@wecom/wecom-openclaw-plugin 2026.3.31-beta.4 → 2026.3.31-beta.5

package/README.md

@@ -31,17 +31,24 @@
 - 🔗 **Dual-mode**: Bot (WebSocket / Webhook) and Agent (HTTP webhook) can run independently or together
 - 💬 Supports both direct messages (DM) and group chat
 - 📤 Proactive messaging to specific users, groups, departments, or tags
-- 🖼️ Receives and processes image, voice, video, and file messages with automatic downloading
+- 🖼️ Receives and processes image, voice, video, file, and **mixed (图文混排)** messages with automatic downloading
+- 🗣️ Voice-to-text: automatically extracts transcribed text from voice messages
+- 💬 Quote message support: processes quoted text, image, voice, and file messages
 - ⏳ Streaming replies with "thinking" placeholder messages (Bot mode)
 - 🔐 Agent mode: AES-256-CBC encrypted XML callbacks with SHA1 signature verification
 - 📝 Markdown formatting support for replies
-- 🃏 Template card messages (text_notice, news_notice, button_interaction, vote_interaction, multiple_interaction)
+- 🃏 Template card messages (text_notice, news_notice, button_interaction, vote_interaction, multiple_interaction) with **event callback handling**
 - 🔒 Built-in access control: DM Policy (pairing / open / allowlist / disabled) and Group Policy (open / allowlist / disabled)
+- 🔑 Command authorization: per-account command permission control with access group support
 - 👥 Multi-account support: run multiple WeCom accounts with independent bot/agent configs
-- 🧩 MCP tool integration (`wecom_mcp`) for calling WeCom MCP Server
+- 🧩 MCP tool integration (`wecom_mcp`) with interceptor pipeline (biz-error, media, smartpage-create, smartpage-export)
+- 🎯 **15 built-in Skills**: contact lookup, doc management, todo, meeting, schedule, messaging, smartsheet, template cards, and more
 - 🔀 Dynamic Agent routing: auto-create isolated agents per user/group
 - 📁 Local file sending with configurable media path allowlist (`mediaLocalRoots`)
-- ⚡ Auto heartbeat keep-alive and reconnection (up to 10 reconnect attempts)
+- 📊 Smart media size limits with auto-downgrade (image 10MB → file, video 10MB → file, voice 2MB/AMR-only → file, max 20MB)
+- 🔄 **Bot-first, Agent-fallback** outbound delivery: auto fallback to Agent HTTP API when Bot WS is unavailable
+- ⚡ Auto heartbeat keep-alive and reconnection (up to 10 reconnect attempts, 5 auth failure retries)
+- 🛡️ Anti-kick protection: suppresses auto-restart on server-side disconnection to prevent mutual kicking loops
 - 🧙 Interactive CLI setup wizard
 
 ---
@@ -152,11 +159,20 @@ The plugin supports two connection modes that can be used independently or toget
 | Config Path | Description | Default |
 |---|---|---|
 | `channels.wecom.mediaLocalRoots` | Extra local paths allowed for media sending (supports `~`) | `[]` |
-| `channels.wecom.media.maxBytes` | Max media file size in bytes | `5242880` (5MB) |
+| `channels.wecom.media.maxBytes` | Max media file size in bytes | `20971520` (20MB) |
 | `channels.wecom.media.tempDir` | Temp directory for media processing | — |
 | `channels.wecom.media.retentionHours` | Media file retention hours | — |
 | `channels.wecom.media.cleanupOnStart` | Clean temp media on startup | — |
 
+**Media Size Limits & Auto-Downgrade:**
+
+| Media Type | Max Size | Downgrade Behavior |
+|---|---|---|
+| Image | 10 MB | Exceeds → sent as file |
+| Video | 10 MB | Exceeds → sent as file |
+| Voice | 2 MB (AMR only) | Non-AMR format or exceeds → sent as file |
+| File | 20 MB | Exceeds → rejected (cannot send) |
+
 #### Network Settings
 
 | Config Path | Description | Default |
@@ -175,13 +191,16 @@ Agent mode uses HTTP webhook callbacks with XML encrypted messages. You need to
 #### Prerequisites
 
 1. Create a self-built app in [WeCom Admin Console](https://work.weixin.qq.com/wework_admin/frame#apps)
-2. In the app settings, go to "API Receive" (API接收) and configure:
-   - **URL**: `https://<your-gateway-host>/plugins/wecom/agent/<accountId>`（例如 `/plugins/wecom/agent/default`）；单账号模式也可使用 `/plugins/wecom/agent`
-   - **Token**: auto-generated or custom
-   - **EncodingAESKey**: auto-generated or custom
-3. Note down the **CorpID**, **CorpSecret**, **Token**, and **EncodingAESKey**
+2. Note down the **CorpID**, **CorpSecret** (from app settings), and **AgentId**
+3. In the app settings, go to "API Receive" (API接收):
+   - Note down the **Token** and **EncodingAESKey** (auto-generated or custom)
+   - **Do NOT click save yet** — WeCom will verify the callback URL immediately when you save
+
+#### Setup Steps
 
-#### Quick Setup
+> **Important**: You must configure the Gateway **before** saving the callback URL in WeCom admin console. WeCom sends a verification request (GET with `echostr`) immediately when you save, and the Gateway needs the `token` and `encodingAESKey` to decrypt and respond correctly.
+
+**Step 1: Configure Gateway**
 
 ```shell
 openclaw config set channels.wecom.agent.corpId <YOUR_CORP_ID>
@@ -193,6 +212,13 @@ openclaw config set channels.wecom.enabled true
 openclaw gateway restart
 ```
 
+**Step 2: Save callback URL in WeCom admin console**
+
+Go back to the "API Receive" settings and enter the callback URL:
+- **URL**: `https://<your-gateway-host>/plugins/wecom/agent/<accountId>` (e.g. `/plugins/wecom/agent/default`); single-account mode can also use `/plugins/wecom/agent`
+
+Now click save — the verification should pass.
+
 #### JSON Configuration
 
 ```json
@@ -255,6 +281,7 @@ The plugin uses a **Bot-first, Agent-fallback** strategy for outbound message de
 This means:
 - **Agent-only accounts** (no Bot configured) can still send proactive messages, cron deliveries, and broadcasts
 - **Target formats** like `party:1`, `tag:Ops`, `user:zhangsan` are fully supported in both paths
+- **Media fallback**: when Bot WS is unavailable, media files are downloaded, uploaded to WeCom via Agent API, then sent; if upload fails, falls back to text + URL
 - No manual switching needed — the plugin handles fallback transparently
 
 ### Using Both Modes Together
@@ -320,7 +347,7 @@ Use `accounts` to configure multiple WeCom accounts, each with optional bot and/
 }
 ```
 
-> **Note**: In multi-account mode, `routing.failClosedOnDefaultRoute` defaults to `true` — accounts without explicit `bindings` will not fall back to the default agent. Configure bindings for each account:
+> **Note**: In multi-account mode, accounts without explicit `bindings` will not fall back to the default agent. Configure bindings for each account:
 > ```json
 > {
 >   "bindings": [
@@ -355,12 +382,6 @@ Dynamic Agent routing automatically creates isolated agents per user or group, e
 | `channels.wecom.dynamicAgents.groupEnabled` | Enable dynamic agent for group chats | `true` |
 | `channels.wecom.dynamicAgents.adminUsers` | Admin users (bypass dynamic routing, use main agent) | `[]` |
 
-### Routing Configuration
-
-| Config Path | Description | Default |
-|---|---|---|
-| `channels.wecom.routing.failClosedOnDefaultRoute` | Reject unbound accounts from falling back to default agent | `true` (multi-account) / `false` (single account) |
-
 ---
 
 ## 🔒 Access Control
@@ -465,17 +486,19 @@ The plugin supports scheduled message delivery via OpenClaw's built-in Cron serv
 The `delivery.to` field supports the following target formats:
 
 | Format | Target | Example |
-|--------|--------|---------|
+|--------|--------|--------|
 | `party:<id>` | Department (all members) | `party:1` (root dept = all employees) |
-| `dept:<id>` | Department (alias) | `dept:5` |
+| `dept:<id>` | Department (alias for party) | `dept:5` |
 | `tag:<id>` | Tag group | `tag:Ops` |
 | `user:<id>` | Specific user | `user:zhangsan` |
 | `group:<id>` | External group chat | `group:wr123abc` |
-| `chat:<id>` | Group chat (alias) | `chat:wc456def` |
+| `chat:<id>` | Group chat (alias for group) | `chat:wc456def` |
 | Pure number | Auto-detected as department | `1` → `party:1` |
 | `wr...` / `wc...` | Auto-detected as group chat | `wr123` → `chatid` |
 | Other string | Auto-detected as user | `zhangsan` → `touser` |
 
+> **Namespace prefixes** (`wecom:`, `qywx:`, `wework:`, `wechatwork:`, `wecom-agent:`) are automatically stripped before parsing.
+
 ### Method 1: CLI (Recommended — takes effect immediately)
 
 ```shell
@@ -483,20 +506,26 @@ openclaw cron add \
   --name "daily-report" \
   --agent main \
   --cron "0 9 * * 1-5" \
+  --tz "Asia/Shanghai" \
   --message "Good morning! Here is your daily briefing." \
-  --deliver \
-  --delivery-channel wecom \
-  --delivery-to "party:1"
+  --announce \
+  --channel wecom \
+  --to "party:1"
 ```
 
+> **Note**: `--announce` enables delivery mode (broadcasts the AI response to the target chat). Use `--no-deliver` to keep output internal. The deprecated `--deliver` flag is an alias for `--announce`.
+
 Common CLI commands:
 
 ```shell
 openclaw cron list              # List all cron jobs
+openclaw cron show <id>         # Show job details
 openclaw cron enable <id>       # Enable a job
 openclaw cron disable <id>      # Disable a job
 openclaw cron remove <id>       # Remove a job
 openclaw cron run <id>          # Manually trigger a job
+openclaw cron runs --id <id>    # View run history
+openclaw cron edit <id> --message "New prompt"  # Edit a job
 ```
 
 ### Method 2: Edit `jobs.json` (requires gateway restart)
@@ -517,10 +546,7 @@ File path: `~/.openclaw/cron/jobs.json`
       "wakeMode": "now",
       "payload": {
         "kind": "agentTurn",
-        "message": "Generate today's briefing and send it.",
-        "deliver": true,
-        "channel": "wecom",
-        "to": "party:1"
+        "message": "Generate today's briefing and send it."
       },
       "delivery": {
         "mode": "announce",

package/dist/index.js

@@ -5,57 +5,6 @@ import { setWeComRuntime } from "./src/runtime.js";
 import { CHANNEL_ID, WEBHOOK_PATHS } from "./src/const.js";
 import { createWecomAgentWebhookHandler } from "./src/agent/webhook.js";
 import { handleWecomWebhookRequest } from "./src/webhook/index.js";
-// ============================================================================
-// 需要自动注入 tools.alsoAllow 的工具名列表
-// ============================================================================
-const REQUIRED_ALSO_ALLOW_TOOLS = ["wecom_mcp"];
-/**
- * 确保 tools.alsoAllow 中包含插件所需的工具名。
- *
- * 逻辑：
- *  1. 读取当前配置
- *  2. 检查 tools.alsoAllow 中是否已包含所需工具
- *  3. 如有缺失，合并写入配置文件
- *
- * 幂等操作——重复调用不会产生副作用。
- * 若 tools.allow 已显式设置（与 alsoAllow 互斥），则跳过注入并打印提示。
- */
-async function ensureToolsAlsoAllow(api) {
-    try {
-        const cfg = api.runtime.config.loadConfig();
-        const tools = cfg.tools ?? {};
-        // 如果用户显式配置了 tools.allow（全量白名单），则 alsoAllow 与之互斥，
-        // 此时不应自动追加 alsoAllow，避免产生校验冲突
-        if (tools.allow && tools.allow.length > 0) {
-            const missing = REQUIRED_ALSO_ALLOW_TOOLS.filter((t) => !tools.allow.includes(t));
-            if (missing.length > 0) {
-                console.warn(`[wecom] tools.allow 已显式设置，无法自动注入 alsoAllow。` +
-                    `请手动将 ${JSON.stringify(missing)} 加入 tools.allow。`);
-            }
-            return;
-        }
-        const existing = tools.alsoAllow ?? [];
-        const missing = REQUIRED_ALSO_ALLOW_TOOLS.filter((t) => !existing.includes(t));
-        if (missing.length === 0) {
-            // 所有工具已在白名单中，无需操作
-            return;
-        }
-        const merged = [...existing, ...missing];
-        const nextConfig = {
-            ...cfg,
-            tools: {
-                ...tools,
-                alsoAllow: merged,
-            },
-        };
-        await api.runtime.config.writeConfigFile(nextConfig);
-        console.log(`[wecom] 已自动将 ${JSON.stringify(missing)} 加入 tools.alsoAllow`);
-    }
-    catch (err) {
-        // 配置写入失败不应阻断 Gateway 启动
-        console.error(`[wecom] 自动注入 tools.alsoAllow 失败:`, err instanceof Error ? err.message : String(err));
-    }
-}
 const plugin = {
     id: "wecom-openclaw-plugin",
     name: "企业微信",
@@ -91,11 +40,6 @@ const plugin = {
                 match: "prefix",
             });
         }
-        // ── Gateway 启动时自动确保 tools.alsoAllow 包含 wecom_mcp ──────────
-        // 在 gateway_start 阶段检测并写入，保证插件安装/更新后首次启动即生效
-        // api.on("gateway_start", async () => {
-        //   await ensureToolsAlsoAllow(api);
-        // });
         // 注入媒体发送指令和文件大小限制提示词（仅对企业微信 channel 生效）
         api.on("before_prompt_build", (_event, ctx) => {
             // 只在企业微信 channel 的会话中注入，避免影响其他 channel 插件

package/dist/src/agent/handler.js

@@ -9,7 +9,7 @@ import { extractMsgType, extractFromUser, extractContent, extractChatId, extract
 import { sendText, downloadMedia, uploadMedia, sendMedia as sendAgentMedia } from "./api-client.js";
 import { buildWecomUnauthorizedCommandPrompt, resolveWecomCommandAuthorization } from "../shared/command-auth.js";
 import { processDynamicRouting } from "../dynamic-routing.js";
-import { DEFAULT_MEDIA_MAX_MB } from "../const.js";
+import { CHANNEL_ID, DEFAULT_MEDIA_MAX_MB } from "../const.js";
 function resolveWecomMediaMaxBytes(config) {
     return config.channels?.wecom?.media?.maxBytes ?? DEFAULT_MEDIA_MAX_MB * 1024 * 1024;
 }
@@ -404,9 +404,9 @@ async function processAgentMessage(params) {
         ConversationLabel: fromLabel,
         SenderName: fromUser,
         SenderId: fromUser,
-        Provider: "wecom",
+        Provider: CHANNEL_ID,
         Surface: "webchat",
-        OriginatingChannel: "wecom",
+        OriginatingChannel: CHANNEL_ID,
         // 标记为 Agent 会话的回复路由目标，避免与 Bot 会话混淆：
         // - 用于让 /new /reset 这类命令回执不被 Bot 侧策略拦截
         // - 群聊场景也统一路由为私信触发者（与 deliver 策略一致）

package/dist/src/channel.js

@@ -432,7 +432,13 @@ export const wecomPlugin = {
                         WEBHOOK_PATHS.AGENT_PLUGIN,
                         WEBHOOK_PATHS.AGENT,
                     ]
-                    : [WEBHOOK_PATHS.AGENT_PLUGIN, WEBHOOK_PATHS.AGENT];
+                    : [
+                        // 单账号模式：同时注册 /default 路径以支持显式指定
+                        WEBHOOK_PATHS.AGENT_PLUGIN,
+                        WEBHOOK_PATHS.AGENT,
+                        `${WEBHOOK_PATHS.AGENT_PLUGIN}/${DEFAULT_ACCOUNT_ID}`,
+                        `${WEBHOOK_PATHS.AGENT}/${DEFAULT_ACCOUNT_ID}`,
+                    ];
                 for (const p of paths) {
                     registerAgentWebhookTarget({
                         agent,

package/dist/src/group-policy.js

@@ -76,21 +76,11 @@ function isGroupSenderAllowed(params) {
  * @returns 检查结果，包含是否允许继续处理
  */
 export function checkGroupPolicy(params) {
-    const { chatId, senderId, account, config, runtime } = params;
-    const wecomConfig = (config.channels?.[CHANNEL_ID] ?? {});
-    const defaultGroupPolicy = config.channels?.[CHANNEL_ID]?.groupPolicy;
-    const groupPolicy = account.config.groupPolicy ?? defaultGroupPolicy ?? "open";
-    // const { groupPolicy, providerMissingFallbackApplied } = resolveOpenProviderRuntimeGroupPolicy({
-    //   providerConfigPresent: config.channels?.[CHANNEL_ID] !== undefined,
-    //   groupPolicy: wecomConfig.groupPolicy,
-    //   defaultGroupPolicy,
-    // });
-    // warnMissingProviderGroupPolicyFallbackOnce({
-    //   providerMissingFallbackApplied,
-    //   providerKey: CHANNEL_ID,
-    //   accountId: account.accountId,
-    //   log: (msg) => runtime.log?.(msg),
-    // });
+    const { chatId, senderId, account, runtime } = params;
+    // 使用 account.config（已经过多账号合并），而非顶层 config.channels.wecom
+    // 避免多账户模式下 groupAllowFrom / groups 等字段取不到账号级配置
+    const wecomConfig = account.config;
+    const groupPolicy = wecomConfig.groupPolicy ?? "open";
     const groupAllowFrom = wecomConfig.groupAllowFrom ?? [];
     const groupAllowed = isWeComGroupAllowed({
         groupPolicy,

package/dist/src/monitor.js

@@ -22,7 +22,7 @@ import { getWeComRuntime } from "./runtime.js";
 import { getDefaultMediaLocalRoots, resolveStateDir } from "./openclaw-compat.js";
 import { CHANNEL_ID, THINKING_MESSAGE, MEDIA_IMAGE_PLACEHOLDER, MEDIA_DOCUMENT_PLACEHOLDER, WS_HEARTBEAT_INTERVAL_MS, WS_MAX_RECONNECT_ATTEMPTS, WS_MAX_AUTH_FAILURE_ATTEMPTS, EVENT_ENTER_CHECK_UPDATE, CMD_ENTER_EVENT_REPLY, SCENE_WECOM_OPENCLAW, } from "./const.js";
 import { parseMessageContent } from "./message-parser.js";
-import { sendWeComReply, sendWeComReplyNonBlocking, StreamExpiredError } from "./message-sender.js";
+import { sendWeComReply, StreamExpiredError } from "./message-sender.js";
 import { downloadAndSaveImages, downloadAndSaveFiles } from "./media-handler.js";
 import { uploadAndSendMedia } from "./media-uploader.js";
 import { maskTemplateCardBlocks } from "./template-card-parser.js";
@@ -32,14 +32,6 @@ import { checkDmPolicy } from "./dm-policy.js";
 import { setWeComWebSocket, setMessageState, deleteMessageState, setReqIdForChat, warmupReqIdStore, startMessageStateCleanup, stopMessageStateCleanup, cleanupAccount, } from "./state-manager.js";
 import { PLUGIN_VERSION } from "./version.js";
 import { processDynamicRouting } from "./dynamic-routing.js";
-/**
- * 去除文本中的 `<think>...</think>` 标签（支持跨行），返回剩余可见文本。
- * 用于判断大模型回复中是否包含实际用户可见内容（而非仅有 thinking 推理过程）。
- */
-function stripThinkTags(text) {
-    return text;
-    // return text.replace(/<think>[\s\S]*?<\/think>/g, "").trim();
-}
 // ============================================================================
 // 媒体本地路径白名单扩展
 // ============================================================================
@@ -270,7 +262,7 @@ async function finishThinkingStream(ctx) {
     const { wsClient, frame, state, runtime } = ctx;
     const body = frame.body;
     const chatId = body.chatid || body.from.userid;
-    const visibleText = stripThinkTags(state.accumulatedText);
+    const visibleText = state.accumulatedText;
     let finishText = state.accumulatedText;
     if (visibleText) {
         finishText = state.accumulatedText;
@@ -348,20 +340,6 @@ async function routeAndDispatchMessage(params) {
         await core.channel.reply.dispatchReplyWithBufferedBlockDispatcher({
             ctx: ctxPayload,
             cfg: config,
-            // replyResolver: async (ctx, opts) => {
-            //   const startTime = Date.now();
-            //   const TEN_MINUTES = 7 * 60 * 1000;
-            //
-            //   console.log('开始输出内容');
-            //
-            //   while (Date.now() - startTime < TEN_MINUTES) {
-            //     // 每隔一段时间发送一个 block reply
-            //     await opts?.onBlockReply?.({ text: `输出内容 ${new Date().toISOString()}` });
-            //     await new Promise(resolve => setTimeout(resolve, 30 * 1000)); // 每5秒输出一次
-            //   }
-            //
-            //   return { text: "10分钟输出完成" }; // 最终回复
-            // },
             dispatcherOptions: {
                 onReplyStart: async () => {
                     if (!isShowThink && state.streamId && !state.accumulatedText) {
@@ -375,10 +353,10 @@ async function routeAndDispatchMessage(params) {
                     }
                 },
                 deliver: async (payload, info) => {
-                    // runtime.log?.(`[openclaw -> plugin] kind=${info.kind}, text=${payload.text ?? ''}, mediaUrl=${payload.mediaUrl ?? ''}, mediaUrls=${JSON.stringify(payload.mediaUrls ?? [])}`);
-                    // 累积文本（快速指令等非流式回复只走 deliver，不走 onPartialReply）
+                    runtime.log?.(`[openclaw -> plugin] kind=${info.kind}, text=${payload.text ?? ''}, mediaUrl=${payload.mediaUrl ?? ''}, mediaUrls=${JSON.stringify(payload.mediaUrls ?? [])}`);
+                    // 累积文本
                     if (payload.text) {
-                        state.accumulatedText = payload.text;
+                        state.accumulatedText += `${payload.text || ''}`;
                     }
                     // 发送媒体（统一走主动发送）
                     const mediaUrls = payload.mediaUrls?.length ? payload.mediaUrls : payload.mediaUrl ? [payload.mediaUrl] : [];
@@ -398,41 +376,30 @@ async function routeAndDispatchMessage(params) {
                             runtime.error?.(`[wecom] sendMediaBatch threw: ${errMsg}`);
                         }
                     }
-                    // 中间帧流式更新已迁移到 onPartialReply（token 级别，更细粒度）
-                    // deliver 只负责文本累积和媒体发送
-                },
-                onError: (err, info) => {
-                    runtime.error?.(`[wecom] ${info.kind} reply failed: ${String(err)}`);
-                },
-            },
-            // onPartialReply：token 级别的流式 UI 刷新
-            // 使用 SDK 的 replyStreamNonBlocking：上一条同 reqId 的消息未收到 ack 时跳过本次发送，
-            // 避免流式中间帧排队积压导致延迟。丢帧在流式场景下可接受（text 是累积文本，最终由 finishThinkingStream 兜底）。
-            replyOptions: {
-                onPartialReply: (data) => {
-                    // 累积文本
-                    if (data.text) {
-                        state.accumulatedText = (data.text || '');
-                    }
-                    if (!state.streamId || !data.text || state.streamExpired)
-                        return;
+                    // 中间帧：有可见文本时流式更新（流式过期后跳过，等 deliver 完成后主动发送）
                     // 使用 maskTemplateCardBlocks 遮罩正在构建中的模板卡片代码块，
                     // 避免 JSON 源码在流式输出过程中暴露给终端用户
-                    const displayText = maskTemplateCardBlocks(data.text, (...args) => runtime.log?.(...args));
-                    if (displayText !== data.text) {
-                        // runtime.log?.(`[wecom][template-card] Partial-reply masked: original=${data.text.length}chars, masked=${displayText.length}chars`);
-                    }
-                    // fire-and-forget：不阻塞 onPartialReply 回调
-                    sendWeComReplyNonBlocking({
-                        wsClient, frame, text: displayText, runtime,
-                        streamId: state.streamId,
-                    }).catch((err) => {
-                        if (err instanceof StreamExpiredError) {
-                            state.streamExpired = true;
-                            runtime.log?.(`[wecom] Stream expired during partial reply, will fallback to proactive send`);
+                    if (state.accumulatedText && !state.streamExpired) {
+                        try {
+                            const displayText = maskTemplateCardBlocks(state.accumulatedText, (...args) => runtime.log?.(...args));
+                            // if (displayText !== state.accumulatedText) {
+                            //   runtime.log?.(`[wecom][template-card] Mid-frame masked: original=${state.accumulatedText.length}chars, masked=${displayText.length}chars`);
+                            // }
+                            await sendWeComReply({ wsClient, frame, text: displayText, runtime, finish: false, streamId: state.streamId });
                         }
-                        // 其他错误静默忽略，不影响主流程（deliver + finishThinkingStream 兜底）
-                    });
+                        catch (err) {
+                            if (err instanceof StreamExpiredError) {
+                                state.streamExpired = true;
+                                runtime.log?.(`[wecom] Stream expired during intermediate reply, will fallback to proactive send`);
+                            }
+                            else {
+                                throw err;
+                            }
+                        }
+                    }
+                },
+                onError: (err, info) => {
+                    runtime.error?.(`[wecom] ${info.kind} reply failed: ${String(err)}`);
                 },
             },
         });
@@ -628,7 +595,6 @@ function createSdkLogger(runtime, accountId) {
 export async function monitorWeComProvider(options) {
     const { account, config, runtime, abortSignal, setStatus } = options;
     runtime.log?.(`[${account.accountId}] [${PLUGIN_VERSION}] Initializing WSClient with SDK...`);
-    runtime.error?.(`[${account.accountId}] [diag] monitor boot marker: build=20260325-event-debug-1`);
     // 启动消息状态定期清理
     startMessageStateCleanup();
     return new Promise((resolve, reject) => {
@@ -802,7 +768,6 @@ export async function monitorWeComProvider(options) {
                 const eventBody = frame.body;
                 const eventType = eventBody.event?.eventtype;
                 runtime.log?.(`[${account.accountId}] Received event callback: eventtype=${eventType ?? ""}, msgid=${eventBody.msgid ?? ""}`);
-                runtime.error?.(`[${account.accountId}] [diag] event-listener fired: eventtype=${eventType ?? ""}, msgid=${eventBody.msgid ?? ""}`);
                 if (eventType !== "template_card_event") {
                     return;
                 }
@@ -835,7 +800,6 @@ export async function monitorWeComProvider(options) {
             }
         });
         runtime.log?.(`[${account.accountId}] Event listeners attached: message + event(template_card_event)`);
-        runtime.error?.(`[${account.accountId}] [diag] listeners-ready marker`);
         // 启动前预热 reqId 缓存，确保完成后再建立连接，避免 getSync 在预热完成前返回 undefined
         warmupReqIdStore(account.accountId, (...args) => runtime.log?.(...args))
             .then((count) => {

package/dist/src/webhook/gateway.js

@@ -15,6 +15,7 @@ import { monitorState } from "./state.js";
 import { registerWecomWebhookTarget, hasActiveTargets } from "./target.js";
 import { startAgentForStream } from "./monitor.js";
 import { hasMultiAccounts } from "../accounts.js";
+import { DEFAULT_ACCOUNT_ID } from "../openclaw-compat.js";
 import { getWeComRuntime } from "../runtime.js";
 // ============================================================================
 // 全局状态
@@ -51,7 +52,14 @@ function resolveBotRegistrationPaths(params) {
             WEBHOOK_PATHS.BOT_ALT,
         ]);
     }
-    return uniquePaths([WEBHOOK_PATHS.BOT_PLUGIN, WEBHOOK_PATHS.BOT, WEBHOOK_PATHS.BOT_ALT]);
+    // 单账号模式：同时注册 /default 路径以支持显式指定
+    return uniquePaths([
+        WEBHOOK_PATHS.BOT_PLUGIN,
+        WEBHOOK_PATHS.BOT,
+        WEBHOOK_PATHS.BOT_ALT,
+        `${WEBHOOK_PATHS.BOT_PLUGIN}/${DEFAULT_ACCOUNT_ID}`,
+        `${WEBHOOK_PATHS.BOT_ALT}/${DEFAULT_ACCOUNT_ID}`,
+    ]);
 }
 // ============================================================================
 // 公共 API

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@wecom/wecom-openclaw-plugin",
-  "version": "2026.3.31-beta.4",
+  "version": "2026.3.31-beta.5",
   "type": "module",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",