system_prompt: str = None,
-        tool_calls_result: ToolCallsResult = None,
+        session_id=None,
+        image_urls=None,
+        func_tool=None,
+        contexts=None,
+        system_prompt=None,
+        tool_calls_result=None,
         **kwargs,
     ) -> LLMResponse:
         if contexts is None:
@@ -485,7 +520,11 @@ class ProviderGoogleGenAI(Provider):
 
         # tool calls result
         if tool_calls_result:
-            context_query.extend(tool_calls_result.to_openai_messages())
+            if not isinstance(tool_calls_result, list):
+                context_query.extend(tool_calls_result.to_openai_messages())
+            else:
+                for tcr in tool_calls_result:
+                    context_query.extend(tcr.to_openai_messages())
 
         model_config = self.provider_config.get("model_config", {})
         model_config["model"] = self.get_model()
@@ -589,9 +628,10 @@ class ProviderGoogleGenAI(Provider):
                 if not image_data:
                     logger.warning(f"图片 {image_url} 得到的结果为空，将忽略。")
                     continue
-                user_content["content"].append(
-                    {"type": "image_url", "image_url": {"url": image_data}}
-                )
+                user_content["content"].append({
+                    "type": "image_url",
+                    "image_url": {"url": image_data},
+                })
             return user_content
         else:
             return {"role": "user", "content": text}

astrbot/core/provider/sources/gemini_tts_source.py

@@ -0,0 +1,79 @@
+import os
+import uuid
+import wave
+
+from google import genai
+from google.genai import types
+
+from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+
+from ..entities import ProviderType
+from ..provider import TTSProvider
+from ..register import register_provider_adapter
+
+
+@register_provider_adapter(
+    "gemini_tts", "Gemini TTS API", provider_type=ProviderType.TEXT_TO_SPEECH
+)
+class ProviderGeminiTTSAPI(TTSProvider):
+    def __init__(
+        self,
+        provider_config: dict,
+        provider_settings: dict,
+    ) -> None:
+        super().__init__(provider_config, provider_settings)
+        api_key: str = provider_config.get("gemini_tts_api_key", "")
+        api_base: str | None = provider_config.get("gemini_tts_api_base")
+        timeout: int = int(provider_config.get("gemini_tts_timeout", 20))
+        http_options = types.HttpOptions(timeout=timeout * 1000)
+
+        if api_base:
+            if api_base.endswith("/"):
+                api_base = api_base[:-1]
+            http_options.base_url = api_base
+
+        self.client = genai.Client(api_key=api_key, http_options=http_options).aio
+        self.model: str = provider_config.get(
+            "gemini_tts_model", "gemini-2.5-flash-preview-tts"
+        )
+        self.prefix: str | None = provider_config.get(
+            "gemini_tts_prefix",
+        )
+        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")
+        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(
+            model=self.model,
+            contents=prompt,
+            config=types.GenerateContentConfig(
+                response_modalities=["AUDIO"],
+                speech_config=types.SpeechConfig(
+                    voice_config=types.VoiceConfig(
+                        prebuilt_voice_config=types.PrebuiltVoiceConfig(
+                            voice_name=self.voice_name,
+                        )
+                    )
+                ),
+            ),
+        )
+
+        # 不想看类型检查报错
+        if (
+            not response.candidates
+            or not response.candidates[0].content
+            or not response.candidates[0].content.parts
+            or not response.candidates[0].content.parts[0].inline_data
+            or not response.candidates[0].content.parts[0].inline_data.data
+        ):
+            raise Exception("No audio content returned from Gemini TTS API.")
+
+        with wave.open(path, "wb") as wf:
+            wf.setnchannels(1)
+            wf.setsampwidth(2)
+            wf.setframerate(24000)
+            wf.writeframes(response.candidates[0].content.parts[0].inline_data.data)
+
+        return path

astrbot/core/provider/sources/gsv_selfhosted_source.py

@@ -0,0 +1,148 @@
+import asyncio
+import os
+import uuid
+
+import aiohttp
+from ..provider import TTSProvider
+from ..entities import ProviderType
+from ..register import register_provider_adapter
+from astrbot import logger
+from astrbot.core.utils.astrbot_path import get_astrbot_data_path
+
+
+@register_provider_adapter(
+    provider_type_name="gsv_tts_selfhost",
+    desc="GPT-SoVITS TTS(本地加载)",
+    provider_type=ProviderType.TEXT_TO_SPEECH,
+)
+class ProviderGSVTTS(TTSProvider):
+    def __init__(
+        self,
+        provider_config: dict,
+        provider_settings: dict,
+    ) -> None:
+        super().__init__(provider_config, provider_settings)
+
+        self.api_base = provider_config.get("api_base", "http://127.0.0.1:9880").rstrip(
+            "/"
+        )
+        self.gpt_weights_path: str = provider_config.get("gpt_weights_path", "")
+        self.sovits_weights_path: str = provider_config.get("sovits_weights_path", "")
+
+        # TTS 请求的默认参数，移除前缀gsv_
+        self.default_params: dict = {
+            key.removeprefix("gsv_"): str(value).lower()
+            for key, value in provider_config.get("gsv_default_parms", {}).items()
+        }
+        self.timeout = provider_config.get("timeout", 60)
+        self._session: aiohttp.ClientSession | None = None
+
+    async def initialize(self):
+        """异步初始化：在 ProviderManager 中被调用"""
+        self._session = aiohttp.ClientSession(
+            timeout=aiohttp.ClientTimeout(total=self.timeout)
+        )
+        try:
+            await self._set_model_weights()
+            logger.info("[GSV TTS] 初始化完成")
+        except Exception as e:
+            logger.error(f"[GSV TTS] 初始化失败：{e}")
+            raise
+
+    def get_session(self) -> aiohttp.ClientSession:
+        if not self._session or self._session.closed:
+            raise RuntimeError(
+                "[GSV TTS] Provider HTTP session is not ready or closed."
+            )
+        return self._session
+
+    async def _make_request(
+        self, endpoint: str, params=None, retries: int = 3
+    ) -> bytes | None:
+        """发起请求"""
+        for attempt in range(retries):
+            logger.debug(f"[GSV TTS] 请求地址：{endpoint}，参数：{params}")
+            try:
+                async with self.get_session().get(endpoint, params=params) as response:
+                    if response.status != 200:
+                        error_text = await response.text()
+                        raise Exception(
+                            f"[GSV TTS] Request to {endpoint} failed with status {response.status}: {error_text}"
+                        )
+                    return await response.read()
+            except Exception as e:
+                if attempt < retries - 1:
+                    logger.warning(
+                        f"[GSV TTS] 请求 {endpoint} 第 {attempt + 1} 次失败：{e}，重试中..."
+                    )
+                    await asyncio.sleep(1)
+                else:
+                    logger.error(f"[GSV TTS] 请求 {endpoint} 最终失败：{e}")
+                    raise
+
+    async def _set_model_weights(self):
+        """设置模型路径"""
+        try:
+            if self.gpt_weights_path:
+                await self._make_request(
+                    f"{self.api_base}/set_gpt_weights",
+                    {"weights_path": self.gpt_weights_path},
+                )
+                logger.info(f"[GSV TTS] 成功设置 GPT 模型路径：{self.gpt_weights_path}")
+            else:
+                logger.info("[GSV TTS] GPT 模型路径未配置，将使用内置 GPT 模型")
+
+            if self.sovits_weights_path:
+                await self._make_request(
+                    f"{self.api_base}/set_sovits_weights",
+                    {"weights_path": self.sovits_weights_path},
+                )
+                logger.info(
+                    f"[GSV TTS] 成功设置 SoVITS 模型路径：{self.sovits_weights_path}"
+                )
+            else:
+                logger.info("[GSV TTS] SoVITS 模型路径未配置，将使用内置 SoVITS 模型")
+        except aiohttp.ClientError as e:
+            logger.error(f"[GSV TTS] 设置模型路径时发生网络错误：{e}")
+        except Exception as e:
+            logger.error(f"[GSV TTS] 设置模型路径时发生未知错误：{e}")
+
+    async def get_audio(self, text: str) -> str:
+        """实现 TTS 核心方法，根据文本内容自动切换情绪"""
+        if not text.strip():
+            raise ValueError("[GSV TTS] TTS 文本不能为空")
+
+        endpoint = f"{self.api_base}/tts"
+
+        params = self.build_synthesis_params(text)
+
+        temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+        os.makedirs(temp_dir, exist_ok=True)
+        path = os.path.join(temp_dir, f"gsv_tts_{uuid.uuid4().hex}.wav")
+
+        logger.debug(f"[GSV TTS] 正在调用语音合成接口，参数：{params}")
+
+        result = await self._make_request(endpoint, params)
+        if isinstance(result, bytes):
+            with open(path, "wb") as f:
+                f.write(result)
+            return path
+        else:
+            raise Exception(f"[GSV TTS] 合成失败，输入文本：{text}，错误信息：{result}")
+
+    def build_synthesis_params(self, text: str) -> dict:
+        """
+        构建语音合成所需的参数字典。
+
+        当前仅包含默认参数 + 文本，未来可在此基础上动态添加如情绪、角色等语义控制字段。
+        """
+        params = self.default_params.copy()
+        params["text"] = text
+        # TODO: 在此处添加情绪分析，例如 params["emotion"] = detect_emotion(text)
+        return params
+
+    async def terminate(self):
+        """终止释放资源：在 ProviderManager 中被调用"""
+        if self._session and not self._session.closed:
+            await self._session.close()
+            logger.info("[GSV TTS] Session 已关闭")

astrbot/core/provider/sources/llmtuner_source.py

@@ -1,132 +0,0 @@
-import os
-from llmtuner.chat import ChatModel
-from typing import List
-from .. import Provider
-from ..entities import LLMResponse
-from ..func_tool_manager import FuncCall
-from astrbot.core.db import BaseDatabase
-from ..register import register_provider_adapter
-
-
-@register_provider_adapter(
-    "llm_tuner", "LLMTuner 适配器, 用于装载使用 LlamaFactory 微调后的模型"
-)
-class LLMTunerModelLoader(Provider):
-    def __init__(
-        self,
-        provider_config: dict,
-        provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=True,
-        default_persona=None,
-    ) -> None:
-        super().__init__(
-            provider_config,
-            provider_settings,
-            persistant_history,
-            db_helper,
-            default_persona,
-        )
-        if not os.path.exists(provider_config["base_model_path"]) or not os.path.exists(
-            provider_config["adapter_model_path"]
-        ):
-            raise FileNotFoundError("模型文件路径不存在。")
-        self.base_model_path = provider_config["base_model_path"]
-        self.adapter_model_path = provider_config["adapter_model_path"]
-        self.model = ChatModel(
-            {
-                "model_name_or_path": self.base_model_path,
-                "adapter_name_or_path": self.adapter_model_path,
-                "template": provider_config["llmtuner_template"],
-                "finetuning_type": provider_config["finetuning_type"],
-                "quantization_bit": provider_config["quantization_bit"],
-            }
-        )
-        self.set_model(
-            os.path.basename(self.base_model_path)
-            + "_"
-            + os.path.basename(self.adapter_model_path)
-        )
-
-    async def assemble_context(self, text: str, image_urls: List[str] = None):
-        """
-        组装上下文。
-        """
-        return {"role": "user", "content": text}
-
-    async def text_chat(
-        self,
-        prompt: str,
-        session_id: str = None,
-        image_urls: List[str] = None,
-        func_tool: FuncCall = None,
-        contexts: List = [],
-        system_prompt: str = None,
-        **kwargs,
-    ) -> LLMResponse:
-        system_prompt = ""
-        new_record = {"role": "user", "content": prompt}
-        query_context = [*contexts, new_record]
-
-        # 提取出系统提示
-        system_idxs = []
-        for idx, context in enumerate(query_context):
-            if context["role"] == "system":
-                system_idxs.append(idx)
-
-            if "_no_save" in context:
-                del context["_no_save"]
-
-        for idx in reversed(system_idxs):
-            system_prompt += " " + query_context.pop(idx)["content"]
-
-        conf = {
-            "messages": query_context,
-            "system": system_prompt,
-        }
-        if func_tool:
-            tool_list = func_tool.get_func_desc_openai_style()
-            if tool_list:
-                conf["tools"] = tool_list
-
-        responses = await self.model.achat(**conf)
-
-        llm_response = LLMResponse("assistant", responses[-1].response_text)
-
-        return llm_response
-
-    async def text_chat_stream(
-        self,
-        prompt,
-        session_id=None,
-        image_urls=...,
-        func_tool=None,
-        contexts=...,
-        system_prompt=None,
-        tool_calls_result=None,
-        **kwargs,
-    ):
-        # raise NotImplementedError("This method is not implemented yet.")
-        # 调用 text_chat 模拟流式
-        llm_response = await self.text_chat(
-            prompt=prompt,
-            session_id=session_id,
-            image_urls=image_urls,
-            func_tool=func_tool,
-            contexts=contexts,
-            system_prompt=system_prompt,
-            tool_calls_result=tool_calls_result,
-        )
-        llm_response.is_chunk = True
-        yield llm_response
-        llm_response.is_chunk = False
-        yield llm_response
-
-    async def get_current_key(self):
-        return "none"
-
-    async def set_key(self, key):
-        pass
-
-    async def get_models(self):
-        return [self.get_model()]

astrbot/core/provider/sources/openai_embedding_source.py

@@ -0,0 +1,43 @@
+from openai import AsyncOpenAI
+from ..provider import EmbeddingProvider
+from ..register import register_provider_adapter
+from ..entities import ProviderType
+
+
+@register_provider_adapter(
+    "openai_embedding",
+    "OpenAI API Embedding 提供商适配器",
+    provider_type=ProviderType.EMBEDDING,
+)
+class OpenAIEmbeddingProvider(EmbeddingProvider):
+    def __init__(self, provider_config: dict, provider_settings: dict) -> None:
+        super().__init__(provider_config, provider_settings)
+        self.provider_config = provider_config
+        self.provider_settings = provider_settings
+        self.client = AsyncOpenAI(
+            api_key=provider_config.get("embedding_api_key"),
+            base_url=provider_config.get(
+                "embedding_api_base", "https://api.openai.com/v1"
+            ),
+            timeout=int(provider_config.get("timeout", 20)),
+        )
+        self.model = provider_config.get("embedding_model", "text-embedding-3-small")
+        self.dimension = provider_config.get("embedding_dimensions", 1536)
+
+    async def get_embedding(self, text: str) -> list[float]:
+        """
+        获取文本的嵌入
+        """
+        embedding = await self.client.embeddings.create(input=text, model=self.model)
+        return embedding.data[0].embedding
+
+    async def get_embeddings(self, texts: list[str]) -> list[list[float]]:
+        """
+        批量获取文本的嵌入
+        """
+        embeddings = await self.client.embeddings.create(input=texts, model=self.model)
+        return [item.embedding for item in embeddings.data]
+
+    def get_dim(self) -> int:
+        """获取向量的维度"""
+        return self.dimension

astrbot/core/provider/sources/openai_source.py

@@ -9,14 +9,12 @@ import astrbot.core.message.components as Comp
 from openai import AsyncOpenAI, AsyncAzureOpenAI
 from openai.types.chat.chat_completion import ChatCompletion
 
-# from openai.types.chat.chat_completion_chunk import ChatCompletionChunk
 from openai._exceptions import NotFoundError, UnprocessableEntityError
 from openai.lib.streaming.chat._completions import ChatCompletionStreamState
 from astrbot.core.utils.io import download_image_by_url
 from astrbot.core.message.message_event_result import MessageChain
 
-from astrbot.core.db import BaseDatabase
-from astrbot.api.provider import Provider, Personality
+from astrbot.api.provider import Provider
 from astrbot import logger
 from astrbot.core.provider.func_tool_manager import FuncCall
 from typing import List, AsyncGenerator
@@ -30,17 +28,13 @@ from astrbot.core.provider.entities import LLMResponse, ToolCallsResult
 class ProviderOpenAIOfficial(Provider):
     def __init__(
         self,
-        provider_config: dict,
-        provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=True,
-        default_persona: Personality = None,
+        provider_config,
+        provider_settings,
+        default_persona = None,
     ) -> None:
         super().__init__(
             provider_config,
             provider_settings,
-            persistant_history,
-            db_helper,
             default_persona,
         )
         self.chosen_api_key = None
@@ -195,7 +189,11 @@ class ProviderOpenAIOfficial(Provider):
             for tool_call in choice.message.tool_calls:
                 for tool in tools.func_list:
                     if tool.name == tool_call.function.name:
-                        args = json.loads(tool_call.function.arguments)
+                        # workaround for #1454
+                        if isinstance(tool_call.function.arguments, str):
+                            args = json.loads(tool_call.function.arguments)
+                        else:
+                            args = tool_call.function.arguments
                         args_ls.append(args)
                         func_name_ls.append(tool_call.function.name)
                         tool_call_ids.append(tool_call.id)
@@ -220,12 +218,10 @@ class ProviderOpenAIOfficial(Provider):
     async def _prepare_chat_payload(
         self,
         prompt: str,
-        session_id: str = None,
-        image_urls: list[str] = None,
-        func_tool: FuncCall = None,
-        contexts: list=None,
-        system_prompt: str=None,
-        tool_calls_result: ToolCallsResult=None,
+        image_urls: list[str] | None = None,
+        contexts: list | None = None,
+        system_prompt: str | None = None,
+        tool_calls_result: ToolCallsResult | list[ToolCallsResult] | None = None,
         **kwargs,
     ) -> tuple:
         """准备聊天所需的有效载荷和上下文"""
@@ -242,14 +238,18 @@ class ProviderOpenAIOfficial(Provider):
 
         # tool calls result
         if tool_calls_result:
-            context_query.extend(tool_calls_result.to_openai_messages())
+            if isinstance(tool_calls_result, ToolCallsResult):
+                context_query.extend(tool_calls_result.to_openai_messages())
+            else:
+                for tcr in tool_calls_result:
+                    context_query.extend(tcr.to_openai_messages())
 
         model_config = self.provider_config.get("model_config", {})
         model_config["model"] = self.get_model()
 
         payloads = {"messages": context_query, **model_config}
 
-        return payloads, context_query, func_tool
+        return payloads, context_query
 
     async def _handle_api_error(
         self,
@@ -340,19 +340,17 @@ class ProviderOpenAIOfficial(Provider):
     async def text_chat(
         self,
         prompt,
-        session_id = None,
-        image_urls = None,
-        func_tool = None,
+        session_id=None,
+        image_urls=None,
+        func_tool=None,
         contexts=None,
         system_prompt=None,
         tool_calls_result=None,
         **kwargs,
     ) -> LLMResponse:
-        payloads, context_query, func_tool = await self._prepare_chat_payload(
+        payloads, context_query = await self._prepare_chat_payload(
             prompt,
-            session_id,
             image_urls,
-            func_tool,
             contexts,
             system_prompt,
             tool_calls_result,
@@ -418,11 +416,9 @@ 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, func_tool = await self._prepare_chat_payload(
+        payloads, context_query = await self._prepare_chat_payload(
             prompt,
-            session_id,
             image_urls,
-            func_tool,
             contexts,
             system_prompt,
             tool_calls_result,

astrbot/core/provider/sources/volcengine_tts.py

@@ -0,0 +1,107 @@
+import uuid
+import base64
+import json
+import os
+import traceback
+import asyncio
+import aiohttp
+import requests
+from ..provider import TTSProvider
+from ..entities import ProviderType
+from ..register import register_provider_adapter
+from astrbot import logger
+
+@register_provider_adapter(
+    "volcengine_tts", "火山引擎 TTS", provider_type=ProviderType.TEXT_TO_SPEECH
+)
+class ProviderVolcengineTTS(TTSProvider):
+    def __init__(self, provider_config: dict, provider_settings: dict) -> None:
+        super().__init__(provider_config, provider_settings)
+        self.api_key = provider_config.get("api_key", "")
+        self.appid = provider_config.get("appid", "")
+        self.cluster = provider_config.get("volcengine_cluster", "")
+        self.voice_type = provider_config.get("volcengine_voice_type", "")
+        self.speed_ratio = provider_config.get("volcengine_speed_ratio", 1.0)
+        self.api_base = provider_config.get("api_base", f"https://openspeech.bytedance.com/api/v1/tts")
+        self.timeout = provider_config.get("timeout", 20)
+
+    def _build_request_payload(self, text: str) -> dict:
+        return {
+            "app": {
+                "appid": self.appid,
+                "token": self.api_key,
+                "cluster": self.cluster
+            },
+            "user": {
+                "uid": str(uuid.uuid4())  
+            },
+            "audio": {
+                "voice_type": self.voice_type,
+                "encoding": "mp3",
+                "speed_ratio": self.speed_ratio,
+                "volume_ratio": 1.0,
+                "pitch_ratio": 1.0,
+            },
+            "request": {
+                "reqid": str(uuid.uuid4()),
+                "text": text,
+                "text_type": "plain",
+                "operation": "query",
+                "with_frontend": 1,
+                "frontend_type": "unitTson"
+            }
+        }
+
+    async def get_audio(self, text: str) -> str:
+        """异步方法获取语音文件路径"""
+        headers = {
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer; {self.api_key}"
+        }
+        
+        payload = self._build_request_payload(text)
+        
+        logger.debug(f"请求头: {headers}")
+        logger.debug(f"请求 URL: {self.api_base}")
+        logger.debug(f"请求体: {json.dumps(payload, ensure_ascii=False)[:100]}...")
+        
+        try:
+            async with aiohttp.ClientSession() as session:
+                async with session.post(
+                    self.api_base,
+                    data=json.dumps(payload), 
+                    headers=headers,
+                    timeout=self.timeout
+                ) as response:
+                    logger.debug(f"响应状态码: {response.status}")
+                    
+                    response_text = await response.text()
+                    logger.debug(f"响应内容: {response_text[:200]}...")
+                    
+                    if response.status == 200:
+                        resp_data = json.loads(response_text)
+                        
+                        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"
+                            
+                            loop = asyncio.get_running_loop()
+                            await loop.run_in_executor(
+                                None, 
+                                lambda: open(file_path, "wb").write(audio_data)
+                            )
+                            
+                            return file_path
+                        else:
+                            error_msg = resp_data.get("message", "未知错误")
+                            raise Exception(f"火山引擎 TTS API 返回错误: {error_msg}")
+                    else:
+                        raise Exception(f"火山引擎 TTS API 请求失败: {response.status}, {response_text}")
+        
+        except Exception as e:
+            error_details = traceback.format_exc()
+            logger.debug(f"火山引擎 TTS 异常详情: {error_details}")
+            raise Exception(f"火山引擎 TTS 异常: {str(e)}")

astrbot/core/provider/sources/zhipu_source.py

@@ -1,4 +1,3 @@
-from astrbot.core.db import BaseDatabase
 from astrbot import logger
 from astrbot.core.provider.func_tool_manager import FuncCall
 from typing import List
@@ -13,15 +12,11 @@ class ProviderZhipu(ProviderOpenAIOfficial):
         self,
         provider_config: dict,
         provider_settings: dict,
-        db_helper: BaseDatabase,
-        persistant_history=True,
         default_persona=None,
     ) -> None:
         super().__init__(
             provider_config,
             provider_settings,
-            db_helper,
-            persistant_history,
             default_persona,
         )
 
@@ -31,10 +26,12 @@ class ProviderZhipu(ProviderOpenAIOfficial):
         session_id: str = None,
         image_urls: List[str] = None,
         func_tool: FuncCall = None,
-        contexts=[],
+        contexts=None,
         system_prompt=None,
         **kwargs,
     ) -> LLMResponse:
+        if contexts is None:
+            contexts = []
         new_record = await self.assemble_context(prompt, image_urls)
         context_query = []
 

astrbot/core/rag/embedding/openai_source.py

@@ -1,20 +0,0 @@
-from typing import List
-from openai import AsyncOpenAI
-
-
-class SimpleOpenAIEmbedding:
-    def __init__(
-        self,
-        model,
-        api_key,
-        api_base=None,
-    ) -> None:
-        self.client = AsyncOpenAI(api_key=api_key, base_url=api_base)
-        self.model = model
-
-    async def get_embedding(self, text) -> List[float]:
-        """
-        获取文本的嵌入
-        """
-        embedding = await self.client.embeddings.create(input=text, model=self.model)
-        return embedding.data[0].embedding

astrbot/core/rag/knowledge_db_mgr.py

@@ -1,95 +0,0 @@
-import os
-from typing import List, Dict
-from astrbot.core import logger
-from .store import Store
-from astrbot.core.config import AstrBotConfig
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
-
-
-class KnowledgeDBManager:
-    def __init__(self, astrbot_config: AstrBotConfig) -> None:
-        self.db_path = os.path.join(get_astrbot_data_path(), "knowledge_db")
-        self.config = astrbot_config.get("knowledge_db", {})
-        self.astrbot_config = astrbot_config
-        if not os.path.exists(self.db_path):
-            os.makedirs(self.db_path)
-        self.store_insts: Dict[str, Store] = {}
-        for name, cfg in self.config.items():
-            if cfg["strategy"] == "embedding":
-                logger.info(f"加载 Chroma Vector Store：{name}")
-                try:
-                    from .store.chroma_db import ChromaVectorStore
-                except ImportError as ie:
-                    logger.error(f"{ie} 可能未安装 chromadb 库。")
-                    continue
-                self.store_insts[name] = ChromaVectorStore(
-                    name, cfg["embedding_config"]
-                )
-            else:
-                logger.error(f"不支持的策略：{cfg['strategy']}")
-
-    async def list_knowledge_db(self) -> List[str]:
-        return [
-            f
-            for f in os.listdir(self.db_path)
-            if os.path.isfile(os.path.join(self.db_path, f))
-        ]
-
-    async def create_knowledge_db(self, name: str, config: Dict):
-        """
-        config 格式：
-        ```
-        {
-            "strategy": "embedding", # 目前只支持 embedding
-            "chunk_method": {
-                "strategy": "fixed",
-                "chunk_size": 100,
-                "overlap_size": 10
-            },
-            "embedding_config": {
-                "strategy": "openai",
-                "base_url": "",
-                "model": "",
-                "api_key": ""
-            }
-        }
-        ```
-        """
-        if name in self.config:
-            raise ValueError(f"知识库已存在：{name}")
-
-        self.config[name] = config
-        self.astrbot_config["knowledge_db"] = self.config
-        self.astrbot_config.save_config()
-
-    async def insert_record(self, name: str, text: str):
-        if name not in self.store_insts:
-            raise ValueError(f"未找到知识库：{name}")
-
-        ret = []
-        match self.config[name]["chunk_method"]["strategy"]:
-            case "fixed":
-                chunk_size = self.config[name]["chunk_method"]["chunk_size"]
-                chunk_overlap = self.config[name]["chunk_method"]["overlap_size"]
-                ret = self._fixed_chunk(text, chunk_size, chunk_overlap)
-            case _:
-                pass
-
-        for chunk in ret:
-            await self.store_insts[name].save(chunk)
-
-    async def retrive_records(self, name: str, query: str, top_n: int = 3) -> List[str]:
-        if name not in self.store_insts:
-            raise ValueError(f"未找到知识库：{name}")
-
-        inst = self.store_insts[name]
-        return await inst.query(query, top_n)
-
-    def _fixed_chunk(self, text: str, chunk_size: int, chunk_overlap: int) -> List[str]:
-        chunks = []
-        start = 0
-        while start < len(text):
-            end = start + chunk_size
-            chunks.append(text[start:end])
-            start += chunk_size - chunk_overlap
-        return chunks

astrbot/core/rag/store/__init__.py

@@ -1,9 +0,0 @@
-from typing import List
-
-
-class Store:
-    async def save(self, text: str):
-        pass
-
-    async def query(self, query: str, top_n: int = 3) -> List[str]:
-        pass

astrbot/core/rag/store/chroma_db.py

@@ -1,44 +0,0 @@
-import chromadb
-import uuid
-from typing import List, Dict
-from astrbot.api import logger
-from ..embedding.openai_source import SimpleOpenAIEmbedding
-from . import Store
-from astrbot.core.utils.astrbot_path import get_astrbot_data_path
-
-
-class ChromaVectorStore(Store):
-    def __init__(self, name: str, embedding_cfg: Dict) -> None:
-        import os
-        self.chroma_client = chromadb.PersistentClient(
-            path=os.path.join(get_astrbot_data_path(), "long_term_memory_chroma.db")
-        )
-        self.collection = self.chroma_client.get_or_create_collection(name=name)
-        self.embedding = None
-        if embedding_cfg["strategy"] == "openai":
-            self.embedding = SimpleOpenAIEmbedding(
-                model=embedding_cfg["model"],
-                api_key=embedding_cfg["api_key"],
-                api_base=embedding_cfg.get("base_url", None),
-            )
-
-    async def save(self, text: str, metadata: Dict = None):
-        logger.debug(f"Saving text: {text}")
-        embedding = await self.embedding.get_embedding(text)
-
-        self.collection.upsert(
-            documents=text,
-            metadatas=metadata,
-            ids=str(uuid.uuid4()),
-            embeddings=embedding,
-        )
-
-    async def query(
-        self, query: str, top_n=3, metadata_filter: Dict = None
-    ) -> List[str]:
-        embedding = await self.embedding.get_embedding(query)
-
-        results = self.collection.query(
-            query_embeddings=embedding, n_results=top_n, where=metadata_filter
-        )
-        return results["documents"][0]

astrbot/core/star/__init__.py

@@ -18,10 +18,12 @@ class Star(CommandParserMixin):
         """将文本转换为图片"""
         return await html_renderer.render_t2i(text, return_url=return_url)
 
-    async def html_render(self, tmpl: str, data: dict, return_url=True) -> str:
+    async def html_render(
+        self, tmpl: str, data: dict, return_url=True, options: dict = None
+    ) -> str:
         """渲染 HTML"""
         return await html_renderer.render_custom_template(
-            tmpl, data, return_url=return_url
+            tmpl, data, return_url=return_url, options=options
         )
 
     async def terminate(self):

astrbot/core/star/context.py

@@ -3,6 +3,7 @@ from typing import List, Union
 
 from astrbot.core import sp
 from astrbot.core.provider.provider import Provider, TTSProvider, STTProvider
+from astrbot.core.provider.entities import ProviderType
 from astrbot.core.db import BaseDatabase
 from astrbot.core.config.astrbot_config import AstrBotConfig
 from astrbot.core.provider.func_tool_manager import FuncCall
@@ -16,7 +17,6 @@ from .star_handler import star_handlers_registry, StarHandlerMetadata, EventType
 from .filter.command import CommandFilter
 from .filter.regex import RegexFilter
 from typing import Awaitable
-from astrbot.core.rag.knowledge_db_mgr import KnowledgeDBManager
 from astrbot.core.conversation_mgr import ConversationManager
 from astrbot.core.star.filter.platform_adapter_type import (
     PlatformAdapterType,
@@ -42,6 +42,8 @@ class Context:
 
     platform_manager: PlatformManager = None
 
+    registered_web_apis: list = []
+
     # back compatibility
     _register_tasks: List[Awaitable] = []
     _star_manager = None
@@ -54,14 +56,12 @@ class Context:
         provider_manager: ProviderManager = None,
         platform_manager: PlatformManager = None,
         conversation_manager: ConversationManager = None,
-        knowledge_db_manager: KnowledgeDBManager = None,
     ):
         self._event_queue = event_queue
         self._config = config
         self._db = db
         self.provider_manager = provider_manager
         self.platform_manager = platform_manager
-        self.knowledge_db_manager = knowledge_db_manager
         self.conversation_manager = conversation_manager
 
     def get_registered_star(self, star_name: str) -> StarMetadata:
@@ -126,11 +126,8 @@ class Context:
         self.provider_manager.provider_insts.append(provider)
 
     def get_provider_by_id(self, provider_id: str) -> Provider:
-        """通过 ID 获取用于文本生成任务的 LLM Provider(Chat_Completion 类型)。"""
-        for provider in self.provider_manager.provider_insts:
-            if provider.meta().id == provider_id:
-                return provider
-        return None
+        """通过 ID 获取对应的 LLM Provider(Chat_Completion 类型)。"""
+        return self.provider_manager.inst_map.get(provider_id)
 
     def get_all_providers(self) -> List[Provider]:
         """获取所有用于文本生成任务的 LLM Provider(Chat_Completion 类型)。"""
@@ -144,24 +141,46 @@ class Context:
         """获取所有用于 STT 任务的 Provider。"""
         return self.provider_manager.stt_provider_insts
 
-    def get_using_provider(self) -> Provider:
+    def get_using_provider(self, umo: str = None) -> Provider:
         """
-        获取当前使用的用于文本生成任务的 LLM Provider(Chat_Completion 类型)。
+        获取当前使用的用于文本生成任务的 LLM Provider(Chat_Completion 类型)。通过 /provider 指令切换。
 
-        通过 /provider 指令切换。
+        Args:
+            umo(str): unified_message_origin 值，如果传入并且用户启用了提供商会话隔离，则使用该会话偏好的提供商。
         """
+        if umo and self._config["provider_settings"]["separate_provider"]:
+            perf = sp.get("session_provider_perf", {})
+            prov_id = perf.get(umo, {}).get(ProviderType.CHAT_COMPLETION.value, None)
+            if inst := self.provider_manager.inst_map.get(prov_id, None):
+                return inst
         return self.provider_manager.curr_provider_inst
 
-    def get_using_tts_provider(self) -> TTSProvider:
+    def get_using_tts_provider(self, umo: str = None) -> TTSProvider:
         """
         获取当前使用的用于 TTS 任务的 Provider。
+
+        Args:
+            umo(str): unified_message_origin 值，如果传入，则使用该会话偏好的提供商。
         """
+        if umo and self._config["provider_settings"]["separate_provider"]:
+            perf = sp.get("session_provider_perf", {})
+            prov_id = perf.get(umo, {}).get(ProviderType.TEXT_TO_SPEECH.value, None)
+            if inst := self.provider_manager.inst_map.get(prov_id, None):
+                return inst
         return self.provider_manager.curr_tts_provider_inst
 
-    def get_using_stt_provider(self) -> STTProvider:
+    def get_using_stt_provider(self, umo: str = None) -> STTProvider:
         """
         获取当前使用的用于 STT 任务的 Provider。
+
+        Args:
+            umo(str): unified_message_origin 值，如果传入，则使用该会话偏好的提供商。
         """
+        if umo and self._config["provider_settings"]["separate_provider"]:
+            perf = sp.get("session_provider_perf", {})
+            prov_id = perf.get(umo, {}).get(ProviderType.SPEECH_TO_TEXT.value, None)
+            if inst := self.provider_manager.inst_map.get(prov_id, None):
+                return inst
         return self.provider_manager.curr_stt_provider_inst
 
     def get_config(self) -> AstrBotConfig:
@@ -301,3 +320,12 @@ class Context:
         注册一个异步任务。
         """
         self._register_tasks.append(task)
+
+    def register_web_api(
+        self, route: str, view_handler: Awaitable, methods: list, desc: str
+    ):
+        for idx, api in enumerate(self.registered_web_apis):
+            if api[0] == route and methods == api[2]:
+                self.registered_web_apis[idx] = (route, view_handler, methods, desc)
+                return
+        self.registered_web_apis.append((route, view_handler, methods, desc))

astrbot/core/star/filter/command.py

@@ -8,6 +8,12 @@ from .custom_filter import CustomFilter
 from ..star_handler import StarHandlerMetadata
 
 
+class GreedyStr(str):
+    """标记指令完成其他参数接收后的所有剩余文本。"""
+
+    pass
+
+
 # 标准指令受到 wake_prefix 的制约。
 class CommandFilter(HandlerFilter):
     """标准指令过滤器"""
@@ -15,8 +21,8 @@ class CommandFilter(HandlerFilter):
     def __init__(
         self,
         command_name: str,
-        alias: set = None,
-        handler_md: StarHandlerMetadata = None,
+        alias: set | None = None,
+        handler_md: StarHandlerMetadata | None = None,
         parent_command_names: List[str] = [""],
     ):
         self.command_name = command_name
@@ -68,7 +74,22 @@ class CommandFilter(HandlerFilter):
     ) -> Dict[str, Any]:
         """将参数列表 params 根据 param_type 转换为参数字典。"""
         result = {}
-        for i, (param_name, param_type_or_default_val) in enumerate(param_type.items()):
+        param_items = list(param_type.items())
+        for i, (param_name, param_type_or_default_val) in enumerate(param_items):
+            is_greedy = param_type_or_default_val is GreedyStr
+
+            if is_greedy:
+                # GreedyStr 必须是最后一个参数
+                if i != len(param_items) - 1:
+                    raise ValueError(
+                        f"参数 '{param_name}' (GreedyStr) 必须是最后一个参数。"
+                    )
+
+                # 将剩余的所有部分合并成一个字符串
+                remaining_params = params[i:]
+                result[param_name] = " ".join(remaining_params)
+                break
+            # 没有 GreedyStr 的情况
             if i >= len(params):
                 if (
                     isinstance(param_type_or_default_val, Type)
@@ -92,6 +113,17 @@ class CommandFilter(HandlerFilter):
                     elif isinstance(param_type_or_default_val, str):
                         # 如果 param_type_or_default_val 是字符串，直接赋值
                         result[param_name] = params[i]
+                    elif isinstance(param_type_or_default_val, bool):
+                        # 处理布尔类型
+                        lower_param = str(params[i]).lower()
+                        if lower_param in ["true", "yes", "1"]:
+                            result[param_name] = True
+                        elif lower_param in ["false", "no", "0"]:
+                            result[param_name] = False
+                        else:
+                            raise ValueError(
+                                f"参数 {param_name} 必须是布尔值（true/false, yes/no, 1/0）。"
+                            )
                     elif isinstance(param_type_or_default_val, int):
                         result[param_name] = int(params[i])
                     elif isinstance(param_type_or_default_val, float):

astrbot/core/star/star_handler.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 import enum
-import heapq
 from dataclasses import dataclass, field
 from typing import Awaitable, List, Dict, TypeVar, Generic
 from .filter import HandlerFilter
@@ -8,100 +7,66 @@ from .star import star_map
 
 T = TypeVar("T", bound="StarHandlerMetadata")
 
-
 class StarHandlerRegistry(Generic[T]):
-    """用于存储所有的 Star Handler"""
-
-    star_handlers_map: Dict[str, StarHandlerMetadata] = {}
-    """用于快速查找。key 是 handler_full_name"""
-    _handlers = []
+    def __init__(self):
+        self.star_handlers_map: Dict[str, StarHandlerMetadata] = {}
+        self._handlers: List[StarHandlerMetadata] = []
 
     def append(self, handler: StarHandlerMetadata):
-        """添加一个 Handler"""
+        """添加一个 Handler，并保持按优先级有序"""
         if "priority" not in handler.extras_configs:
             handler.extras_configs["priority"] = 0
 
-        heapq.heappush(self._handlers, (-handler.extras_configs["priority"], handler))
         self.star_handlers_map[handler.handler_full_name] = handler
+        self._handlers.append(handler)
+        self._handlers.sort(key=lambda h: -h.extras_configs["priority"])
 
     def _print_handlers(self):
-        """打印所有的 Handler"""
-        for _, handler in self._handlers:
+        for handler in self._handlers:
             print(handler.handler_full_name)
 
     def get_handlers_by_event_type(
         self, event_type: EventType, only_activated=True, platform_id=None
     ) -> List[StarHandlerMetadata]:
-        """通过事件类型获取 Handler
-
-        Args:
-            event_type: 事件类型
-            only_activated: 是否只返回已激活的插件的处理器
-            platform_id: 平台ID，如果提供此参数，将过滤掉在此平台不兼容的处理器
-
-        Returns:
-            List[StarHandlerMetadata]: 处理器列表
-        """
         handlers = []
-        for _, handler in self._handlers:
+        for handler in self._handlers:
             if handler.event_type != event_type:
                 continue
-
-            # 只激活的插件处理器
             if only_activated:
                 plugin = star_map.get(handler.handler_module_path)
                 if not (plugin and plugin.activated):
                     continue
-
-            # 平台兼容性过滤
             if platform_id and event_type != EventType.OnAstrBotLoadedEvent:
                 if not handler.is_enabled_for_platform(platform_id):
                     continue
-
             handlers.append(handler)
-
         return handlers
 
     def get_handler_by_full_name(self, full_name: str) -> StarHandlerMetadata:
-        """通过 Handler 的全名获取 Handler"""
         return self.star_handlers_map.get(full_name, None)
 
     def get_handlers_by_module_name(
         self, module_name: str
     ) -> List[StarHandlerMetadata]:
-        """通过模块名获取 Handler"""
         return [
-            handler
-            for _, handler in self._handlers
+            handler for handler in self._handlers
             if handler.handler_module_path == module_name
         ]
 
     def clear(self):
-        """清空所有的 Handler"""
         self.star_handlers_map.clear()
         self._handlers.clear()
 
     def remove(self, handler: StarHandlerMetadata):
-        """删除一个 Handler"""
-        # self._handlers.remove(handler)
-        for i, h in enumerate(self._handlers):
-            if h[1] == handler:
-                self._handlers.pop(i)
-                break
-        try:
-            del self.star_handlers_map[handler.handler_full_name]
-        except KeyError:
-            pass
+        self.star_handlers_map.pop(handler.handler_full_name, None)
+        self._handlers = [h for h in self._handlers if h != handler]
 
     def __iter__(self):
-        """使 StarHandlerRegistry 支持迭代"""
-        return (handler for _, handler in self._handlers)
+        return iter(self._handlers)
 
     def __len__(self):
-        """返回 Handler 的数量"""
         return len(self._handlers)
 
-
 star_handlers_registry = StarHandlerRegistry()
 
 

astrbot/core/star/star_manager.py

@@ -37,6 +37,12 @@ except ImportError:
     if os.getenv("ASTRBOT_RELOAD", "0") == "1":
         logger.warning("未安装 watchfiles，无法实现插件的热重载。")
 
+try:
+    import nh3
+except ImportError:
+    logger.warning("未安装 nh3 库，无法清理插件 README.md 中的 HTML 标签。")
+    nh3 = None
+
 
 class PluginManager:
     def __init__(self, context: Context, config: AstrBotConfig):
@@ -140,11 +146,13 @@ class PluginManager:
                 if os.path.exists(os.path.join(path, d, "main.py")) or os.path.exists(
                     os.path.join(path, d, d + ".py")
                 ):
-                    modules.append({
-                        "pname": d,
-                        "module": module_str,
-                        "module_path": os.path.join(path, d, module_str),
-                    })
+                    modules.append(
+                        {
+                            "pname": d,
+                            "module": module_str,
+                            "module_path": os.path.join(path, d, module_str),
+                        }
+                    )
         return modules
 
     def _get_plugin_modules(self) -> List[dict]:
@@ -158,7 +166,7 @@ class PluginManager:
             plugins.extend(_p)
         return plugins
 
-    def _check_plugin_dept_update(self, target_plugin: str = None):
+    async def _check_plugin_dept_update(self, target_plugin: str = None):
         """检查插件的依赖
         如果 target_plugin 为 None，则检查所有插件的依赖
         """
@@ -177,7 +185,7 @@ class PluginManager:
                 pth = os.path.join(plugin_path, "requirements.txt")
                 logger.info(f"正在安装插件 {p} 所需的依赖库: {pth}")
                 try:
-                    pip_installer.install(requirements_path=pth)
+                    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: {str(e)}")
 
@@ -399,7 +407,7 @@ class PluginManager:
                     module = __import__(path, fromlist=[module_str])
                 except (ModuleNotFoundError, ImportError):
                     # 尝试安装依赖
-                    self._check_plugin_dept_update(target_plugin=root_dir_name)
+                    await self._check_plugin_dept_update(target_plugin=root_dir_name)
                     module = __import__(path, fromlist=[module_str])
                 except Exception as e:
                     logger.error(traceback.format_exc())
@@ -443,11 +451,11 @@ class PluginManager:
                             metadata.repo = metadata_yaml.repo
                     except Exception:
                         pass
-
+                    metadata.config = plugin_config
                     if path not in inactivated_plugins:
                         # 只有没有禁用插件时才实例化插件类
                         if plugin_config:
-                            metadata.config = plugin_config
+                            # metadata.config = plugin_config
                             try:
                                 metadata.star_cls = metadata.star_cls_type(
                                     context=self.context, config=plugin_config
@@ -634,16 +642,21 @@ class PluginManager:
         if not os.path.exists(readme_path):
             readme_path = os.path.join(plugin_path, "readme.md")
 
-        if os.path.exists(readme_path):
+        if os.path.exists(readme_path) and nh3:
             try:
                 with open(readme_path, "r", encoding="utf-8") as f:
                     readme_content = f.read()
+                cleaned_content = nh3.clean(readme_content)
             except Exception as e:
                 logger.warning(f"读取插件 {dir_name} 的 README.md 文件失败: {str(e)}")
 
         plugin_info = None
         if plugin:
-            plugin_info = {"repo": plugin.repo, "readme": readme_content}
+            plugin_info = {
+                "repo": plugin.repo,
+                "readme": cleaned_content,
+                "name": plugin.name,
+            }
 
         return plugin_info
 
@@ -838,6 +851,10 @@ class PluginManager:
 
         plugin_info = None
         if plugin:
-            plugin_info = {"repo": plugin.repo, "readme": readme_content}
+            plugin_info = {
+                "repo": plugin.repo,
+                "readme": readme_content,
+                "name": plugin.name,
+            }
 
         return plugin_info

astrbot/core/star/updator.py

@@ -18,7 +18,8 @@ class PluginUpdator(RepoZipUpdator):
         return self.plugin_store_path
 
     async def install(self, repo_url: str, proxy="") -> str:
-        repo_name = self.format_repo_name(repo_url)
+        _, repo_name, _ = self.parse_github_url(repo_url)
+        repo_name = self.format_name(repo_name)
         plugin_path = os.path.join(self.plugin_store_path, repo_name)
         await self.download_from_repo_url(plugin_path, repo_url, proxy)
         self.unzip_file(plugin_path + ".zip", plugin_path)
@@ -31,10 +32,6 @@ class PluginUpdator(RepoZipUpdator):
         if not repo_url:
             raise Exception(f"插件 {plugin.name} 没有指定仓库地址。")
 
-        if proxy:
-            proxy = proxy.removesuffix("/")
-            repo_url = f"{proxy}/{repo_url}"
-
         plugin_path = os.path.join(self.plugin_store_path, plugin.root_dir_name)
 
         logger.info(f"正在更新插件，路径: {plugin_path}，仓库地址: {repo_url}")
@@ -54,7 +51,7 @@ class PluginUpdator(RepoZipUpdator):
     def unzip_file(self, zip_path: str, target_dir: str):
         os.makedirs(target_dir, exist_ok=True)
         update_dir = ""
-        logger.info(f"解压文件: {zip_path}")
+        logger.info(f"正在解压压缩包: {zip_path}")
         with zipfile.ZipFile(zip_path, "r") as z:
             update_dir = z.namelist()[0]
             z.extractall(target_dir)

astrbot/core/utils/pip_installer.py

@@ -1,5 +1,6 @@
 import logging
-from pip import main as pip_main
+import asyncio
+import sys
 
 logger = logging.getLogger("astrbot")
 
@@ -9,7 +10,7 @@ class PipInstaller:
         self.pip_install_arg = pip_install_arg
         self.pypi_index_url = pypi_index_url
 
-    def install(
+    async def install(
         self,
         package_name: str = None,
         requirements_path: str = None,
@@ -29,12 +30,33 @@ class PipInstaller:
             args.extend(self.pip_install_arg.split())
 
         logger.info(f"Pip 包管理器: pip {' '.join(args)}")
+        try:
+            process = await asyncio.create_subprocess_exec(
+                sys.executable,
+                "-m",
+                "pip",
+                *args,
+                stdout=asyncio.subprocess.PIPE,
+                stderr=asyncio.subprocess.STDOUT,
+            )
 
-        result_code = pip_main(args)
+            assert process.stdout is not None
+            async for line in process.stdout:
+                logger.info(line.decode().strip())
 
-        # 清除 pip.main 导致的多余的 logging handlers
-        for handler in logging.root.handlers[:]:
-            logging.root.removeHandler(handler)
+            await process.wait()
 
-        if result_code != 0:
-            raise Exception(f"安装失败，错误码：{result_code}")
+            if process.returncode != 0:
+                raise Exception(f"安装失败，错误码：{process.returncode}")
+        except FileNotFoundError:
+            # 没有 pip
+            from pip import main as pip_main
+
+            result_code = await asyncio.to_thread(pip_main, args)
+
+            # 清除 pip.main 导致的多余的 logging handlers
+            for handler in logging.root.handlers[:]:
+                logging.root.removeHandler(handler)
+
+            if result_code != 0:
+                raise Exception(f"安装失败，错误码：{result_code}")

astrbot/core/utils/t2i/network_strategy.py

@@ -11,7 +11,7 @@ ASTRBOT_T2I_DEFAULT_ENDPOINT = "https://t2i.soulter.top/text2img"
 
 
 class NetworkRenderStrategy(RenderStrategy):
-    def __init__(self, base_url: str = ASTRBOT_T2I_DEFAULT_ENDPOINT) -> None:
+    def __init__(self, base_url: str | None = None) -> None:
         super().__init__()
         if not base_url:
             base_url = ASTRBOT_T2I_DEFAULT_ENDPOINT
@@ -34,18 +34,22 @@ class NetworkRenderStrategy(RenderStrategy):
             self.BASE_RENDER_URL += "/text2img"
 
     async def render_custom_template(
-        self, tmpl_str: str, tmpl_data: dict, return_url: bool = True
+        self,
+        tmpl_str: str,
+        tmpl_data: dict,
+        return_url: bool = True,
+        options: dict | None = None,
     ) -> str:
         """使用自定义文转图模板"""
+        default_options = {"full_page": True, "type": "jpeg", "quality": 40}
+        if options:
+            default_options |= options
+
         post_data = {
             "tmpl": tmpl_str,
             "json": return_url,
             "tmpldata": tmpl_data,
-            "options": {
-                "full_page": True,
-                "type": "jpeg",
-                "quality": 40,
-            },
+            "options": default_options,
         }
         if return_url:
             ssl_context = ssl.create_default_context(cafile=certifi.where())

astrbot/core/utils/t2i/renderer.py

@@ -6,7 +6,7 @@ logger = LogManager.GetLogger(log_name="astrbot")
 
 
 class HtmlRenderer:
-    def __init__(self, endpoint_url: str = None):
+    def __init__(self, endpoint_url: str | None = None):
         self.network_strategy = NetworkRenderStrategy(endpoint_url)
         self.local_strategy = LocalRenderStrategy()
 
@@ -16,19 +16,24 @@ class HtmlRenderer:
         self.network_strategy.set_endpoint(endpoint_url)
 
     async def render_custom_template(
-        self, tmpl_str: str, tmpl_data: dict, return_url: bool = False
+        self,
+        tmpl_str: str,
+        tmpl_data: dict,
+        return_url: bool = False,
+        options: dict | None = None,
     ):
         """使用自定义文转图模板。该方法会通过网络调用 t2i 终结点图文渲染API。
         @param tmpl_str: HTML Jinja2 模板。
         @param tmpl_data: jinja2 模板数据。
+        @param options: 渲染选项。
 
         @return: 图片 URL 或者文件路径，取决于 return_url 参数。
 
         @example: 参见 https://astrbot.app 插件开发部分。
         """
-        local = locals()
-        local.pop("self")
-        return await self.network_strategy.render_custom_template(**local)
+        return await self.network_strategy.render_custom_template(
+            tmpl_str, tmpl_data, return_url, options
+        )
 
     async def render_t2i(
         self, text: str, use_network: bool = True, return_url: bool = False

astrbot/core/utils/tencent_record_helper.py

@@ -1,5 +1,12 @@
+import base64
 import wave
+import os
+import subprocess
 from io import BytesIO
+import asyncio
+import tempfile
+from astrbot.core import logger
+from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 
 
 async def tencent_silk_to_wav(silk_path: str, output_path: str) -> str:
@@ -50,3 +57,104 @@ async def wav_to_tencent_silk(wav_path: str, output_path: str) -> int:
         rate = wav.getframerate()
         duration = pilk.encode(wav_path, output_path, pcm_rate=rate, tencent=True)
         return duration
+
+
+async def convert_to_pcm_wav(input_path: str, output_path: str) -> str:
+    """
+    将 MP3 或其他音频格式转换为 PCM 16bit WAV，采样率24000Hz，单声道。
+    若转换失败则抛出异常。
+    """
+    try:
+        from pyffmpeg import FFmpeg
+
+        ff = FFmpeg()
+        ff.convert(input=input_path, output=output_path)
+    except Exception as e:
+        logger.debug(f"pyffmpeg 转换失败: {e}, 尝试使用 ffmpeg 命令行进行转换")
+
+        p = await asyncio.create_subprocess_exec(
+            "ffmpeg",
+            "-y",
+            "-i",
+            input_path,
+            "-acodec",
+            "pcm_s16le",
+            "-ar",
+            "24000",
+            "-ac",
+            "1",
+            "-af",
+            "apad=pad_dur=2",
+            "-fflags",
+            "+genpts",
+            "-hide_banner",
+            output_path,
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+        )
+        stdout, stderr = await p.communicate()
+        logger.info(f"[FFmpeg] stdout: {stdout.decode().strip()}")
+        logger.debug(f"[FFmpeg] stderr: {stderr.decode().strip()}")
+        logger.info(f"[FFmpeg] return code: {p.returncode}")
+
+    if os.path.exists(output_path) and os.path.getsize(output_path) > 0:
+        return output_path
+    else:
+        raise RuntimeError("生成的WAV文件不存在或为空")
+
+
+async def audio_to_tencent_silk_base64(audio_path: str) -> tuple[str, float]:
+    """
+    将 MP3/WAV 文件转为 Tencent Silk 并返回 base64 编码与时长（秒）。
+
+    参数:
+    - audio_path: 输入音频文件路径（.mp3 或 .wav）
+
+    返回:
+    - silk_b64: Base64 编码的 Silk 字符串
+    - duration: 音频时长（秒）
+    """
+    try:
+        import pilk
+    except ImportError as e:
+        raise Exception("未安装 pysilk，请执行: pip install pysilk") from e
+
+    temp_dir = os.path.join(get_astrbot_data_path(), "temp")
+    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(
+        suffix=".wav", delete=False, dir=temp_dir
+    ).name
+
+    if ext != ".wav":
+        await convert_to_pcm_wav(audio_path, temp_wav)
+        # 删除原文件
+        os.remove(audio_path)
+        wav_path = temp_wav
+    else:
+        wav_path = audio_path
+
+    with wave.open(wav_path, "rb") as wav_file:
+        rate = wav_file.getframerate()
+
+    silk_path = tempfile.NamedTemporaryFile(
+        suffix=".silk", delete=False, dir=temp_dir
+    ).name
+
+    try:
+        duration = await asyncio.to_thread(
+            pilk.encode, wav_path, silk_path, pcm_rate=rate, tencent=True
+        )
+
+        with open(silk_path, "rb") as f:
+            silk_bytes = await asyncio.to_thread(f.read)
+            silk_b64 = base64.b64encode(silk_bytes).decode("utf-8")
+
+        return silk_b64, duration  # 已是秒
+    finally:
+        if os.path.exists(wav_path) and wav_path != audio_path:
+            os.remove(wav_path)
+        if os.path.exists(silk_path):
+            os.remove(silk_path)

astrbot/core/zip_updator.py

@@ -1,5 +1,6 @@
 import aiohttp
 import os
+import re
 import zipfile
 import shutil
 
@@ -119,28 +120,61 @@ class RepoZipUpdator:
         )
 
     async def download_from_repo_url(self, target_path: str, repo_url: str, proxy=""):
-        repo_namespace = repo_url.split("/")[-2:]
-        author = repo_namespace[0]
-        repo = repo_namespace[1]
+        author, repo, branch = self.parse_github_url(repo_url)
 
         logger.info(f"正在下载更新 {repo} ...")
-        release_url = f"https://api.github.com/repos/{author}/{repo}/releases"
-        releases = await self.fetch_release_info(url=release_url)
-        if not releases:
-            # download from the default branch directly.
-            logger.info(f"正在从默认分支下载 {author}/{repo} ")
+
+        if branch:
+            logger.info(f"正在从指定分支 {branch} 下载 {author}/{repo}")
             release_url = (
-                f"https://github.com/{author}/{repo}/archive/refs/heads/master.zip"
+                f"https://github.com/{author}/{repo}/archive/refs/heads/{branch}.zip"
             )
         else:
-            release_url = releases[0]["zipball_url"]
+            try:
+                release_url = f"https://api.github.com/repos/{author}/{repo}/releases"
+                releases = await self.fetch_release_info(url=release_url)
+            except Exception as e:
+                logger.warning(
+                    f"获取 {author}/{repo} 的 GitHub Releases 失败: {e}，将尝试下载默认分支"
+                )
+                releases = []
+            if not releases:
+                # 如果没有最新版本，下载默认分支
+                logger.info(f"正在从默认分支下载 {author}/{repo}")
+                release_url = (
+                    f"https://github.com/{author}/{repo}/archive/refs/heads/master.zip"
+                )
+            else:
+                release_url = releases[0]["zipball_url"]
 
         if proxy:
+            proxy = proxy.rstrip("/")
             release_url = f"{proxy}/{release_url}"
-            logger.info(f"使用代理下载: {release_url}")
+            logger.info(
+                f"检查到设置了镜像站，将使用镜像站下载 {author}/{repo} 仓库源码: {release_url}"
+            )
 
         await download_file(release_url, target_path + ".zip")
 
+    def parse_github_url(self, url: str):
+        """使用正则表达式解析 GitHub 仓库 URL，支持 `.git` 后缀和 `tree/branch` 结构
+        Returns:
+            tuple[str, str, str]: 返回作者名、仓库名和分支名
+        Raises:
+            ValueError: 如果 URL 格式不正确
+        """
+        cleaned_url = url.rstrip("/")
+        pattern = r"^https://github\.com/([a-zA-Z0-9_-]+)/([a-zA-Z0-9_-]+)(\.git)?(?:/tree/([a-zA-Z0-9_-]+))?$"
+        match = re.match(pattern, cleaned_url)
+
+        if match:
+            author = match.group(1)
+            repo = match.group(2)
+            branch = match.group(4)
+            return author, repo, branch
+        else:
+            raise ValueError("无效的 GitHub URL")
+
     def unzip_file(self, zip_path: str, target_dir: str):
         """
         解压缩文件, 并将压缩包内**第一个**文件夹内的文件移动到 target_dir
@@ -174,16 +208,5 @@ class RepoZipUpdator:
                 f"删除更新文件失败，可以手动删除 {zip_path} 和 {os.path.join(target_dir, update_dir)}"
             )
 
-    def format_repo_name(self, repo_url: str) -> str:
-        if repo_url.endswith("/"):
-            repo_url = repo_url[:-1]
-
-        repo_namespace = repo_url.split("/")[-2:]
-        repo = repo_namespace[1]
-
-        repo = self.format_name(repo)
-
-        return repo
-
     def format_name(self, name: str) -> str:
         return name.replace("-", "_").lower()

astrbot/dashboard/routes/auth.py

@@ -1,8 +1,9 @@
 import jwt
 import datetime
+import asyncio
 from .route import Route, Response, RouteContext
 from quart import request
-from astrbot.core import WEBUI_SK, DEMO_MODE
+from astrbot.core import DEMO_MODE
 from astrbot import logger
 
 
@@ -21,7 +22,11 @@ class AuthRoute(Route):
         post_data = await request.json
         if post_data["username"] == username and post_data["password"] == password:
             change_pwd_hint = False
-            if username == "astrbot" and password == "77b90590a8945a7d36c963981a307dc9":
+            if (
+                username == "astrbot"
+                and password == "77b90590a8945a7d36c963981a307dc9"
+                and not DEMO_MODE
+            ):
                 change_pwd_hint = True
                 logger.warning("为了保证安全，请尽快修改默认密码。")
 
@@ -37,6 +42,7 @@ class AuthRoute(Route):
                 .__dict__
             )
         else:
+            await asyncio.sleep(3)
             return Response().error("用户名或密码错误").__dict__
 
     async def edit_account(self):
@@ -72,7 +78,10 @@ class AuthRoute(Route):
     def generate_jwt(self, username):
         payload = {
             "username": username,
-            "exp": datetime.datetime.utcnow() + datetime.timedelta(days=30),
+            "exp": datetime.datetime.utcnow() + datetime.timedelta(days=7),
         }
-        token = jwt.encode(payload, WEBUI_SK, algorithm="HS256")
+        jwt_token = self.config["dashboard"].get("jwt_secret", None)
+        if not jwt_token:
+            raise ValueError("JWT secret is not set in the cmd_config.")
+        token = jwt.encode(payload, jwt_token, algorithm="HS256")
         return token

astrbot/dashboard/routes/chat.py

@@ -26,6 +26,7 @@ class ChatRoute(Route):
             "/chat/conversations": ("GET", self.get_conversations),
             "/chat/get_conversation": ("GET", self.get_conversation),
             "/chat/delete_conversation": ("GET", self.delete_conversation),
+            "/chat/rename_conversation": ("POST", self.rename_conversation),
             "/chat/get_file": ("GET", self.get_file),
             "/chat/post_image": ("POST", self.post_image),
             "/chat/post_file": ("POST", self.post_file),
@@ -61,16 +62,25 @@ class ChatRoute(Route):
             return Response().error("Missing key: filename").__dict__
 
         try:
-            with open(os.path.join(self.imgs_dir, filename), "rb") as f:
-                if filename.endswith(".wav"):
+            file_path = os.path.join(self.imgs_dir, os.path.basename(filename))
+            real_file_path = os.path.realpath(file_path)
+            real_imgs_dir = os.path.realpath(self.imgs_dir)
+
+            if not real_file_path.startswith(real_imgs_dir):
+                return Response().error("Invalid file path").__dict__
+
+            with open(real_file_path, "rb") as f:
+                filename_ext = os.path.splitext(filename)[1].lower()
+
+                if filename_ext == ".wav":
                     return QuartResponse(f.read(), mimetype="audio/wav")
-                elif filename.split(".")[-1] in self.supported_imgs:
+                elif filename_ext[1:] in self.supported_imgs:
                     return QuartResponse(f.read(), mimetype="image/jpeg")
                 else:
                     return QuartResponse(f.read())
 
-        except FileNotFoundError:
-            return Response().error("File not found").__dict__
+        except (FileNotFoundError, OSError):
+            return Response().error("File access error").__dict__
 
     async def post_image(self):
         post_data = await request.files
@@ -91,7 +101,6 @@ class ChatRoute(Route):
 
         file = post_data["file"]
         filename = f"{str(uuid.uuid4())}"
-        print(file)
         # 通过文件格式判断文件类型
         if file.content_type.startswith("audio"):
             filename += ".wav"
@@ -143,7 +152,7 @@ class ChatRoute(Route):
         try:
             history = json.loads(conversation.history)
         except BaseException as e:
-            print(e)
+            logger.error(f"Failed to parse conversation history: {e}")
             history = []
         new_his = {"type": "user", "message": message}
         if image_url:
@@ -197,6 +206,9 @@ class ChatRoute(Route):
                     if streaming and type != "end":
                         continue
 
+                    if type == "update_title":
+                        continue
+
                     if result_text:
                         conversation = self.db.get_conversation_by_user_id(
                             username, cid
@@ -204,7 +216,7 @@ class ChatRoute(Route):
                         try:
                             history = json.loads(conversation.history)
                         except BaseException as e:
-                            print(e)
+                            logger.error(f"Failed to parse conversation history: {e}")
                             history = []
                         history.append({"type": "bot", "message": result_text})
                         self.db.update_conversation(
@@ -242,6 +254,18 @@ class ChatRoute(Route):
         self.db.new_conversation(username, conversation_id)
         return Response().ok(data={"conversation_id": conversation_id}).__dict__
 
+    async def rename_conversation(self):
+        username = g.get("username", "guest")
+        post_data = await request.json
+        if "conversation_id" not in post_data or "title" not in post_data:
+            return Response().error("Missing key: conversation_id or title").__dict__
+
+        conversation_id = post_data["conversation_id"]
+        title = post_data["title"]
+
+        self.db.update_conversation_title(username, conversation_id, title=title)
+        return Response().ok(message="重命名成功！").__dict__
+
     async def get_conversations(self):
         username = g.get("username", "guest")
         conversations = self.db.get_conversations(username)

astrbot/dashboard/routes/config.py

@@ -9,6 +9,7 @@ from astrbot.core.platform.register import platform_registry
 from astrbot.core.provider.register import provider_registry
 from astrbot.core.star.star import star_registry
 from astrbot.core import logger
+import asyncio
 
 
 def try_cast(value: str, type_: str):
@@ -153,6 +154,7 @@ class ConfigRoute(Route):
     ) -> None:
         super().__init__(context)
         self.core_lifecycle = core_lifecycle
+        self.config: AstrBotConfig = core_lifecycle.astrbot_config
         self.routes = {
             "/config/get": ("GET", self.get_configs),
             "/config/astrbot/update": ("POST", self.post_astrbot_configs),
@@ -164,9 +166,125 @@ class ConfigRoute(Route):
             "/config/provider/update": ("POST", self.post_update_provider),
             "/config/provider/delete": ("POST", self.post_delete_provider),
             "/config/llmtools": ("GET", self.get_llm_tools),
+            "/config/provider/check_status": ("GET", self.check_all_providers_status),
+            "/config/provider/list": ("GET", self.get_provider_config_list),
+            "/config/provider/get_session_seperate": (
+                "GET",
+                lambda: Response()
+                .ok({"enable": self.config["provider_settings"]["separate_provider"]})
+                .__dict__,
+            ),
+            "/config/provider/set_session_seperate": (
+                "POST",
+                self.post_session_seperate,
+            ),
         }
         self.register_routes()
 
+    async def _test_single_provider(self, provider):
+        """辅助函数：测试单个 provider 的可用性"""
+        meta = provider.meta()
+        provider_name = provider.provider_config.get("id", "Unknown Provider")
+        logger.debug(f"Got provider meta: {meta}")
+        if not provider_name and meta:
+            provider_name = meta.id
+        elif not provider_name:
+            provider_name = "Unknown Provider"
+        status_info = {
+            "id": getattr(meta, "id", "Unknown ID"),
+            "model": getattr(meta, "model", "Unknown Model"),
+            "type": getattr(meta, "type", "Unknown Type"),
+            "name": provider_name,
+            "status": "unavailable",  # 默认为不可用
+            "error": None,
+        }
+        logger.debug(
+            f"Attempting to check provider: {status_info['name']} (ID: {status_info['id']}, Type: {status_info['type']}, Model: {status_info['model']})"
+        )
+        try:
+            logger.debug(f"Sending 'Ping' to provider: {status_info['name']}")
+            response = await asyncio.wait_for(
+                provider.text_chat(prompt="REPLY `PONG` ONLY"), timeout=45.0
+            )
+            logger.debug(f"Received response from {status_info['name']}: {response}")
+            # 只要 text_chat 调用成功返回一个 LLMResponse 对象 (即 response 不为 None)，就认为可用
+            if response is not None:
+                status_info["status"] = "available"
+                response_text_snippet = ""
+                if hasattr(response, "completion_text") and response.completion_text:
+                    response_text_snippet = (
+                        response.completion_text[:70] + "..."
+                        if len(response.completion_text) > 70
+                        else response.completion_text
+                    )
+                elif hasattr(response, "result_chain") and response.result_chain:
+                    try:
+                        response_text_snippet = (
+                            response.result_chain.get_plain_text()[:70] + "..."
+                            if len(response.result_chain.get_plain_text()) > 70
+                            else response.result_chain.get_plain_text()
+                        )
+                    except Exception as _:
+                        pass
+                logger.info(
+                    f"Provider {status_info['name']} (ID: {status_info['id']}) is available. Response snippet: '{response_text_snippet}'"
+                )
+            else:
+                # 这个分支理论上不应该被走到，除非 text_chat 实现可能返回 None
+                status_info["error"] = (
+                    "Test call returned None, but expected an LLMResponse object."
+                )
+                logger.warning(
+                    f"Provider {status_info['name']} (ID: {status_info['id']}) test call returned None."
+                )
+
+        except asyncio.TimeoutError:
+            status_info["error"] = (
+                "Connection timed out after 45 seconds during test call."
+            )
+            logger.warning(
+                f"Provider {status_info['name']} (ID: {status_info['id']}) timed out."
+            )
+        except Exception as e:
+            error_message = str(e)
+            status_info["error"] = error_message
+            logger.warning(
+                f"Provider {status_info['name']} (ID: {status_info['id']}) is unavailable. Error: {error_message}"
+            )
+            logger.debug(
+                f"Traceback for {status_info['name']}:\n{traceback.format_exc()}"
+            )
+        return status_info
+
+    async def check_all_providers_status(self):
+        """
+        API 接口: 检查所有 LLM Providers 的状态
+        """
+        logger.info("API call received: /config/provider/check_status")
+        try:
+            all_providers: typing.List = (
+                self.core_lifecycle.star_context.get_all_providers()
+            )
+            logger.debug(f"Found {len(all_providers)} providers to check.")
+
+            if not all_providers:
+                logger.info("No providers found to check.")
+                return Response().ok([]).__dict__
+
+            tasks = [self._test_single_provider(p) for p in all_providers]
+            logger.debug(f"Created {len(tasks)} tasks for concurrent provider checks.")
+
+            results = await asyncio.gather(*tasks)
+            logger.info(f"Provider status check completed. Results: {results}")
+
+            return Response().ok(results).__dict__
+        except Exception as e:
+            logger.error(f"Critical error in check_all_providers_status: {str(e)}")
+            logger.error(traceback.format_exc())
+            return (
+                Response().error(f"检查 Provider 状态时发生严重错误: {str(e)}").__dict__
+            )
+
     async def get_configs(self):
         # plugin_name 为空时返回 AstrBot 配置
         # 否则返回指定 plugin_name 的插件配置
@@ -175,6 +293,32 @@ class ConfigRoute(Route):
             return Response().ok(await self._get_astrbot_config()).__dict__
         return Response().ok(await self._get_plugin_config(plugin_name)).__dict__
 
+    async def post_session_seperate(self):
+        """设置提供商会话隔离"""
+        post_config = await request.json
+        enable = post_config.get("enable", None)
+        if enable is None:
+            return Response().error("缺少参数 enable").__dict__
+
+        astrbot_config = self.core_lifecycle.astrbot_config
+        astrbot_config["provider_settings"]["separate_provider"] = enable
+        try:
+            astrbot_config.save_config()
+        except Exception as e:
+            return Response().error(str(e)).__dict__
+        return Response().ok(None, "设置成功~").__dict__
+
+    async def get_provider_config_list(self):
+        provider_type = request.args.get("provider_type", None)
+        if not provider_type:
+            return Response().error("缺少参数 provider_type").__dict__
+        provider_list = []
+        astrbot_config = self.core_lifecycle.astrbot_config
+        for provider in astrbot_config["provider"]:
+            if provider.get("provider_type", None) == provider_type:
+                provider_list.append(provider)
+        return Response().ok(provider_list).__dict__
+
     async def post_astrbot_configs(self):
         post_configs = await request.json
         try:

astrbot/dashboard/routes/log.py

@@ -23,6 +23,7 @@ class LogRoute(Route):
                         **message,  # see astrbot/core/log.py
                     }
                     yield f"data: {json.dumps(payload, ensure_ascii=False)}\n\n"
+                    await asyncio.sleep(0.07)  # 控制发送频率，避免过快
             except asyncio.CancelledError:
                 pass
             except BaseException as e:

astrbot/dashboard/routes/plugin.py

@@ -18,6 +18,12 @@ from astrbot.core.star.filter.regex import RegexFilter
 from astrbot.core.star.star_handler import EventType
 from astrbot.core import DEMO_MODE
 
+try:
+    import nh3
+except ImportError:
+    logger.warning("未安装 nh3 库，无法清理插件 README.md 中的 HTML 标签。")
+    nh3 = None
+
 
 class PluginRoute(Route):
     def __init__(
@@ -102,7 +108,10 @@ class PluginRoute(Route):
 
     async def get_plugins(self):
         _plugin_resp = []
+        plugin_name = request.args.get("name")
         for plugin in self.plugin_manager.context.get_all_stars():
+            if plugin_name and plugin.name != plugin_name:
+                continue
             _t = {
                 "name": plugin.name,
                 "repo": "" if plugin.repo is None else plugin.repo,
@@ -145,9 +154,7 @@ class PluginRoute(Route):
             if handler.event_type == EventType.AdapterMessageEvent:
                 # 处理平台适配器消息事件
                 has_admin = False
-                for (
-                    filter
-                ) in (
+                for filter in (
                     handler.event_filters
                 ):  # 正常handler就只有 1~2 个 filter，因此这里时间复杂度不会太高
                     if isinstance(filter, CommandFilter):
@@ -325,6 +332,9 @@ class PluginRoute(Route):
             return Response().error(str(e)).__dict__
 
     async def get_plugin_readme(self):
+        if not nh3:
+            return Response().error("未安装 nh3 库").__dict__
+
         plugin_name = request.args.get("name")
         logger.debug(f"正在获取插件 {plugin_name} 的README文件内容")
 
@@ -360,9 +370,11 @@ class PluginRoute(Route):
             with open(readme_path, "r", encoding="utf-8") as f:
                 readme_content = f.read()
 
+            cleaned_content = nh3.clean(readme_content)
+
             return (
                 Response()
-                .ok({"content": readme_content}, "成功获取README内容")
+                .ok({"content": cleaned_content}, "成功获取README内容")
                 .__dict__
             )
         except Exception as e:
@@ -383,14 +395,12 @@ class PluginRoute(Route):
                 platform_type = platform.get("type", "")
                 platform_id = platform.get("id", "")
 
-                platforms.append(
-                    {
-                        "name": platform_id,  # 使用type作为name，这是系统内部使用的平台名称
-                        "id": platform_id,  # 保留id字段以便前端可以显示
-                        "type": platform_type,
-                        "display_name": f"{platform_type}({platform_id})",
-                    }
-                )
+                platforms.append({
+                    "name": platform_id,  # 使用type作为name，这是系统内部使用的平台名称
+                    "id": platform_id,  # 保留id字段以便前端可以显示
+                    "type": platform_type,
+                    "display_name": f"{platform_type}({platform_id})",
+                })
 
             adjusted_platform_enable = {}
             for platform_id, plugins in platform_enable.items():
@@ -399,13 +409,11 @@ class PluginRoute(Route):
             # 获取所有插件，包括系统内部插件
             plugins = []
             for plugin in self.plugin_manager.context.get_all_stars():
-                plugins.append(
-                    {
-                        "name": plugin.name,
-                        "desc": plugin.desc,
-                        "reserved": plugin.reserved,  # 添加reserved标志
-                    }
-                )
+                plugins.append({
+                    "name": plugin.name,
+                    "desc": plugin.desc,
+                    "reserved": plugin.reserved,  # 添加reserved标志
+                })
 
             logger.debug(
                 f"获取插件平台配置: 原始配置={platform_enable}, 调整后={adjusted_platform_enable}"
@@ -413,13 +421,11 @@ class PluginRoute(Route):
 
             return (
                 Response()
-                .ok(
-                    {
-                        "platforms": platforms,
-                        "plugins": plugins,
-                        "platform_enable": adjusted_platform_enable,
-                    }
-                )
+                .ok({
+                    "platforms": platforms,
+                    "plugins": plugins,
+                    "platform_enable": adjusted_platform_enable,
+                })
                 .__dict__
             )
         except Exception as e:

astrbot/dashboard/routes/stat.py

@@ -8,6 +8,7 @@ from quart import request
 from astrbot.core.core_lifecycle import AstrBotCoreLifecycle
 from astrbot.core.db import BaseDatabase
 from astrbot.core.config import VERSION
+from astrbot.core.utils.io import get_dashboard_version
 from astrbot.core import DEMO_MODE
 
 
@@ -40,13 +41,37 @@ class StatRoute(Route):
         await self.core_lifecycle.restart()
         return Response().ok().__dict__
 
-    def format_sec(self, sec: int):
-        m, s = divmod(sec, 60)
-        h, m = divmod(m, 60)
-        return f"{h}小时{m}分{s}秒"
+    def _get_running_time_components(self, total_seconds: int):
+        """将总秒数转换为时分秒组件"""
+        minutes, seconds = divmod(total_seconds, 60)
+        hours, minutes = divmod(minutes, 60)
+        return {
+            "hours": hours,
+            "minutes": minutes,
+            "seconds": seconds
+        }
+
+    def is_default_cred(self):
+        username = self.config["dashboard"]["username"]
+        password = self.config["dashboard"]["password"]
+        return (
+            username == "astrbot"
+            and password == "77b90590a8945a7d36c963981a307dc9"
+            and not DEMO_MODE
+        )
 
     async def get_version(self):
-        return Response().ok({"version": VERSION}).__dict__
+        return (
+            Response()
+            .ok(
+                {
+                    "version": VERSION,
+                    "dashboard_version": await get_dashboard_version(),
+                    "change_pwd_hint": self.is_default_cred(),
+                }
+            )
+            .__dict__
+        )
 
     async def get_start_time(self):
         return Response().ok({"start_time": self.core_lifecycle.start_time}).__dict__
@@ -87,6 +112,11 @@ class StatRoute(Route):
                 }
                 plugin_info.append(info)
 
+            # 计算运行时长组件
+            running_time = self._get_running_time_components(
+                int(time.time()) - self.core_lifecycle.start_time
+            )
+
             stat_dict.update(
                 {
                     "platform": self.db_helper.get_grouped_base_stats(
@@ -99,9 +129,7 @@ class StatRoute(Route):
                     "plugin_count": len(plugins),
                     "plugins": plugin_info,
                     "message_time_series": message_time_based_stats,
-                    "running": self.format_sec(
-                        int(time.time()) - self.core_lifecycle.start_time
-                    ),
+                    "running": running_time,  # 现在返回时间组件而不是格式化的字符串
                     "memory": {
                         "process": psutil.Process().memory_info().rss >> 20,
                         "system": psutil.virtual_memory().total >> 20,

astrbot/dashboard/routes/static_file.py

@@ -12,7 +12,10 @@ class StaticFileRoute(Route):
             "/logs",
             "/extension",
             "/dashboard/default",
-            "/project-atri",
+            "/alkaid",
+            "/alkaid/knowledge-base",
+            "/alkaid/long-term-memory",
+            "/alkaid/other",
             "/console",
             "/chat",
             "/settings",

astrbot/dashboard/routes/update.py

@@ -91,7 +91,7 @@ class UpdateRoute(Route):
             # pip 更新依赖
             logger.info("更新依赖中...")
             try:
-                pip_installer.install(requirements_path="requirements.txt")
+                await pip_installer.install(requirements_path="requirements.txt")
             except Exception as e:
                 logger.error(f"更新依赖失败: {e}")
 
@@ -140,7 +140,7 @@ class UpdateRoute(Route):
         if not package:
             return Response().error("缺少参数 package 或不合法。").__dict__
         try:
-            pip_installer.install(package, mirror=mirror)
+            await pip_installer.install(package, mirror=mirror)
             return Response().ok(None, "安装成功。").__dict__
         except Exception as e:
             logger.error(f"/api/update_pip: {traceback.format_exc()}")

astrbot/dashboard/server.py

@@ -10,11 +10,13 @@ from quart.logging import default_handler
 from astrbot.core.core_lifecycle import AstrBotCoreLifecycle
 from .routes import *
 from .routes.route import RouteContext, Response
-from astrbot.core import logger, WEBUI_SK
+from astrbot.core import logger
 from astrbot.core.db import BaseDatabase
 from astrbot.core.utils.io import get_local_ip_addresses
 from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 
+APP: Quart = None
+
 
 class AstrBotDashboard:
     def __init__(
@@ -27,6 +29,7 @@ class AstrBotDashboard:
         self.config = core_lifecycle.astrbot_config
         self.data_path = os.path.abspath(os.path.join(get_astrbot_data_path(), "dist"))
         self.app = Quart("dashboard", static_folder=self.data_path, static_url_path="/")
+        APP = self.app  # noqa
         self.app.config["MAX_CONTENT_LENGTH"] = (
             128 * 1024 * 1024
         )  # 将 Flask 允许的最大上传文件体大小设置为 128 MB
@@ -51,12 +54,31 @@ class AstrBotDashboard:
         self.conversation_route = ConversationRoute(self.context, db, core_lifecycle)
         self.file_route = FileRoute(self.context)
 
+        self.app.add_url_rule(
+            "/api/plug/<path:subpath>",
+            view_func=self.srv_plug_route,
+            methods=["GET", "POST"],
+        )
+
         self.shutdown_event = shutdown_event
 
+        self._init_jwt_secret()
+
+    async def srv_plug_route(self, subpath, *args, **kwargs):
+        """
+        插件路由
+        """
+        registered_web_apis = self.core_lifecycle.star_context.registered_web_apis
+        for api in registered_web_apis:
+            route, view_handler, methods, _ = api
+            if route == f"/{subpath}" and request.method in methods:
+                return await view_handler(*args, **kwargs)
+        return jsonify(Response().error("未找到该路由").__dict__)
+
     async def auth_middleware(self):
         if not request.path.startswith("/api"):
             return
-        allowed_endpoints = ["/api/auth/login", "/api/chat/get_file", "/api/file"]
+        allowed_endpoints = ["/api/auth/login", "/api/file"]
         if any(request.path.startswith(prefix) for prefix in allowed_endpoints):
             return
         # claim jwt
@@ -68,7 +90,7 @@ class AstrBotDashboard:
         if token.startswith("Bearer "):
             token = token[7:]
         try:
-            payload = jwt.decode(token, WEBUI_SK, algorithms=["HS256"])
+            payload = jwt.decode(token, self._jwt_secret, algorithms=["HS256"])
             g.username = payload["username"]
         except jwt.ExpiredSignatureError:
             r = jsonify(Response().error("Token 过期").__dict__)
@@ -120,6 +142,15 @@ class AstrBotDashboard:
         except Exception as e:
             return f"获取进程信息失败: {str(e)}"
 
+    def _init_jwt_secret(self):
+        if not self.config.get("dashboard", {}).get("jwt_secret", None):
+            # 如果没有设置 JWT 密钥，则生成一个新的密钥
+            jwt_secret = os.urandom(32).hex()
+            self.config["dashboard"]["jwt_secret"] = jwt_secret
+            self.config.save_config()
+            logger.info("Initialized random JWT secret for dashboard.")
+        self._jwt_secret = self.config["dashboard"]["jwt_secret"]
+
     def run(self):
         ip_addr = []
         if p := os.environ.get("DASHBOARD_PORT"):

changelogs/v3.5.11.md

@@ -0,0 +1,7 @@
+# What's Changed
+
+1. 新增：火山引擎 TTS
+2. 修复：修复了 WeChatPadPro 在重新登录时为新设备的问题
+2. ‼️修复：微信公众号（个人认证或者未认证）的情况下能接收但无法回复消息的问题
+3. 修复：Minimax TTS 相关问题
+4. 优化：登录界面侧边栏、关于页面样式，修复如果此前已经登录但未自行跳转的问题

changelogs/v3.5.12.md

@@ -0,0 +1,18 @@
+# What's Changed
+
+1. 新增：支持 MCP 的 Streamable HTTP 传输方式。详见 [#1637](https://github.com/Soulter/AstrBot/issues/1637)
+2. 新增：支持 MCP 的 SSE 传输方式的自定义请求头。详见 [#1659](https://github.com/Soulter/AstrBot/issues/1659)
+3. 优化：将 /llm 和 /model 和 /provider 指令设置为管理员指令
+4. 修复：修复插件的 priority 部分失效的问题
+5. 修复：修复 QQ 下合并转发消息内无法发送文件等问题，尽可能修复了各种文件、语音、视频、图片无法发送的问题
+6. 优化：Telegram 支持长消息分段发送，优化消息编辑的逻辑
+7. 优化：WebUI 强制默认修改密码
+8. 优化：移除了 vpet
+9. 新增：插件接口：支持动态路由注册
+10. 优化：CLI 模式下的插件下载
+11. 新增：WeChatPadPro 对接获取联系人接口
+12. 新增：T2I、语音、视频支持文件服务
+13. 优化：硅基流动下某些工具调用返回的 argument 格式适配
+14. 优化：在使用 /llm 指令关闭后重启 AstrBot 后，模型提供商未被加载
+15. 新增：新增基于 FAISS + SQLite 的向量存储接口
+16. 新增：Alkaid Page

changelogs/v3.5.13.md

@@ -0,0 +1,9 @@
+# What's Changed
+
+1. 新增：WebUI 支持暗夜模式。
+2. 修复：修复 WebUI Chat 接口的未授权访问安全漏洞、插件 README 可能存在的 XSS 注入漏洞。
+3. 优化：优化 Vec DB 在 indexing 过程时的数据库事务处理。
+4. 修复：WebUI 下，插件市场的推荐卡片无法点击帮助文档的问题。
+5. 新增：知识库。
+6. 新增：WebUI 提供商测试功能，一键检测可用性。
+7. 新增：WebUI 提供商分类功能，按能力分类提供商。