chore: bump version to 4.22.3

* feat(provider/vllm_rerank): add configurable rerank_api_suffix option

Add rerank_api_suffix config option to the VLLM Rerank provider so
users can control the API URL path suffix instead of having /v1/rerank
hardcoded.

- Default value is /v1/rerank (preserves existing behavior)
- Users can set it to empty string to disable auto-append
- Handles suffix without leading slash by auto-adding one
- Schema, default config, and i18n metadata all updated

Issue: Fixes #7238

* fix(provider/vllm_rerank): handle null suffix and improve hint descriptions

- Add explicit None check for rerank_api_suffix (fixes HIGH from Gemini)
- Update rerank_api_base hint to describe actual behavior without
  mentioning specific provider options (fixes 3x MEDIUM from Gemini)
- Add ru-RU i18n for rerank_api_suffix (fixes P2 from Codex)

Co-authored-by: gemini-code-assist[bot]
Co-authored-by: chatgpt-codex-connector[bot]

---------

Co-authored-by: LehaoLin <linlehao@cuhk.edu.cn>

README.md

@@ -92,6 +92,9 @@ Update `astrbot`:
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> AstrBot deployed via `uv` **does not support upgrading through the WebUI**. To update, please run the command above from the command line.
+
 ### Docker Deployment
 
 For users familiar with containers and looking for a more stable, production-ready deployment method, we recommend deploying AstrBot with Docker / Docker Compose.

README_fr.md

@@ -92,6 +92,9 @@ Mettre à jour `astrbot` :
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> AstrBot déployé via `uv` **ne prend pas en charge la mise à jour via le WebUI**. Pour mettre à jour, exécutez la commande ci-dessus depuis le terminal.
+
 ### Déploiement Docker
 
 Pour les utilisateurs familiers avec les conteneurs et qui souhaitent une méthode plus stable et adaptée à la production, nous recommandons de déployer AstrBot avec Docker / Docker Compose.

README_ja.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> `uv` 経由でデプロイした AstrBot は、**WebUI からのバージョンアップグレードに対応していません**。更新するには、上記のコマンドをコマンドラインで実行してください。
+
 ### Docker デプロイ
 
 コンテナ運用に慣れており、より安定した本番向けのデプロイ方法を求めるユーザーには、Docker / Docker Compose での AstrBot デプロイをおすすめします。

README_ru.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> AstrBot, развёрнутый через `uv`, **не поддерживает обновление через WebUI**. Для обновления выполните указанную выше команду из командной строки.
+
 ### Развёртывание Docker
 
 Для пользователей, знакомых с контейнерами и которым нужен более стабильный и подходящий для production способ, мы рекомендуем разворачивать AstrBot через Docker / Docker Compose.

README_zh-TW.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> 透過 `uv` 部署的 AstrBot **不支援在 WebUI 中進行版本升級**。如需更新，請透過命令列執行上述命令。
+
 ### Docker 部署
 
 對於熟悉容器、希望獲得更穩定且更適合正式環境部署方式的使用者，我們推薦使用 Docker / Docker Compose 部署 AstrBot。

README_zh.md

@@ -92,6 +92,9 @@ astrbot run
 uv tool upgrade astrbot
 ```
 
+> [!WARNING]
+> 通过 `uv` 部署的 AstrBot **不支持在 WebUI 中进行版本升级**。如需更新，请通过命令行执行上述命令。
+
 ### Docker 部署
 
 对于熟悉容器、希望获得更稳定且更适合生产环境部署方式的用户，我们推荐使用 Docker / Docker Compose 部署 AstrBot。

astrbot/cli/__init__.py

@@ -1 +1 @@
-__version__ = "4.22.2"
+__version__ = "4.22.3"

astrbot/core/agent/mcp_client.py

@@ -51,18 +51,29 @@ def _prepare_stdio_env(config: dict) -> dict:
     """Preserve Windows executable resolution for stdio subprocesses."""
     if sys.platform != "win32":
         return config
-
-    pathext = os.environ.get("PATHEXT")
-    if not pathext:
-        return config
-
     prepared = config.copy()
     env = dict(prepared.get("env") or {})
-    env.setdefault("PATHEXT", pathext)
+    env = _merge_environment_variables(env)
     prepared["env"] = env
     return prepared
 
 
+def _merge_environment_variables(env: dict) -> dict:
+    """合并环境变量，处理Windows不区分大小写的情况"""
+    merged = env.copy()
+
+    # 将用户环境变量转换为统一的大小写形式便于比较
+    user_keys_lower = {k.lower(): k for k in merged.keys()}
+
+    for sys_key, sys_value in os.environ.items():
+        sys_key_lower = sys_key.lower()
+        if sys_key_lower not in user_keys_lower:
+            # 使用系统环境变量中的原始大小写
+            merged[sys_key] = sys_value
+
+    return merged
+
+
 async def _quick_test_mcp_connection(config: dict) -> tuple[bool, str]:
     """Quick test MCP server connectivity"""
     import aiohttp

astrbot/core/config/default.py

@@ -5,7 +5,7 @@ from typing import Any, TypedDict
 
 from astrbot.core.utils.astrbot_path import get_astrbot_data_path
 
-VERSION = "4.22.2"
+VERSION = "4.22.3"
 DB_PATH = os.path.join(get_astrbot_data_path(), "data_v4.db")
 PERSONAL_WECHAT_CONFIG_METADATA = {
     "weixin_oc_base_url": {
@@ -1761,6 +1761,7 @@ CONFIG_METADATA_2 = {
                         "enable": True,
                         "rerank_api_key": "",
                         "rerank_api_base": "http://127.0.0.1:8000",
+                        "rerank_api_suffix": "/v1/rerank",
                         "rerank_model": "BAAI/bge-reranker-base",
                         "timeout": 20,
                     },
@@ -1826,7 +1827,12 @@ CONFIG_METADATA_2 = {
                     "rerank_api_base": {
                         "description": "重排序模型 API Base URL",
                         "type": "string",
-                        "hint": "AstrBot 会在请求时在末尾加上 /v1/rerank。",
+                        "hint": "最终请求路径由 Base URL 和路径后缀拼接而成（默认为 /v1/rerank）。",
+                    },
+                    "rerank_api_suffix": {
+                        "description": "API URL 路径后缀",
+                        "type": "string",
+                        "hint": "追加到 base_url 后的路径，如 /v1/rerank。留空则不追加。",
                     },
                     "rerank_api_key": {
                         "description": "API Key",

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

@@ -35,11 +35,11 @@ class DiscordBotClient(discord.Bot):
     async def on_ready(self) -> None:
         """当机器人成功连接并准备就绪时触发"""
         if self.user is None:
-            logger.error("[Discord] 客户端未正确加载用户信息 (self.user is None)")
+            logger.error("[Discord] Bot user not loaded correctly (self.user is None)")
             return
 
-        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。")
+        logger.info(f"[Discord] Logged in as {self.user} (ID: {self.user.id})")
+        logger.info("[Discord] Client is ready.")
 
         if self.on_ready_once_callback and not self._ready_once_fired:
             self._ready_once_fired = True
@@ -47,7 +47,7 @@ class DiscordBotClient(discord.Bot):
                 await self.on_ready_once_callback()
             except Exception as e:
                 logger.error(
-                    f"[Discord] on_ready_once_callback 执行失败: {e}",
+                    f"[Discord] Failed to execute on_ready_once_callback: {e}",
                     exc_info=True,
                 )
 
@@ -99,7 +99,7 @@ class DiscordBotClient(discord.Bot):
             return
 
         logger.debug(
-            f"[Discord] 收到原始消息 from {message.author.name}: {message.content}",
+            f"[Discord] Received raw message from {message.author.name}: {message.content}",
         )
 
         if self.on_message_received:

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

@@ -46,7 +46,7 @@ class DiscordPlatformAdapter(Platform):
     ) -> None:
         super().__init__(platform_config, event_queue)
         self.settings = platform_settings
-        self.client_self_id: str | None = None
+        self.bot_self_id: str | None = None
         self.registered_handlers = []
         # 指令注册相关
         self.enable_command_register = self.config.get("discord_command_register", True)
@@ -64,7 +64,7 @@ class DiscordPlatformAdapter(Platform):
         """通过会话发送消息"""
         if self.client.user is None:
             logger.error(
-                "[Discord] 客户端未就绪 (self.client.user is None)，无法发送消息"
+                "[Discord] Client is not ready (self.client.user is None); message send skipped"
             )
             return
 
@@ -92,10 +92,10 @@ class DiscordPlatformAdapter(Platform):
 
         message_obj.message_str = message_chain.get_plain_text()
         message_obj.sender = MessageMember(
-            user_id=str(self.client_self_id),
+            user_id=str(self.bot_self_id),
             nickname=self.client.user.display_name,
         )
-        message_obj.self_id = cast(str, self.client_self_id)
+        message_obj.self_id = cast(str, self.bot_self_id)
         message_obj.session_id = session.session_id
         message_obj.message = message_chain.chain
 
@@ -115,7 +115,7 @@ class DiscordPlatformAdapter(Platform):
         """返回平台元数据"""
         return PlatformMetadata(
             "discord",
-            "Discord 适配器",
+            "Discord Adapter",
             id=cast(str, self.config.get("id")),
             default_config_tmpl=self.config,
             support_streaming_message=False,
@@ -127,16 +127,18 @@ class DiscordPlatformAdapter(Platform):
 
         # 初始化回调函数
         async def on_received(message_data) -> None:
-            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")
+            logger.debug(f"[Discord] Message received: {message_data}")
+            if self.bot_self_id is None:
+                self.bot_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。")
+            logger.error(
+                "[Discord] Bot token is not configured. Please set a valid token in the config file."
+            )
             return
 
         proxy = self.config.get("discord_proxy") or None
@@ -144,12 +146,17 @@ class DiscordPlatformAdapter(Platform):
         self.client.on_message_received = on_received
 
         async def callback() -> None:
-            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),
+            try:
+                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),
+                    )
+            except Exception as e:
+                logger.error(
+                    f"[Discord] on_ready_once_callback err: {e}", exc_info=True
                 )
 
         self.client.on_ready_once_callback = callback
@@ -158,11 +165,16 @@ class DiscordPlatformAdapter(Platform):
             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。")
+            logger.error(
+                "[Discord] Login failed. Please check whether the bot token is correct."
+            )
         except discord.errors.ConnectionClosed:
-            logger.warning("[Discord] 与 Discord 的连接已关闭。")
+            logger.warning("[Discord] Connection with Discord has been closed.")
         except Exception as e:
-            logger.error(f"[Discord] 适配器运行时发生意外错误: {e}", exc_info=True)
+            logger.error(
+                f"[Discord] Unexpected error while adapter is running: {e}",
+                exc_info=True,
+            )
 
     def _get_message_type(
         self,
@@ -241,7 +253,7 @@ class DiscordPlatformAdapter(Platform):
                     )
         abm.message = message_chain
         abm.raw_message = message
-        abm.self_id = cast(str, self.client_self_id)
+        abm.self_id = cast(str, self.bot_self_id)
         abm.session_id = str(message.channel.id)
         abm.message_id = str(message.id)
         return abm
@@ -264,7 +276,7 @@ class DiscordPlatformAdapter(Platform):
 
         if self.client.user is None:
             logger.error(
-                "[Discord] 客户端未就绪 (self.client.user is None)，无法处理消息"
+                "[Discord] Client is not ready (self.client.user is None); message handling skipped"
             )
             return
 
@@ -283,7 +295,7 @@ class DiscordPlatformAdapter(Platform):
         raw_message = message.raw_message
         if not isinstance(raw_message, discord.Message):
             logger.warning(
-                f"[Discord] 收到非 Message 类型的消息: {type(raw_message)}，已忽略。"
+                f"[Discord] Non-Message type received and ignored: {type(raw_message)}"
             )
             return
 
@@ -324,20 +336,9 @@ class DiscordPlatformAdapter(Platform):
 
     @override
     async def terminate(self) -> None:
-        """终止适配器"""
-        logger.info("[Discord] 正在终止适配器... (step 1: cancel polling task)")
+        logger.info("[Discord] Shutting down adapter...")
         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
+        logger.info("[Discord] Cleaning up commands...")
         if self.enable_command_register and self.client:
             try:
                 await asyncio.wait_for(
@@ -347,16 +348,29 @@ class DiscordPlatformAdapter(Platform):
                     ),
                     timeout=10,
                 )
-                logger.info("[Discord] 指令清理完成。")
+                logger.info("[Discord] Commands cleaned up successfully.")
             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)")
+                logger.warning(
+                    f"[Discord] Error occurred while cleaning up commands: {e}"
+                )
+
+        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 cancelled successfully.")
+            except Exception as e:
+                logger.warning(
+                    f"[Discord] Error occurred while cancelling polling task: {e}"
+                )
+        logger.info("[Discord] Closing client connection...")
         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。")
+                logger.warning(f"[Discord] Error occurred while closing client: {e}")
+        logger.info("[Discord] Adapter shutdown complete.")
 
     def register_handler(self, handler_info) -> None:
         """注册处理器信息"""
@@ -364,7 +378,7 @@ class DiscordPlatformAdapter(Platform):
 
     async def _collect_and_register_commands(self) -> None:
         """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...")
+        logger.info("[Discord] Collecting and registering slash commands...")
         registered_commands = []
 
         for handler_md in star_handlers_registry:
@@ -405,15 +419,15 @@ class DiscordPlatformAdapter(Platform):
 
         if registered_commands:
             logger.info(
-                f"[Discord] 准备同步 {len(registered_commands)} 个指令: {', '.join(registered_commands)}",
+                f"[Discord] Ready to sync {len(registered_commands)} commands: {', '.join(registered_commands)}",
             )
         else:
-            logger.info("[Discord] 没有发现可注册的指令。")
+            logger.info("[Discord] No commands found for registration.")
 
         # 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。")
+        logger.info("[Discord] Command synchronization completed.")
 
     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"""
@@ -422,17 +436,17 @@ class DiscordPlatformAdapter(Platform):
             ctx: discord.ApplicationContext, params: str | None = None
         ) -> 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}")
+            logger.debug(f"[Discord] Callback triggered: {cmd_name}")
+            logger.debug(f"[Discord] Callback context: {ctx}")
+            logger.debug(f"[Discord] Callback params: {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}'",
+                f"[Discord] Slash command '{cmd_name}' triggered. "
+                f"Raw params: '{params}'. "
+                f"Built command string: '{message_str_for_filter}'",
             )
 
             # 尝试立即响应，防止超时
@@ -441,7 +455,7 @@ class DiscordPlatformAdapter(Platform):
                 await ctx.defer()
                 followup_webhook = ctx.followup
             except Exception as e:
-                logger.warning(f"[Discord] 指令 '{cmd_name}' defer 失败: {e}")
+                logger.warning(f"[Discord] Failed to defer command '{cmd_name}': {e}")
 
             # 2. 构建 AstrBotMessage
             channel = ctx.channel
@@ -465,7 +479,7 @@ class DiscordPlatformAdapter(Platform):
             )
             abm.message = [Plain(text=message_str_for_filter)]
             abm.raw_message = ctx.interaction
-            abm.self_id = cast(str, self.client_self_id)
+            abm.self_id = cast(str, self.bot_self_id)
             abm.session_id = str(ctx.channel_id)
             abm.message_id = str(ctx.interaction.id)
 
@@ -503,10 +517,10 @@ class DiscordPlatformAdapter(Platform):
 
         # 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}")
+            logger.debug(f"[Discord] Skipping invalid slash command format: {cmd_name}")
             return None
 
-        description = handler_metadata.desc or f"指令: {cmd_name}"
+        description = handler_metadata.desc or f"Command: {cmd_name}"
         if len(description) > 100:
             description = f"{description[:97]}..."
 

astrbot/core/platform/sources/misskey/misskey_adapter.py

@@ -93,7 +93,7 @@ class MisskeyPlatformAdapter(Platform):
 
         self.api: MisskeyAPI | None = None
         self._running = False
-        self.client_self_id = ""
+        self.bot_self_id = ""
         self._bot_username = ""
         self._user_cache = {}
 
@@ -138,10 +138,10 @@ class MisskeyPlatformAdapter(Platform):
 
         try:
             user_info = await self.api.get_current_user()
-            self.client_self_id = str(user_info.get("id", ""))
+            self.bot_self_id = str(user_info.get("id", ""))
             self._bot_username = user_info.get("username", "")
             logger.info(
-                f"[Misskey] 已连接用户: {self._bot_username} (ID: {self.client_self_id})",
+                f"[Misskey] 已连接用户: {self._bot_username} (ID: {self.bot_self_id})",
             )
         except Exception as e:
             logger.error(f"[Misskey] 获取用户信息失败: {e}")
@@ -312,9 +312,9 @@ class MisskeyPlatformAdapter(Platform):
             )
             room_id = data.get("toRoomId")
             logger.debug(
-                f"[Misskey] 收到聊天事件: sender_id={sender_id}, room_id={room_id}, is_self={sender_id == self.client_self_id}",
+                f"[Misskey] 收到聊天事件: sender_id={sender_id}, room_id={room_id}, is_self={sender_id == self.bot_self_id}",
             )
-            if sender_id == self.client_self_id:
+            if sender_id == self.bot_self_id:
                 return
 
             if room_id:
@@ -354,13 +354,13 @@ class MisskeyPlatformAdapter(Platform):
         mentions = note.get("mentions", [])
         if self._bot_username and f"@{self._bot_username}" in text:
             return True
-        if self.client_self_id in [str(uid) for uid in mentions]:
+        if self.bot_self_id in [str(uid) for uid in mentions]:
             return True
 
         reply = note.get("reply")
         if reply and isinstance(reply, dict):
             reply_user_id = str(reply.get("user", {}).get("id", ""))
-            if reply_user_id == self.client_self_id:
+            if reply_user_id == self.bot_self_id:
                 return bool(self._bot_username and f"@{self._bot_username}" in text)
 
         return False
@@ -598,7 +598,7 @@ class MisskeyPlatformAdapter(Platform):
                     visibility, visible_user_ids = resolve_message_visibility(
                         user_id=user_id_for_cache,
                         user_cache=self._user_cache,
-                        self_id=self.client_self_id,
+                        self_id=self.bot_self_id,
                         default_visibility=self.default_visibility,
                     )
                     logger.debug(
@@ -637,14 +637,14 @@ class MisskeyPlatformAdapter(Platform):
         message = create_base_message(
             raw_data,
             sender_info,
-            self.client_self_id,
+            self.bot_self_id,
             is_chat=False,
         )
         cache_user_info(
             self._user_cache,
             sender_info,
             raw_data,
-            self.client_self_id,
+            self.bot_self_id,
             is_chat=False,
         )
 
@@ -656,7 +656,7 @@ class MisskeyPlatformAdapter(Platform):
                 message,
                 raw_text,
                 self._bot_username,
-                self.client_self_id,
+                self.bot_self_id,
             )
             message_parts.extend(text_parts)
 
@@ -685,14 +685,14 @@ class MisskeyPlatformAdapter(Platform):
         message = create_base_message(
             raw_data,
             sender_info,
-            self.client_self_id,
+            self.bot_self_id,
             is_chat=True,
         )
         cache_user_info(
             self._user_cache,
             sender_info,
             raw_data,
-            self.client_self_id,
+            self.bot_self_id,
             is_chat=True,
         )
 
@@ -713,7 +713,7 @@ class MisskeyPlatformAdapter(Platform):
         message = create_base_message(
             raw_data,
             sender_info,
-            self.client_self_id,
+            self.bot_self_id,
             is_chat=False,
             room_id=room_id,
         )
@@ -722,10 +722,10 @@ class MisskeyPlatformAdapter(Platform):
             self._user_cache,
             sender_info,
             raw_data,
-            self.client_self_id,
+            self.bot_self_id,
             is_chat=False,
         )
-        cache_room_info(self._user_cache, raw_data, self.client_self_id)
+        cache_room_info(self._user_cache, raw_data, self.bot_self_id)
 
         raw_text = raw_data.get("text", "")
         message_parts = []
@@ -736,7 +736,7 @@ class MisskeyPlatformAdapter(Platform):
                     message,
                     raw_text,
                     self._bot_username,
-                    self.client_self_id,
+                    self.bot_self_id,
                 )
                 message_parts.extend(text_parts)
             else:

astrbot/core/platform/sources/misskey/misskey_utils.py

@@ -335,7 +335,7 @@ def extract_sender_info(
 def create_base_message(
     raw_data: dict[str, Any],
     sender_info: dict[str, Any],
-    client_self_id: str,
+    bot_self_id: str,
     is_chat: bool = False,
     room_id: str | None = None,
 ) -> AstrBotMessage:
@@ -367,7 +367,7 @@ def create_base_message(
         session_id if sender_info["sender_id"] else f"{session_prefix}%unknown"
     )
     message.message_id = str(raw_data.get("id", ""))
-    message.self_id = client_self_id
+    message.self_id = bot_self_id
 
     return message
 
@@ -376,7 +376,7 @@ def process_at_mention(
     message: AstrBotMessage,
     raw_text: str,
     bot_username: str,
-    client_self_id: str,
+    bot_self_id: str,
 ) -> tuple[list[str], str]:
     """处理@提及逻辑，返回消息部分列表和处理后的文本"""
     message_parts = []
@@ -386,7 +386,7 @@ def process_at_mention(
 
     if bot_username and raw_text.startswith(f"@{bot_username}"):
         at_mention = f"@{bot_username}"
-        message.message.append(Comp.At(qq=client_self_id))
+        message.message.append(Comp.At(qq=bot_self_id))
         remaining_text = raw_text[len(at_mention) :].strip()
         if remaining_text:
             message.message.append(Comp.Plain(remaining_text))
@@ -401,7 +401,7 @@ def cache_user_info(
     user_cache: dict[str, Any],
     sender_info: dict[str, Any],
     raw_data: dict[str, Any],
-    client_self_id: str,
+    bot_self_id: str,
     is_chat: bool = False,
 ) -> None:
     """缓存用户信息"""
@@ -410,7 +410,7 @@ def cache_user_info(
             "username": sender_info["username"],
             "nickname": sender_info["nickname"],
             "visibility": "specified",
-            "visible_user_ids": [client_self_id, sender_info["sender_id"]],
+            "visible_user_ids": [bot_self_id, sender_info["sender_id"]],
         }
     else:
         user_cache_data = {
@@ -428,7 +428,7 @@ def cache_user_info(
 def cache_room_info(
     user_cache: dict[str, Any],
     raw_data: dict[str, Any],
-    client_self_id: str,
+    bot_self_id: str,
 ) -> None:
     """缓存房间信息"""
     room_data = raw_data.get("toRoom")
@@ -442,7 +442,7 @@ def cache_room_info(
             "room_description": room_data.get("description", ""),
             "owner_id": room_data.get("ownerId", ""),
             "visibility": "specified",
-            "visible_user_ids": [client_self_id],
+            "visible_user_ids": [bot_self_id],
         }
 
 

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

@@ -50,7 +50,6 @@ class TelegramPlatformAdapter(Platform):
     ) -> None:
         super().__init__(platform_config, event_queue)
         self.settings = platform_settings
-        self.client_self_id = uuid.uuid4().hex[:8]
 
         base_url = self.config.get(
             "telegram_api_base_url",
@@ -336,6 +335,8 @@ class TelegramPlatformAdapter(Platform):
             return None
 
         def _apply_caption() -> None:
+            if not update.message:
+                return
             if update.message.caption:
                 message.message_str = update.message.caption
                 message.message.append(Comp.Plain(message.message_str))

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

@@ -148,7 +148,6 @@ class WecomPlatformAdapter(Platform):
     ) -> None:
         super().__init__(platform_config, event_queue)
         self.settingss = platform_settings
-        self.client_self_id = uuid.uuid4().hex[:8]
         self.api_base_url = platform_config.get(
             "api_base_url",
             "https://qyapi.weixin.qq.com/cgi-bin/",

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

@@ -2,7 +2,6 @@ import asyncio
 import os
 import sys
 import time
-import uuid
 from collections.abc import Callable, Coroutine
 from typing import Any, cast
 
@@ -324,7 +323,6 @@ class WeixinOfficialAccountPlatformAdapter(Platform):
     ) -> None:
         super().__init__(platform_config, event_queue)
         self.settingss = platform_settings
-        self.client_self_id = uuid.uuid4().hex[:8]
         self.api_base_url = platform_config.get(
             "api_base_url",
             "https://api.weixin.qq.com/cgi-bin/",

astrbot/core/provider/sources/gemini_source.py

@@ -241,15 +241,10 @@ class ProviderGoogleGenAI(Provider):
                 thinking_config = types.ThinkingConfig(
                     thinking_budget=thinking_budget,
                 )
-        elif model_name in [
-            "gemini-3-pro",
-            "gemini-3-pro-preview",
-            "gemini-3-flash",
-            "gemini-3-flash-preview",
-            "gemini-3-flash-lite",
-            "gemini-3-flash-lite-preview",
-        ]:
-            # The thinkingLevel parameter, recommended for Gemini 3 models and onwards
+        elif any(model_name.startswith(p) for p in ("gemini-3-", "gemini-3.")):
+            # The thinkingLevel parameter, recommended for Gemini 3 models and onwards.
+            # Use prefix match so new variants (3.1, 3-flash-lite-preview, etc.) are
+            # covered without needing to keep an exhaustive list up to date.
             # Gemini 2.5 series models don't support thinkingLevel; use thinkingBudget instead.
             thinking_level = self.provider_config.get("gm_thinking_config", {}).get(
                 "level", "HIGH"
@@ -517,7 +512,11 @@ class ProviderGoogleGenAI(Provider):
         ):
             chain.append(Comp.Plain("这是图片"))
         for part in result_parts:
-            if part.text:
+            # Skip thinking parts — their text is already captured via
+            # _extract_reasoning_content above.  Including them here would
+            # leak the model's internal reasoning into the user-facing message,
+            # which also causes duplicate/triple replies on some platforms.
+            if part.text and not part.thought:
                 chain.append(Comp.Plain(part.text))
 
             if (

astrbot/core/provider/sources/openai_source.py

@@ -532,6 +532,7 @@ class ProviderOpenAIOfficial(Provider):
             **payloads,
             stream=True,
             extra_body=extra_body,
+            stream_options={"include_usage": True},
         )
 
         llm_response = LLMResponse("assistant", is_chunk=True)
@@ -539,12 +540,10 @@ class ProviderOpenAIOfficial(Provider):
         state = ChatCompletionStreamState()
 
         async for chunk in stream:
-            if not chunk.choices:
-                continue
-            choice = chunk.choices[0]
-            delta = choice.delta
+            choice = chunk.choices[0] if chunk.choices else None
+            delta = choice.delta if choice else None
 
-            if dtcs := delta.tool_calls:
+            if delta and (dtcs := delta.tool_calls):
                 for idx, tc in enumerate(dtcs):
                     # siliconflow workaround
                     if tc.function and tc.function.arguments:
@@ -574,7 +573,7 @@ class ProviderOpenAIOfficial(Provider):
                 _y = True
             if chunk.usage:
                 llm_response.usage = self._extract_usage(chunk.usage)
-            elif choice_usage := getattr(choice, "usage", None):
+            elif choice and (choice_usage := getattr(choice, "usage", None)):
                 # Workaround for some providers that only return usage in choices[].usage, e.g. MoonshotAI
                 # See https://github.com/AstrBotDevs/AstrBot/issues/6614
                 llm_response.usage = self._extract_usage(choice_usage)
@@ -863,6 +862,9 @@ class ProviderOpenAIOfficial(Provider):
 
     def _finally_convert_payload(self, payloads: dict) -> None:
         """Finally convert the payload. Such as think part conversion, tool inject."""
+        model = payloads.get("model", "").lower()
+        is_gemini = "gemini" in model
+
         for message in payloads.get("messages", []):
             if message.get("role") == "assistant" and isinstance(
                 message.get("content"), list
@@ -874,11 +876,24 @@ class ProviderOpenAIOfficial(Provider):
                         reasoning_content += str(part.get("think"))
                     else:
                         new_content.append(part)
-                message["content"] = new_content
-                # reasoning key is "reasoning_content"
+                # Some providers (Grok, etc.) reject empty content lists.
+                # When all parts were think blocks, fall back to None.
+                message["content"] = new_content or None
                 if reasoning_content:
                     message["reasoning_content"] = reasoning_content
 
+            # Gemini 的 function_response 要求 google.protobuf.Struct（即 JSON 对象），
+            # 纯文本会触发 400 Invalid argument，需要包一层 JSON。
+            if is_gemini and message.get("role") == "tool":
+                content = message.get("content", "")
+                if isinstance(content, str):
+                    try:
+                        json.loads(content)
+                    except (json.JSONDecodeError, ValueError):
+                        message["content"] = json.dumps(
+                            {"result": content}, ensure_ascii=False
+                        )
+
     async def _handle_api_error(
         self,
         e: Exception,

astrbot/core/provider/sources/vllm_rerank_source.py

@@ -20,6 +20,11 @@ class VLLMRerankProvider(RerankProvider):
         self.auth_key = provider_config.get("rerank_api_key", "")
         self.base_url = provider_config.get("rerank_api_base", "http://127.0.0.1:8000")
         self.base_url = self.base_url.rstrip("/")
+        self.api_suffix = provider_config.get("rerank_api_suffix", "/v1/rerank")
+        if self.api_suffix is None:
+            self.api_suffix = "/v1/rerank"
+        if self.api_suffix and not self.api_suffix.startswith("/"):
+            self.api_suffix = "/" + self.api_suffix
         self.timeout = provider_config.get("timeout", 20)
         self.model = provider_config.get("rerank_model", "BAAI/bge-reranker-base")
 
@@ -45,8 +50,9 @@ class VLLMRerankProvider(RerankProvider):
         if top_n is not None:
             payload["top_n"] = top_n
         assert self.client is not None
+        rerank_url = f"{self.base_url}{self.api_suffix}"
         async with self.client.post(
-            f"{self.base_url}/v1/rerank",
+            rerank_url,
             json=payload,
         ) as response:
             response_data = await response.json()

changelogs/v4.22.3.md

@@ -0,0 +1,68 @@
+- [更新日志(简体中文)](#chinese)
+- [Changelog(English)](#english)
+
+<a id="chinese"></a>
+
+## What's Changed
+
+### 新增
+
+- 新增 `vllm_rerank` 提供商支持可配置的 `rerank_api_suffix`，可灵活适配不同 API 路径。（[#7278](https://github.com/AstrBotDevs/AstrBot/pull/7278)）
+- 适配 llama.cpp 的 token 用量提取支持，增强 token 统计与计费分析能力。（[#7358](https://github.com/AstrBotDevs/AstrBot/pull/7358)）
+- WebUI 捆绑 Monaco Editor 的 worker 文件，防止从 CDN 下载失败。（[#7249](https://github.com/AstrBotDevs/AstrBot/pull/7249)）
+- 全新的模型/平台总览统计面板。（[#7152](https://github.com/AstrBotDevs/AstrBot/pull/7152)）
+
+### 优化
+
+- 优化 OpenAI 消息处理，若 `content` 列表为空时将其置为 `None`，并降低空内容下的异常率。（[#6551](https://github.com/AstrBotDevs/AstrBot/pull/6551)）
+- 优化 provider page 布局与样式，提升整体设置页面排版一致性。（[#7248](https://github.com/AstrBotDevs/AstrBot/pull/7248)）
+- 用户已登录时将直接跳转欢迎页，减少再次进入登录页的步骤。（[commit](https://github.com/AstrBotDevs/AstrBot/commit/9896b48c)）
+
+### 修复
+
+- 修复 Windows 下 MCP runner 未继承完整系统环境变量导致启动异常的问题。（[#7054](https://github.com/AstrBotDevs/AstrBot/pull/7054)）
+- 修复 Gemini 思考片段泄露到用户可见消息链路的问题。（[#7196](https://github.com/AstrBotDevs/AstrBot/pull/7196)）
+- 修复 Gemini(OpenAI 兼容 API) 下，`function_response` 格式要求导致的 `400 Invalid argument` 错误。（[#7216](https://github.com/AstrBotDevs/AstrBot/pull/7216)
+- 修复知识库初始化失败时的错误提示与恢复问题。（[#7243](https://github.com/AstrBotDevs/AstrBot/pull/7243)）
+- 修复 Bailian Rerank 对新旧响应格式兼容性，避免解析失败。（[#7217](https://github.com/AstrBotDevs/AstrBot/pull/7217)）
+- Docker 镜像补充 AMR 编码器依赖，修复部分部署场景下的音频转码缺失问题。（[#7247](https://github.com/AstrBotDevs/AstrBot/pull/7247)）
+- 修复 pip 安装缺少 `pysocks` 依赖导致的 SOCKS5 代理不可用问题。（[#7221](https://github.com/AstrBotDevs/AstrBot/pull/7221)）
+- 修复 `cached_tokens` 在 `_extract_usage` 中的处理逻辑，避免 token 统计错乱。（[#6719](https://github.com/AstrBotDevs/AstrBot/pull/6719)）
+- 增强 SSE 断开保护，压缩传输下主动发送心跳，减少 WebChat 误断连。（[#7003](https://github.com/AstrBotDevs/AstrBot/pull/7003)）
+- 修复严格 API 下空 assistant 消息导致的 400 错误。（[#7202](https://github.com/AstrBotDevs/AstrBot/pull/7202)）
+- 新增二维码功能缺失的 `qrcode` 依赖修复，恢复二维码生成功能。（[commit](https://github.com/AstrBotDevs/AstrBot/commit/5886c437)）
+- 修复配置器数字输入组件的交互体验问题，避免异常输入场景下的处理偏差。（[#7153](https://github.com/AstrBotDevs/AstrBot/pull/7153)）
+- 修复 `Record.path` 在 Pydantic v2 下被错误视为必填字段的问题。（[#7048](https://github.com/AstrBotDevs/AstrBot/pull/7048)）
+
+<a id="english"></a>
+## What's Changed (EN)
+
+### New Features
+
+- Added configurable `rerank_api_suffix` for the `vllm_rerank` provider to support custom API paths.（[#7278](https://github.com/AstrBotDevs/AstrBot/pull/7278)）
+- Added token usage extraction support for llama.cpp to improve token accounting.（[#7358](https://github.com/AstrBotDevs/AstrBot/pull/7358)）
+- Bundled Monaco Editor workers in WebUI so code editing does not depend on external CDN downloads.（[#7249](https://github.com/AstrBotDevs/AstrBot/pull/7249)）
+- Added a new statistics overview for model and platform usage.（[#7152](https://github.com/AstrBotDevs/AstrBot/pull/7152)）
+- Redirected logged-in users to the welcome page instead of the login page.（[commit](https://github.com/AstrBotDevs/AstrBot/commit/9896b48c)）
+
+### Improvements
+
+- Improved OpenAI message handling by setting message `content` to `None` when content lists are empty, reducing empty-payload edge cases.（[#6551](https://github.com/AstrBotDevs/AstrBot/pull/6551)）
+- Improved provider page layout and styling to keep settings pages consistent.（[#7248](https://github.com/AstrBotDevs/AstrBot/pull/7248)）
+- Refined connection-success handling and messaging when users are already logged in.（[commit](https://github.com/AstrBotDevs/AstrBot/commit/2de23184)）
+
+### Bug Fixes
+
+- Inherited all system environment variables in Windows MCP runner to fix environment-dependent launch issues.（[#7054](https://github.com/AstrBotDevs/AstrBot/pull/7054)）
+- Filtered Gemini thinking parts from user-facing message chains.（[#7196](https://github.com/AstrBotDevs/AstrBot/pull/7196)）
+- Fixed Gemini `function_response` compatibility to prevent `400 Invalid argument` responses.（[#7216](https://github.com/AstrBotDevs/AstrBot/pull/7216)）
+- Improved knowledge base initialization error handling.（[#7243](https://github.com/AstrBotDevs/AstrBot/pull/7243)）
+- Fixed Bailian rerank parsing for both old and new response formats.（[#7217](https://github.com/AstrBotDevs/AstrBot/pull/7217)）
+- Added AMR encoder support in Docker builds to fix missing audio codec support.（[#7247](https://github.com/AstrBotDevs/AstrBot/pull/7247)）
+- Added the `pysocks` dependency so pip installs can use SOCKS5 proxy correctly.（[#7221](https://github.com/AstrBotDevs/AstrBot/pull/7221)）
+- Fixed `_extract_usage` cached token handling.（[#6719](https://github.com/AstrBotDevs/AstrBot/pull/6719)）
+- Sent SSE heartbeat packets to prevent WebChat disconnection under compression.（[#7003](https://github.com/AstrBotDevs/AstrBot/pull/7003)）
+- Fixed `400` errors caused by empty assistant messages on strict APIs.（[#7202](https://github.com/AstrBotDevs/AstrBot/pull/7202)）
+- Added missing `qrcode` dependency to restore QR code generation.（[commit](https://github.com/AstrBotDevs/AstrBot/commit/5886c437)）
+- Fixed number input UX in config renderer.（[#7153](https://github.com/AstrBotDevs/AstrBot/pull/7153)）
+- Fixed `Record.path` being incorrectly treated as required under Pydantic v2.（[#7048](https://github.com/AstrBotDevs/AstrBot/pull/7048)）

dashboard/src/i18n/locales/en-US/features/config-metadata.json

@@ -1075,7 +1075,11 @@
       },
       "rerank_api_base": {
         "description": "Rerank Model API Base URL",
-        "hint": "AstrBot appends /v1/rerank to the request URL."
+        "hint": "The full request URL is formed by combining the Base URL and a path suffix (defaults to /v1/rerank)."
+      },
+      "rerank_api_suffix": {
+        "description": "API URL path suffix",
+        "hint": "Path appended to base_url, e.g. /v1/rerank. Leave empty to disable auto-append."
       },
       "rerank_api_key": {
         "description": "API Key",

dashboard/src/i18n/locales/ru-RU/features/config-metadata.json

@@ -1076,7 +1076,11 @@
             },
             "rerank_api_base": {
                 "description": "Base URL API модели Rerank",
-                "hint": "AstrBot добавляет /v1/rerank к URL запроса."
+                "hint": "Полный URL запроса формируется путём добавления суффикса к Base URL (по умолчанию /v1/rerank)."
+            },
+            "rerank_api_suffix": {
+                "description": "Суффикс пути API",
+                "hint": "Суффикс пути, добавляемый к base_url, например /v1/rerank. Оставьте пустым, чтобы не добавлять."
             },
             "rerank_api_key": {
                 "description": "API Key",

dashboard/src/i18n/locales/zh-CN/features/config-metadata.json

@@ -1077,7 +1077,11 @@
       },
       "rerank_api_base": {
         "description": "重排序模型 API Base URL",
-        "hint": "AstrBot 会在请求时在末尾加上 /v1/rerank。"
+        "hint": "最终请求路径由 Base URL 和路径后缀拼接而成（默认为 /v1/rerank）。"
+      },
+      "rerank_api_suffix": {
+        "description": "API URL 路径后缀",
+        "hint": "追加到 base_url 后的路径后缀，如 /v1/rerank。留空则不追加。"
       },
       "rerank_api_key": {
         "description": "API Key",

docs/en/deploy/astrbot/package.md

@@ -9,6 +9,11 @@ If `uv` is not installed, install it first by following the official guide:
 
 `uv` supports Linux, Windows, and macOS.
 
+## Important Notes
+
+> [!WARNING]
+> AstrBot deployed via `uv` **does not support upgrading through the WebUI**. To update, run `uv tool upgrade astrbot` from the command line.
+
 ## Install and Start
 
 ```bash

docs/zh/deploy/astrbot/package.md

@@ -8,6 +8,11 @@
 
 `uv` 支持 Linux、Windows、macOS。
 
+## 注意事项
+
+> [!WARNING]
+> 通过 `uv` 部署的 AstrBot **不支持在 WebUI 中进行版本升级**。如需更新，请在命令行中执行 `uv tool upgrade astrbot`。
+
 ## 安装并启动
 
 ```bash

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "AstrBot"
-version = "4.22.2"
+version = "4.22.3"
 description = "Easy-to-use multi-platform LLM chatbot and development framework"
 readme = "README.md"
 license = { text = "AGPL-3.0-or-later" }

requirements.txt

@@ -56,3 +56,4 @@ tenacity>=9.1.2
 shipyard-python-sdk>=0.2.4
 shipyard-neo-sdk>=0.2.0
 packaging>=24.2
+qrcode>=8.2

tests/test_openai_source.py

@@ -2,6 +2,7 @@ from types import SimpleNamespace
 
 import pytest
 from openai.types.chat.chat_completion import ChatCompletion
+from openai.types.chat.chat_completion_chunk import ChatCompletionChunk
 from PIL import Image as PILImage
 
 from astrbot.core.exceptions import EmptyModelOutputError
@@ -1175,6 +1176,93 @@ async def test_parse_openai_completion_raises_empty_model_output_error():
         await provider.terminate()
 
 
+@pytest.mark.asyncio
+async def test_query_stream_extracts_usage_from_empty_choices_chunk(monkeypatch):
+    provider = _make_provider()
+    try:
+        chunks = [
+            ChatCompletionChunk.model_validate(
+                {
+                    "id": "chatcmpl-stream",
+                    "object": "chat.completion.chunk",
+                    "created": 0,
+                    "model": "gpt-4o-mini",
+                    "choices": [
+                        {
+                            "index": 0,
+                            "delta": {
+                                "role": "assistant",
+                                "content": "ok",
+                            },
+                            "finish_reason": None,
+                        }
+                    ],
+                }
+            ),
+            ChatCompletionChunk.model_validate(
+                {
+                    "id": "chatcmpl-stream",
+                    "object": "chat.completion.chunk",
+                    "created": 0,
+                    "model": "gpt-4o-mini",
+                    "choices": [
+                        {
+                            "index": 0,
+                            "delta": {},
+                            "finish_reason": "stop",
+                        }
+                    ],
+                }
+            ),
+            ChatCompletionChunk.model_validate(
+                {
+                    "id": "chatcmpl-stream",
+                    "object": "chat.completion.chunk",
+                    "created": 0,
+                    "model": "gpt-4o-mini",
+                    "choices": [],
+                    "usage": {
+                        "prompt_tokens": 2550,
+                        "completion_tokens": 125,
+                        "total_tokens": 2675,
+                        "prompt_tokens_details": {
+                            "cached_tokens": 2488,
+                        },
+                    },
+                }
+            ),
+        ]
+
+        async def fake_stream():
+            for chunk in chunks:
+                yield chunk
+
+        async def fake_create(**kwargs):
+            return fake_stream()
+
+        monkeypatch.setattr(provider.client.chat.completions, "create", fake_create)
+
+        responses = [
+            response
+            async for response in provider._query_stream(
+                payloads={
+                    "model": "gpt-4o-mini",
+                    "messages": [{"role": "user", "content": "hello"}],
+                },
+                tools=None,
+            )
+        ]
+
+        final_response = responses[-1]
+        assert final_response.completion_text == "ok"
+        assert final_response.usage is not None
+        assert final_response.usage.input_other == 62
+        assert final_response.usage.input_cached == 2488
+        assert final_response.usage.output == 125
+    finally:
+        await provider.terminate()
+
+
 @pytest.mark.asyncio
 async def test_query_filters_empty_assistant_message_without_tool_calls(monkeypatch):
     """Test that empty assistant messages without tool_calls are filtered out."""