feat: gate WeChat media analysis by user

Preserve Portal/Goose SSE cursor boundaries and recover Finish after reconnect.

AGENTS.md

@@ -29,6 +29,7 @@ bash scripts/check-release-ready.sh
 | MindSpace 公开页 `edit_file` 落盘 + 聊天 Finish 不丢消息 | [docs/regression-guards/mindspace-publish-and-chat-finish.md](docs/regression-guards/mindspace-publish-and-chat-finish.md) |
 | MindSpace remote 页面 sync + storage 缺失缩略图 fallback | [docs/regression-guards/mindspace-remote-page-sync-and-thumbnail.md](docs/regression-guards/mindspace-remote-page-sync-and-thumbnail.md) |
 | Page Data 数据集注册、绑定与交付验收 | [docs/regression-guards/page-data-delivery-contract.md](docs/regression-guards/page-data-delivery-contract.md) |
+| H5 SSE 断线续播、Portal/Goose 游标映射与 Finish 终态恢复 | [docs/regression-guards/h5-session-stream-replay.md](docs/regression-guards/h5-session-stream-replay.md) |
 
 索引：[docs/regression-guards/README.md](docs/regression-guards/README.md)
 
@@ -38,6 +39,7 @@ bash scripts/check-release-ready.sh
 npm run verify:mindspace-publish-guards
 npm run verify:mindspace-publish-guards:full
 npm run verify:mindspace-page-sync-guards
+npm run verify:h5-session-patches
 ```
 
 发版脚本（`scripts/release-portal-runtime-prod.sh`）在未 `--skip-tests` 时也会执行相关 verify。
@@ -51,6 +53,7 @@ npm run verify:mindspace-page-sync-guards
 - `server.mjs` - session snapshot 需 `hint_mc` 且 `hint_ua` 才走缓存
 - `mindspace-pages.mjs` - storage 缺失时 HTML 页回退读 workspace `relative_path`
 - `mindspace-page-sync-service.mjs` + `server.mjs` - remote 模式也必须 sync public HTML
+- `session-stream.mjs` + `session-stream-store.mjs` + `tkmind-proxy.mjs` - Portal replay ID 不得直接作为 Goose `Last-Event-ID`
 
 代码内搜索 `REGRESSION GUARD` 可定位所有内联说明。
 

docs/local-analytics.md

@@ -24,7 +24,11 @@ local Umami service at `http://127.0.0.1:3100`; it does not contact 103/105.
    ```
 
 4. Restart the local Memind server. Full generated HTML pages will receive a
-   same-origin `/analytics/script.js` tracker and a `page_view` event with
+   same-origin `/analytics/script.js` tracker. The tracker identifies the
+   visitor with a stable pseudonymous owner ID before sending a standard Umami
+   page view, so Users and Pageviews are populated. The Identify properties
+   include the readable Memind username and current public page URL for
+   operational analytics. Click, form, scroll, and engagement events retain the
    pseudonymous `owner_id`, `page_id`, and `channel` dimensions.
 
 The integration is fail-open: missing configuration, disabled analytics, or a

docs/regression-guards/README.md

@@ -9,12 +9,14 @@
 | [mindspace-publish-and-chat-finish.md](./mindspace-publish-and-chat-finish.md) | ① `edit_file` 覆盖 `public/*.html` ② Finish 后聊天不清空、不暴露 agent 内部前缀 |
 | [mindspace-remote-page-sync-and-thumbnail.md](./mindspace-remote-page-sync-and-thumbnail.md) | ① remote 模式 public HTML 入库 sync ② storage 缺失时缩略图/读页回退 workspace HTML |
 | [page-data-delivery-contract.md](./page-data-delivery-contract.md) | 数据集注册、绑定、真实 page UUID 与交付验收 |
+| [h5-session-stream-replay.md](./h5-session-stream-replay.md) | Portal/Goose SSE 游标映射、断线续播与 Finish 终态恢复 |
 
 ## 自动化
 
 ```bash
 npm run verify:mindspace-publish-guards        # 推荐：改相关代码后
 npm run verify:mindspace-publish-guards:full   # 发 Portal runtime 后
+npm run verify:h5-session-patches              # H5 会话、Agent Run 与 SSE 续播
 ```
 
 ## 新增守卫时

docs/regression-guards/h5-session-stream-replay.md

@@ -0,0 +1,40 @@
+# H5 会话 SSE 断线续播守卫
+
+## 症状
+
+后台 `h5_agent_runs` 已经是 `succeeded`、Goose 也已发送 `Finish`，但 H5 输入区仍显示“正在执行任务…”。常见触发条件是手机切后台、锁屏或网络切换发生在 `ActiveRequests` 与 `Finish` 之间。
+
+## 根因
+
+Portal 对 SSE 帧使用本地持久化游标。该游标可能是 Portal 生成的 UUID，并不一定是 Goose 能识别的上游事件 ID。若重连时把 Portal UUID 原样作为 `Last-Event-ID` 发送给 Goose，Goose 无法续播，浏览器就收不到遗漏的 `Finish`。
+
+## 必须保留的行为
+
+1. Portal 游标只用于查询 `h5_session_stream_events`，不得原样传给 Goose。
+2. 连接 Goose 时，只能使用已持久化的 `upstream_event_id`。
+3. 当前游标或重放尾部没有可用 `upstream_event_id` 时，必须省略 `Last-Event-ID`，让 Goose 从权威事件流重新回放。
+4. 本地重放最后一条已经是 `Finish` 或 `Error` 时，不再连接 Goose。
+5. 重放不得破坏现有消息 ID 合并、Finish 幂等计费和 MindSpace Finish 同步守卫。
+
+## 关键路径
+
+- `session-stream.mjs`
+- `session-stream-store.mjs`
+- `tkmind-proxy.mjs`
+- `src/hooks/useTKMindChat.ts`
+
+## 回归测试
+
+```bash
+node --test session-stream.test.mjs session-stream-store.test.mjs chat-agent-run-gate.test.mjs
+node --test --test-name-pattern='proxySessionEvents' tkmind-proxy.test.mjs
+npm run verify:h5-session-patches
+node --test billing-session-concurrency.test.mjs
+```
+
+必须覆盖：
+
+- Portal 游标存在但 `upstream_event_id` 为空时，不向 Goose 发送错误游标，并能收到 `Finish`。
+- 有映射时，Portal 游标正确转换成 Goose 游标。
+- 多轮会话中选择最新可映射的上游游标，不被上一轮 Finish 截断。
+- Finish 重放不重复扣费，消息合并与页面同步守卫继续通过。

mindspace-analytics.mjs

@@ -95,10 +95,10 @@ export function injectMindSpaceAnalytics(html, {
   if (source.includes(ANALYTICS_MARKER)) return source;
   const owner = pseudonymizeAnalyticsId(ownerId, config.idSecret);
   if (!owner) return source;
-  // Public page source must not contain a readable account name. The stable
-  // pseudonym and coarse plan segment are sufficient for page analytics;
-  // readable labels are reserved for server-originated events only.
-  const metadata = { owner_id: owner, owner_segment: String(ownerSegment || 'unknown'), page_id: String(pageId || ''), publication_id: String(publicationId || ''), channel };
+  // The stable pseudonym remains the Umami identity key. The readable username
+  // is an explicitly enabled analytics property so operators can recognize the
+  // Memind user, while the current public page URL is resolved in the browser.
+  const metadata = { owner_id: owner, username: resolveAnalyticsOwnerLabel({ displayName: ownerLabel }), owner_segment: String(ownerSegment || 'unknown'), page_id: String(pageId || ''), publication_id: String(publicationId || ''), channel };
   const attrs = [
     ANALYTICS_MARKER,
     `data-website-id="${config.websiteId.replaceAll('"', '"')}"`,
@@ -106,7 +106,7 @@ export function injectMindSpaceAnalytics(html, {
     `data-host-url="${config.hostPath}"`,
   ];
   if (config.domains) attrs.push(`data-domains="${config.domains.replaceAll('"', '"')}"`);
-  const block = `<script ${attrs.join(' ')} defer src="${config.scriptPath}"></script><script ${ANALYTICS_MARKER}>(function(){var d=${jsonForInlineScript(metadata)},seen={};function t(n,x){if(!window.umami||typeof window.umami.track!=='function')return;var p=Object.assign({},d,{page_url:location.href,page_title:document.title},x||{});window.umami.track(n,p);}function once(n,x){if(seen[n])return;seen[n]=1;t(n,x);}function ready(){t('page_view');document.addEventListener('click',function(e){var el=e.target&&e.target.closest?e.target.closest('a,button,[role="button"],[data-umami-event]'):null;if(!el)return;var custom=el.getAttribute('data-umami-event');var href=el.tagName==='A'?el.getAttribute('href')||'':'';t(custom||'page_click',{element:el.tagName.toLowerCase(),element_id:el.id||'',event_label:(custom||el.getAttribute('aria-label')||'').slice(0,100),target_url:href.slice(0,500)});},{passive:true});document.addEventListener('submit',function(e){var form=e.target;t('page_form_submit',{form_id:form&&form.id||'',form_action:form&&form.getAttribute('action')||''});},{passive:true});var marks=[25,50,75,90];function scroll(){var h=document.documentElement.scrollHeight-window.innerHeight;if(h<=0){once('page_scroll_100');return;}var pct=Math.round(window.scrollY/h*100);marks.forEach(function(m){if(pct>=m)once('page_scroll_'+m);});}window.addEventListener('scroll',scroll,{passive:true});setTimeout(function(){once('page_engaged_10s');},10000);setTimeout(function(){once('page_engaged_30s');},30000);scroll();}if(document.readyState==='loading'){document.addEventListener('DOMContentLoaded',ready,{once:true});}else{ready();}})();</script>`;
+  const block = `<script ${attrs.join(' ')} defer src="${config.scriptPath}"></script><script ${ANALYTICS_MARKER}>(function(){var d=${jsonForInlineScript(metadata)},seen={};function t(n,x){if(!window.umami||typeof window.umami.track!=='function')return;var p=Object.assign({},d,{page_url:location.href,page_title:document.title},x||{});window.umami.track(n,p);}function identify(){if(!window.umami||typeof window.umami.identify!=='function')return;window.umami.identify(d.owner_id,{username:d.username,memind_page_url:location.href,owner_segment:d.owner_segment,channel:d.channel});}function pageview(){if(!window.umami||typeof window.umami.track!=='function')return;window.umami.track();}function once(n,x){if(seen[n])return;seen[n]=1;t(n,x);}function ready(){identify();pageview();document.addEventListener('click',function(e){var el=e.target&&e.target.closest?e.target.closest('a,button,[role="button"],[data-umami-event]'):null;if(!el)return;var custom=el.getAttribute('data-umami-event');var href=el.tagName==='A'?el.getAttribute('href')||'':'';t(custom||'page_click',{element:el.tagName.toLowerCase(),element_id:el.id||'',event_label:(custom||el.getAttribute('aria-label')||'').slice(0,100),target_url:href.slice(0,500)});},{passive:true});document.addEventListener('submit',function(e){var form=e.target;t('page_form_submit',{form_id:form&&form.id||'',form_action:form&&form.getAttribute('action')||''});},{passive:true});var marks=[25,50,75,90];function scroll(){var h=document.documentElement.scrollHeight-window.innerHeight;if(h<=0){once('page_scroll_100');return;}var pct=Math.round(window.scrollY/h*100);marks.forEach(function(m){if(pct>=m)once('page_scroll_'+m);});}window.addEventListener('scroll',scroll,{passive:true});setTimeout(function(){once('page_engaged_10s');},10000);setTimeout(function(){once('page_engaged_30s');},30000);scroll();}if(document.readyState==='loading'){document.addEventListener('DOMContentLoaded',ready,{once:true});}else{ready();}})();</script>`;
   if (/<\/head>/i.test(source)) return source.replace(/<\/head>/i, `${block}</head>`);
   return source.replace(/<body\b/i, `${block}<body`);
 }

mindspace-analytics.test.mjs

@@ -1,5 +1,6 @@
 import test from 'node:test';
 import assert from 'node:assert/strict';
+import vm from 'node:vm';
 
 import {
   injectMindSpaceAnalytics,
@@ -64,10 +65,14 @@ test('injects one local same-origin tracker with page dimensions', () => {
   assert.match(out, /src="\/analytics\/script\.js"/);
   assert.match(out, /data-host-url="\/analytics"/);
   assert.match(out, /data-auto-track="false"/);
+  assert.match(out, /window\.umami\.identify\(d\.owner_id,\{username:d\.username,memind_page_url:location\.href,owner_segment:d\.owner_segment,channel:d\.channel\}\)/);
+  assert.match(out, /function pageview\(\).*window\.umami\.track\(\)/);
+  assert.ok(out.indexOf('identify();pageview();') > 0);
+  assert.doesNotMatch(out, /t\('page_view'\)/);
   assert.match(out, /page_id/);
   assert.match(out, /owner_segment/);
   assert.doesNotMatch(out, /owner_label/);
-  assert.doesNotMatch(out, /张三/);
+  assert.match(out, /"username":"张三"/);
   assert.match(out, /page_click/);
   assert.match(out, /page_form_submit/);
   assert.match(out, /page_scroll_/);
@@ -76,6 +81,55 @@ test('injects one local same-origin tracker with page dimensions', () => {
   assert.equal(injectMindSpaceAnalytics(out, { ownerId: 'user-123', config: { enabled: true, websiteId: 'local-website', idSecret: 'secret' } }), out);
 });
 
+test('identifies the pseudonymous owner before sending a standard page view', () => {
+  const out = injectMindSpaceAnalytics('<!doctype html><html><head></head><body></body></html>', {
+    ownerId: 'user-123',
+    ownerSegment: 'plan:pro',
+    ownerLabel: '张三',
+    channel: 'h5',
+    config: {
+      enabled: true,
+      websiteId: 'local-website',
+      idSecret: 'secret',
+      scriptPath: '/analytics/script.js',
+      hostPath: '/analytics',
+    },
+  });
+  const inlineScript = out.match(/<script data-memind-analytics="1">([\s\S]*?)<\/script>/)?.[1];
+  assert.ok(inlineScript);
+
+  const calls = [];
+  vm.runInNewContext(inlineScript, {
+    window: {
+      umami: {
+        identify: (...args) => calls.push(['identify', ...args]),
+        track: (...args) => calls.push(['track', ...args]),
+      },
+      innerHeight: 800,
+      scrollY: 0,
+      addEventListener: () => {},
+    },
+    document: {
+      readyState: 'complete',
+      title: 'Demo',
+      documentElement: { scrollHeight: 1600 },
+      addEventListener: () => {},
+    },
+    location: { href: 'https://m.tkmind.cn/MindSpace/demo/public/page.html' },
+    setTimeout: () => {},
+  });
+
+  assert.deepEqual(JSON.parse(JSON.stringify(calls)), [
+    ['identify', pseudonymizeAnalyticsId('user-123', 'secret'), {
+      username: '张三',
+      memind_page_url: 'https://m.tkmind.cn/MindSpace/demo/public/page.html',
+      owner_segment: 'plan:pro',
+      channel: 'h5',
+    }],
+    ['track'],
+  ]);
+});
+
 test('does not alter non-full-html or disabled pages', () => {
   const fragment = '<div>hello</div>';
   assert.equal(injectMindSpaceAnalytics(fragment, { ownerId: 'u', config: { enabled: true, websiteId: 'w', idSecret: 's' } }), fragment);

session-stream-store.mjs

@@ -8,6 +8,17 @@ function safeJsonParse(value, fallback = null) {
   }
 }
 
+function mapEventRow(row) {
+  if (!row) return null;
+  return {
+    id: row.id,
+    eventType: row.event_type,
+    payload: safeJsonParse(row.payload_json, null),
+    upstreamEventId: row.upstream_event_id ?? null,
+    createdAt: Number(row.created_at),
+  };
+}
+
 export function createSessionStreamStore({ pool }) {
   if (!pool) {
     throw new Error('createSessionStreamStore requires pool');
@@ -52,9 +63,10 @@ export function createSessionStreamStore({ pool }) {
 
     let afterCreatedAt = null;
     let cursorMiss = false;
+    let cursorEvent = null;
     if (afterEventId) {
       const [cursorRows] = await pool.query(
-        `SELECT created_at
+        `SELECT id, event_type, payload_json, upstream_event_id, created_at
          FROM h5_session_stream_events
          WHERE id = ? AND agent_session_id = ? AND user_id = ?
          LIMIT 1`,
@@ -63,7 +75,8 @@ export function createSessionStreamStore({ pool }) {
       if (!cursorRows[0]) {
         cursorMiss = true;
       } else {
-        afterCreatedAt = Number(cursorRows[0].created_at);
+        cursorEvent = mapEventRow(cursorRows[0]);
+        afterCreatedAt = cursorEvent.createdAt;
       }
     }
 
@@ -85,14 +98,9 @@ export function createSessionStreamStore({ pool }) {
     );
 
     return {
-      events: rows.map((row) => ({
-        id: row.id,
-        eventType: row.event_type,
-        payload: safeJsonParse(row.payload_json, null),
-        upstreamEventId: row.upstream_event_id ?? null,
-        createdAt: Number(row.created_at),
-      })),
+      events: rows.map(mapEventRow),
       cursorMiss,
+      cursorEvent,
     };
   }
 
@@ -107,13 +115,7 @@ export function createSessionStreamStore({ pool }) {
     );
     const row = rows[0];
     if (!row) return null;
-    return {
-      id: row.id,
-      eventType: row.event_type,
-      payload: safeJsonParse(row.payload_json, null),
-      upstreamEventId: row.upstream_event_id ?? null,
-      createdAt: Number(row.created_at),
-    };
+    return mapEventRow(row);
   }
 
   return {

session-stream-store.test.mjs

@@ -23,7 +23,7 @@ function createMockPool() {
       }
       if (sql.includes('WHERE id = ? AND agent_session_id = ?')) {
         const row = events.find((item) => item.id === params[0]);
-        return [[row ? { created_at: row.created_at } : undefined].filter(Boolean)];
+        return [[row].filter(Boolean)];
       }
       if (sql.includes('created_at > ?')) {
         const after = Number(params[2]);
@@ -67,12 +67,14 @@ test('session stream store append and replay with cursor', async () => {
     sessionId: 'sess-1',
     payload: { type: 'Message', message: { role: 'assistant', content: [] } },
     id: 'evt-1',
+    upstreamEventId: 'upstream-1',
   });
   await store.appendEvent({
     userId: 'user-1',
     sessionId: 'sess-1',
     payload: { type: 'Finish' },
     id: 'evt-2',
+    upstreamEventId: 'upstream-2',
   });
 
   const head = await store.listEventsForUser('user-1', 'sess-1');
@@ -82,7 +84,10 @@ test('session stream store append and replay with cursor', async () => {
   assert.equal(tail.events.length, 1);
   assert.equal(tail.events[0].payload.type, 'Finish');
   assert.equal(tail.cursorMiss, false);
+  assert.equal(tail.cursorEvent.id, first.id);
+  assert.equal(tail.cursorEvent.upstreamEventId, 'upstream-1');
 
   const miss = await store.listEventsForUser('user-1', 'sess-1', { afterEventId: 'missing' });
   assert.equal(miss.cursorMiss, true);
+  assert.equal(miss.cursorEvent, null);
 });

session-stream.mjs

@@ -15,6 +15,26 @@ export function parseSessionStreamLastEventId(value) {
   return id || null;
 }
 
+/**
+ * Portal replay ids are local database cursors and are not necessarily valid
+ * Goose SSE ids. Resume Goose from the newest persisted upstream cursor we
+ * actually know about. If none exists, return null so the caller omits
+ * Last-Event-ID and lets Goose replay the authoritative stream from the start.
+ *
+ * @param {Array<{ upstreamEventId?: string | null }>} events
+ * @param {{ upstreamEventId?: string | null } | null} cursorEvent
+ * @returns {string | null}
+ */
+export function resolveUpstreamResumeEventId(events, cursorEvent = null) {
+  const candidates = Array.isArray(events) ? [...events].reverse() : [];
+  if (cursorEvent) candidates.push(cursorEvent);
+  for (const event of candidates) {
+    const id = String(event?.upstreamEventId ?? '').trim();
+    if (id) return id;
+  }
+  return null;
+}
+
 export function isTerminalSessionEvent(event) {
   const type = String(event?.type ?? '').trim();
   return SESSION_STREAM_TERMINAL_TYPES.has(type);

session-stream.test.mjs

@@ -6,6 +6,7 @@ import {
   isTerminalSessionEvent,
   parseSessionSseBlock,
   parseSessionStreamLastEventId,
+  resolveUpstreamResumeEventId,
   shouldPersistSessionStreamEvent,
   shouldSkipUpstreamAfterSessionReplay,
 } from './session-stream.mjs';
@@ -44,3 +45,34 @@ test('terminal detection and upstream skip policy', () => {
   assert.equal(isSessionStreamReplayEnabled({ MEMIND_SESSION_STREAM_REPLAY: '1' }), true);
   assert.equal(parseSessionStreamLastEventId(' abc '), 'abc');
 });
+
+test('upstream resume cursor never falls back to a Portal replay id', () => {
+  assert.equal(
+    resolveUpstreamResumeEventId([], {
+      id: 'portal-uuid',
+      upstreamEventId: null,
+    }),
+    null,
+  );
+  assert.equal(
+    resolveUpstreamResumeEventId([], {
+      id: 'portal-uuid',
+      upstreamEventId: 'upstream-7',
+    }),
+    'upstream-7',
+  );
+});
+
+test('upstream resume cursor uses the newest mappable replay event across turns', () => {
+  assert.equal(
+    resolveUpstreamResumeEventId(
+      [
+        { upstreamEventId: 'upstream-old-finish' },
+        { upstreamEventId: 'upstream-new-message' },
+        { upstreamEventId: null },
+      ],
+      { upstreamEventId: 'upstream-cursor' },
+    ),
+    'upstream-new-message',
+  );
+});

tkmind-proxy.mjs

@@ -44,6 +44,7 @@ import {
   isSessionStreamReplayEnabled,
   parseSessionSseBlock,
   parseSessionStreamLastEventId,
+  resolveUpstreamResumeEventId,
   shouldPersistSessionStreamEvent,
   shouldSkipUpstreamAfterSessionReplay,
 } from './session-stream.mjs';
@@ -2154,6 +2155,7 @@ export function createTkmindProxy({
     let clientClosed = false;
     const replayEnabled = isSessionStreamReplayEnabled() && sessionStreamStore;
     const initialLastEventId = parseSessionStreamLastEventId(req.get('last-event-id'));
+    let upstreamLastEventId = replayEnabled ? null : initialLastEventId;
     const abortUpstream = () => {
       clientClosed = true;
       upstreamAbort.abort();
@@ -2210,17 +2212,22 @@ export function createTkmindProxy({
           if (!res.writableEnded) res.end();
           return;
         }
+        upstreamLastEventId = resolveUpstreamResumeEventId(
+          batch?.events ?? [],
+          batch?.cursorEvent ?? null,
+        );
       }
 
       const pathname = `/sessions/${sessionId}/events`;
       const sessionTarget = await resolveTarget(sessionId);
+      const upstreamHeaders = { Accept: 'text/event-stream' };
+      if (upstreamLastEventId) {
+        upstreamHeaders['Last-Event-ID'] = upstreamLastEventId;
+      }
       const upstream = await apiFetch(sessionTarget, apiSecret, pathname, {
         method: 'GET',
         signal: upstreamAbort.signal,
-        headers: {
-          Accept: 'text/event-stream',
-          'Last-Event-ID': req.get('last-event-id') ?? '',
-        },
+        headers: upstreamHeaders,
       });
       if (clientClosed) return;
 

tkmind-proxy.test.mjs

@@ -533,6 +533,178 @@ test('proxySessionEvents attaches session taxonomy when flag enabled', async ()
   }
 });
 
+test('proxySessionEvents omits an unmapped Portal cursor so Goose can replay Finish', async () => {
+  const previousReplay = process.env.MEMIND_SESSION_STREAM_REPLAY;
+  process.env.MEMIND_SESSION_STREAM_REPLAY = '1';
+  let upstream;
+  let receivedLastEventId = 'not-requested';
+
+  try {
+    upstream = createServer((req, res) => {
+      if (req.method === 'GET' && req.url === '/sessions/session-1/events') {
+        receivedLastEventId = req.headers['last-event-id'] ?? null;
+        res.writeHead(200, { 'Content-Type': 'text/event-stream' });
+        res.end(
+          'id: upstream-finish\n' +
+          'data: {"type":"Finish","reason":"stop","request_id":"req-1"}\n\n',
+        );
+        return;
+      }
+      res.writeHead(404, { 'Content-Type': 'application/json' });
+      res.end(JSON.stringify({ message: `unexpected ${req.method} ${req.url}` }));
+    });
+    const upstreamPort = await listen(upstream);
+    const persisted = [];
+    const proxy = createTkmindProxy({
+      apiTarget: `http://127.0.0.1:${upstreamPort}`,
+      apiSecret: 'test-secret',
+      userAuth: createMemoryTestUserAuth(process.cwd()),
+      sessionStreamStore: {
+        async listEventsForUser(_userId, _sessionId, { afterEventId } = {}) {
+          assert.equal(afterEventId, 'portal-active-request-uuid');
+          return {
+            cursorMiss: false,
+            cursorEvent: {
+              id: 'portal-active-request-uuid',
+              payload: { type: 'ActiveRequests', request_ids: ['req-1'] },
+              upstreamEventId: null,
+              createdAt: 1,
+            },
+            events: [],
+          };
+        },
+        async appendEvent(event) {
+          persisted.push(event);
+        },
+      },
+    });
+
+    const req = new EventEmitter();
+    req.currentUser = { id: 'user-1', username: 'john' };
+    req.get = (name) =>
+      name.toLowerCase() === 'last-event-id' ? 'portal-active-request-uuid' : '';
+    req.once = req.once.bind(req);
+    req.off = req.off.bind(req);
+
+    const chunks = [];
+    const res = new EventEmitter();
+    res.headersSent = false;
+    res.writableEnded = false;
+    res.statusCode = 200;
+    res.status = (code) => {
+      res.statusCode = code;
+      return res;
+    };
+    res.setHeader = () => {};
+    res.flushHeaders = () => {
+      res.headersSent = true;
+    };
+    res.write = (chunk) => {
+      res.headersSent = true;
+      chunks.push(String(chunk));
+      return true;
+    };
+    res.end = () => {
+      res.writableEnded = true;
+    };
+    res.json = () => {
+      throw new Error('json must not be called after SSE started');
+    };
+
+    await proxy.proxySessionEvents(req, res, 'session-1');
+
+    assert.equal(receivedLastEventId, null);
+    assert.match(chunks.join(''), /"type":"Finish"/);
+    await new Promise((resolve) => setImmediate(resolve));
+    assert.equal(persisted.at(-1)?.upstreamEventId, 'upstream-finish');
+  } finally {
+    if (previousReplay == null) delete process.env.MEMIND_SESSION_STREAM_REPLAY;
+    else process.env.MEMIND_SESSION_STREAM_REPLAY = previousReplay;
+    await closeServer(upstream);
+  }
+});
+
+test('proxySessionEvents translates a Portal replay cursor to its Goose cursor', async () => {
+  const previousReplay = process.env.MEMIND_SESSION_STREAM_REPLAY;
+  process.env.MEMIND_SESSION_STREAM_REPLAY = '1';
+  let upstream;
+  let receivedLastEventId = null;
+
+  try {
+    upstream = createServer((req, res) => {
+      if (req.method === 'GET' && req.url === '/sessions/session-1/events') {
+        receivedLastEventId = req.headers['last-event-id'] ?? null;
+        res.writeHead(200, { 'Content-Type': 'text/event-stream' });
+        res.end('id: upstream-finish\ndata: {"type":"Finish","reason":"stop"}\n\n');
+        return;
+      }
+      res.writeHead(404, { 'Content-Type': 'application/json' });
+      res.end(JSON.stringify({ message: `unexpected ${req.method} ${req.url}` }));
+    });
+    const upstreamPort = await listen(upstream);
+    const proxy = createTkmindProxy({
+      apiTarget: `http://127.0.0.1:${upstreamPort}`,
+      apiSecret: 'test-secret',
+      userAuth: createMemoryTestUserAuth(process.cwd()),
+      sessionStreamStore: {
+        async listEventsForUser() {
+          return {
+            cursorMiss: false,
+            cursorEvent: {
+              id: 'portal-message-uuid',
+              payload: { type: 'Message' },
+              upstreamEventId: 'upstream-message-17',
+              createdAt: 1,
+            },
+            events: [],
+          };
+        },
+        async appendEvent() {},
+      },
+    });
+
+    const req = new EventEmitter();
+    req.currentUser = { id: 'user-1', username: 'john' };
+    req.get = (name) => name.toLowerCase() === 'last-event-id' ? 'portal-message-uuid' : '';
+    req.once = req.once.bind(req);
+    req.off = req.off.bind(req);
+
+    const chunks = [];
+    const res = new EventEmitter();
+    res.headersSent = false;
+    res.writableEnded = false;
+    res.statusCode = 200;
+    res.status = (code) => {
+      res.statusCode = code;
+      return res;
+    };
+    res.setHeader = () => {};
+    res.flushHeaders = () => {
+      res.headersSent = true;
+    };
+    res.write = (chunk) => {
+      res.headersSent = true;
+      chunks.push(String(chunk));
+      return true;
+    };
+    res.end = () => {
+      res.writableEnded = true;
+    };
+    res.json = () => {
+      throw new Error('json must not be called after SSE started');
+    };
+
+    await proxy.proxySessionEvents(req, res, 'session-1');
+
+    assert.equal(receivedLastEventId, 'upstream-message-17');
+    assert.match(chunks.join(''), /"type":"Finish"/);
+  } finally {
+    if (previousReplay == null) delete process.env.MEMIND_SESSION_STREAM_REPLAY;
+    else process.env.MEMIND_SESSION_STREAM_REPLAY = previousReplay;
+    await closeServer(upstream);
+  }
+});
+
 test('startSessionForUser resolves memories through Memory V2 facade', async () => {
   let resolveInput = null;
   await withFakeGoosedSession(async ({ apiTarget, workingDir, harnessEntries }) => {

wechat-media.mjs

@@ -9,12 +9,19 @@ const __dirname = path.dirname(fileURLToPath(import.meta.url));
 
 const DEFAULT_WECHAT_MEDIA_URL = 'https://api.weixin.qq.com/cgi-bin/media/get';
 const DEFAULT_MAX_IMAGE_BYTES = 10 * 1024 * 1024;
+const DEFAULT_MAX_ATTACHMENT_BYTES = 30 * 1024 * 1024;
 const ALLOWED_IMAGE_MIME_TYPES = new Map([
   ['image/jpeg', 'jpg'],
   ['image/png', 'png'],
   ['image/webp', 'webp'],
   ['image/gif', 'gif'],
 ]);
+const ALLOWED_ATTACHMENT_EXTENSIONS = new Map([
+  ['.doc', 'application/msword'],
+  ['.docx', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'],
+  ['.xls', 'application/vnd.ms-excel'],
+  ['.xlsx', 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'],
+]);
 
 function resolveImageExtension(contentType = '', fallbackUrl = '') {
   const normalized = String(contentType ?? '')
@@ -44,6 +51,32 @@ function ensureImageWithinLimit(buffer, maxBytes) {
   }
 }
 
+function sanitizeAttachmentFilename(filename = '') {
+  const basename = path.basename(String(filename ?? '').trim()).replace(/[\u0000-\u001f\u007f]/g, '');
+  if (!basename || basename === '.' || basename === '..') {
+    throw new Error('微信文件缺少有效文件名');
+  }
+  const extension = path.extname(basename).toLowerCase();
+  const mimeType = ALLOWED_ATTACHMENT_EXTENSIONS.get(extension);
+  if (!mimeType) {
+    throw new Error('当前服务号文件仅支持 Word（doc/docx）和 Excel（xls/xlsx）');
+  }
+  return {
+    filename: basename.slice(0, 160),
+    extension,
+    mimeType,
+  };
+}
+
+function ensureAttachmentWithinLimit(buffer, maxBytes) {
+  if (!Buffer.isBuffer(buffer) || buffer.length === 0) {
+    throw new Error('微信文件内容为空');
+  }
+  if (buffer.length > maxBytes) {
+    throw new Error(`文件超过大小限制（${maxBytes} bytes）`);
+  }
+}
+
 export async function downloadTemporaryMedia(accessToken, mediaId, { wechatFetch = undiciFetch } = {}) {
   if (!accessToken) throw new Error('缺少微信 access_token');
   if (!mediaId) throw new Error('缺少微信 mediaId');
@@ -154,3 +187,58 @@ export async function persistWechatImage(
     source,
   };
 }
+
+export async function persistWechatAttachment(
+  {
+    userId,
+    appId,
+    openid,
+    msgId,
+    mediaId,
+    filename,
+    publicBaseUrl,
+    maxFileBytes = DEFAULT_MAX_ATTACHMENT_BYTES,
+  },
+  {
+    wechatFetch = undiciFetch,
+    accessToken,
+    h5Root = __dirname,
+  } = {},
+) {
+  if (!userId) throw new Error('缺少 userId');
+  const resolved = sanitizeAttachmentFilename(filename);
+  const downloaded = await downloadTemporaryMedia(accessToken, mediaId, { wechatFetch });
+  ensureAttachmentWithinLimit(downloaded.buffer, maxFileBytes);
+
+  const publishDir = path.join(h5Root, PUBLISH_ROOT_DIR, String(userId), PUBLIC_ZONE_DIR, 'wechat-mp');
+  fs.mkdirSync(publishDir, { recursive: true });
+
+  const timestamp = Date.now();
+  const hash = crypto.createHash('sha1').update(downloaded.buffer).digest('hex').slice(0, 12);
+  const originalStem = path.basename(resolved.filename, resolved.extension)
+    .replace(/[^\p{L}\p{N}._-]+/gu, '_')
+    .replace(/^_+|_+$/g, '')
+    .slice(0, 80) || 'attachment';
+  const identity = [appId || 'wx', openid || 'openid', msgId || timestamp, mediaId || hash]
+    .filter(Boolean)
+    .join('-')
+    .replace(/[^a-zA-Z0-9._-]+/g, '_')
+    .slice(0, 100);
+  const publicFilename = `${originalStem}-${identity}-${hash}${resolved.extension}`;
+  const absolutePath = path.join(publishDir, publicFilename);
+  fs.writeFileSync(absolutePath, downloaded.buffer);
+
+  return {
+    absolutePath,
+    bytes: downloaded.buffer.length,
+    contentType: resolved.mimeType,
+    filename: resolved.filename,
+    publicFilename,
+    publicUrl: buildWechatImagePublicUrl({
+      publicBaseUrl,
+      publishKey: String(userId),
+      filename: publicFilename,
+    }),
+    source: 'wechat_media',
+  };
+}

wechat-mp-config.mjs

@@ -23,6 +23,13 @@ function deriveWechatEndpointFromUrl(baseUrl, suffix) {
   }
 }
 
+function parseCsvList(value = '') {
+  return String(value ?? '')
+    .split(',')
+    .map((item) => item.trim())
+    .filter(Boolean);
+}
+
 export function loadWechatMpConfig(env = process.env) {
   const appId = env.H5_WECHAT_MP_APP_ID?.trim() ?? env.H5_WECHAT_APP_ID?.trim() ?? '';
   const appSecret =
@@ -67,10 +74,13 @@ export function loadWechatMpConfig(env = process.env) {
     mediaPublicBaseUrl:
       env.H5_WECHAT_MP_MEDIA_PUBLIC_BASE_URL?.trim()?.replace(/\/$/, '') || publicBaseUrl,
     maxImageBytes: Math.max(1, Number(env.H5_WECHAT_MP_MAX_IMAGE_BYTES ?? 10 * 1024 * 1024)),
+    maxFileBytes: Math.max(1, Number(env.H5_WECHAT_MP_MAX_FILE_BYTES ?? 30 * 1024 * 1024)),
     acceptVoice: env.H5_WECHAT_MP_ACCEPT_VOICE !== '0',
     acceptImage: env.H5_WECHAT_MP_ACCEPT_IMAGE !== '0',
+    acceptFile: env.H5_WECHAT_MP_ACCEPT_FILE !== '0',
     acceptLocation: env.H5_WECHAT_MP_ACCEPT_LOCATION !== '0',
     acceptLink: env.H5_WECHAT_MP_ACCEPT_LINK !== '0',
+    mediaAnalysisGrayUsers: parseCsvList(env.H5_WECHAT_MP_MEDIA_GRAY_USERS),
     encodingAesKey: env.H5_WECHAT_MP_ENCODING_AES_KEY?.trim() ?? '',
   };
 }

wechat-mp.mjs

@@ -9,7 +9,11 @@ import { resolveSessionAccess } from './session-broker.mjs';
 import { isStubPublicHtmlContent, materializeMissingPublicHtmlWrites } from './mindspace-public-finish-sync.mjs';
 import { loadWechatMpConfig } from './wechat-mp-config.mjs';
 import { buildPublicUrl, PUBLISH_ROOT_DIR } from './user-publish.mjs';
-import { downloadTemporaryMedia, persistWechatImage } from './wechat-media.mjs';
+import {
+  downloadTemporaryMedia,
+  persistWechatAttachment,
+  persistWechatImage,
+} from './wechat-media.mjs';
 import { normalizeWechatName, resolveWechatAddressName } from './wechat/user/display-name.mjs';
 import { buildAckText } from './wechat/ack/ack-provider.mjs';
 import { guardScheduleConfirmationReply } from './wechat/handlers/schedule-guard.mjs';
@@ -100,6 +104,8 @@ function parseWechatMessage(xml) {
     description: parseXmlField(xml, 'Description'),
     url: parseXmlField(xml, 'Url'),
     thumbMediaId: parseXmlField(xml, 'ThumbMediaId'),
+    fileName: parseXmlField(xml, 'FileName') || parseXmlField(xml, 'Filename'),
+    fileSize: parseXmlField(xml, 'FileSize'),
     event: parseXmlField(xml, 'Event').toLowerCase(),
     eventKey: parseXmlField(xml, 'EventKey'),
     latitude: parseXmlField(xml, 'Latitude'),
@@ -1164,6 +1170,23 @@ function normalizeNumber(value) {
   return Number.isFinite(num) ? num : null;
 }
 
+function isWechatMediaGrayUser(user, configuredUsers = []) {
+  const allowlist = Array.isArray(configuredUsers)
+    ? configuredUsers.map((value) => String(value ?? '').trim().toLowerCase()).filter(Boolean)
+    : [];
+  if (allowlist.length === 0) return false;
+  const identities = [
+    user?.userId,
+    user?.username,
+    user?.slug,
+    user?.displayName,
+    user?.nickname,
+  ]
+    .map((value) => String(value ?? '').trim().toLowerCase())
+    .filter(Boolean);
+  return identities.some((identity) => allowlist.includes(identity));
+}
+
 function normalizeWechatInboundIntent(inbound) {
   const msgType = String(inbound?.msgType ?? '').toLowerCase();
   const base = {
@@ -1210,6 +1233,22 @@ function normalizeWechatInboundIntent(inbound) {
     };
   }
 
+  if (msgType === 'file') {
+    const filename = String(inbound?.fileName ?? '').trim();
+    return {
+      ...base,
+      displayText: filename ? `收到文件：${filename}` : '收到文件',
+      agentText: '',
+      media: {
+        mediaId: inbound?.mediaId || '',
+      },
+      attachment: {
+        filename,
+        sizeBytes: normalizeNumber(inbound?.fileSize),
+      },
+    };
+  }
+
   if (msgType === 'location') {
     const latitude = normalizeNumber(inbound?.locationX);
     const longitude = normalizeNumber(inbound?.locationY);
@@ -1434,10 +1473,15 @@ export function createWechatMpService({
     jsapiTicketUrl: config.jsapiTicketUrl || DEFAULT_WECHAT_JSAPI_TICKET_URL,
     mediaPublicBaseUrl: config.mediaPublicBaseUrl || config.publicBaseUrl,
     maxImageBytes: Math.max(1, Number(config.maxImageBytes ?? 10 * 1024 * 1024)),
+    maxFileBytes: Math.max(1, Number(config.maxFileBytes ?? 30 * 1024 * 1024)),
     acceptVoice: config.acceptVoice !== false,
     acceptImage: config.acceptImage !== false,
+    acceptFile: config.acceptFile !== false,
     acceptLocation: config.acceptLocation !== false,
     acceptLink: config.acceptLink !== false,
+    mediaAnalysisGrayUsers: Array.isArray(config.mediaAnalysisGrayUsers)
+      ? config.mediaAnalysisGrayUsers
+      : [],
     asrTarget: config.asrTarget || DEFAULT_ASR_TARGET,
   };
 
@@ -1872,16 +1916,32 @@ export function createWechatMpService({
     }
   };
 
-  const buildIntentMetadata = (intent) => ({
-    source: 'wechat_mp',
-    msgType: intent.msgType,
-    originalMsgId: intent.msgId || null,
-    displayText: intent.displayText || '',
-    mediaPublicUrl: intent.media?.publicUrl || null,
-    recognition: intent.msgType === 'voice' ? intent.agentText || null : null,
-    location: intent.location || null,
-    link: intent.link || null,
-  });
+  const buildIntentMetadata = (intent, { mediaAnalysisEnabled = false } = {}) => {
+    const mediaPublicUrl = intent.media?.publicUrl || null;
+    const fileAttachment =
+      intent.msgType === 'file' && mediaPublicUrl && intent.attachment?.filename
+        ? {
+            assetId: '',
+            downloadUrl: mediaPublicUrl,
+            filename: intent.attachment.filename,
+            mimeType: intent.attachment.mimeType || 'application/octet-stream',
+          }
+        : null;
+    return {
+      source: 'wechat_mp',
+      msgType: intent.msgType,
+      originalMsgId: intent.msgId || null,
+      displayText: intent.displayText || '',
+      mediaPublicUrl,
+      ...(mediaAnalysisEnabled && intent.msgType === 'image' && mediaPublicUrl
+        ? { imageUrls: [mediaPublicUrl] }
+        : {}),
+      ...(mediaAnalysisEnabled && fileAttachment ? { fileAttachments: [fileAttachment] } : {}),
+      recognition: intent.msgType === 'voice' ? intent.agentText || null : null,
+      location: intent.location || null,
+      link: intent.link || null,
+    };
+  };
 
   const persistIntentDetail = async ({ intent, userId = null, rawXmlHash = '' }) => {
     if (typeof userAuth.insertWechatMpMessageDetail !== 'function') return;
@@ -1921,6 +1981,7 @@ export function createWechatMpService({
 
   const runIntentMessage = async ({ inbound, intent, user }) => {
     const wechatIntent = classifyWechatIntent(intent);
+    const mediaAnalysisEnabled = isWechatMediaGrayUser(user, config.mediaAnalysisGrayUsers);
     const resetCandidate =
       intent.msgType === 'text' || intent.msgType === 'voice' ? intent.agentText : '';
     // Page Data delivery owns persistent files, datasets and two publication
@@ -1970,7 +2031,7 @@ export function createWechatMpService({
         sessionId,
         requestId,
         agentPrompt,
-        buildIntentMetadata(intent),
+        buildIntentMetadata(intent, { mediaAnalysisEnabled }),
       );
       const workingDir = publishLayout?.publishDir ?? (await userAuth.resolveWorkingDir(user.userId));
       const linkExistsForRequest = resolveLinkExistsForWorkingDir(workingDir, linkExists);
@@ -2129,7 +2190,7 @@ export function createWechatMpService({
           sessionId,
           retryId,
           retryPrompt,
-          buildIntentMetadata(intent),
+          buildIntentMetadata(intent, { mediaAnalysisEnabled }),
         );
         const workingDir = publishLayout?.publishDir ?? (await userAuth.resolveWorkingDir(user.userId));
         const linkExistsForRequest = resolveLinkExistsForWorkingDir(workingDir, linkExists);
@@ -2302,6 +2363,7 @@ export function createWechatMpService({
     const supportedByConfig =
       (intent.msgType === 'voice' && config.acceptVoice) ||
       (intent.msgType === 'image' && config.acceptImage) ||
+      (intent.msgType === 'file' && config.acceptFile) ||
       (intent.msgType === 'location' && config.acceptLocation) ||
       (intent.msgType === 'link' && config.acceptLink) ||
       intent.msgType === 'text' ||
@@ -2350,6 +2412,25 @@ export function createWechatMpService({
       };
     }
 
+    const mediaAnalysisEnabled = isWechatMediaGrayUser(
+      boundUser,
+      config.mediaAnalysisGrayUsers,
+    );
+
+    if (intent.msgType === 'file' && !mediaAnalysisEnabled) {
+      await persistIntentDetail({ intent, userId: boundUser.userId, rawXmlHash });
+      return {
+        ok: true,
+        status: 200,
+        contentType: 'application/xml; charset=utf-8',
+        body: buildWechatTextReply({
+          toUserName: inbound.fromUserName,
+          fromUserName: inbound.toUserName,
+          content: '当前账号尚未开启服务号文件分析灰度，请先通过 H5 上传文件。',
+        }),
+      };
+    }
+
     if (intent.msgType === 'voice' && !intent.agentText.trim() && intent.media?.mediaId) {
       try {
         const fallbackText = await transcribeWechatVoiceMedia(intent.media.mediaId, intent.media.format);
@@ -2433,6 +2514,56 @@ export function createWechatMpService({
       }
     }
 
+    if (intent.msgType === 'file') {
+      try {
+        const accessToken = await getStableAccessToken();
+        const persisted = await persistWechatAttachment(
+          {
+            userId: boundUser.userId,
+            appId: config.appId,
+            openid: inbound.fromUserName,
+            msgId: inbound.msgId,
+            mediaId: inbound.mediaId,
+            filename: inbound.fileName,
+            publicBaseUrl: config.mediaPublicBaseUrl,
+            maxFileBytes: config.maxFileBytes,
+          },
+          {
+            wechatFetch,
+            accessToken,
+          },
+        );
+        intent.media = {
+          ...intent.media,
+          mediaId: inbound.mediaId || intent.media?.mediaId || '',
+          publicUrl: persisted.publicUrl,
+          format: persisted.contentType,
+          source: persisted.source,
+        };
+        intent.attachment = {
+          ...intent.attachment,
+          filename: persisted.filename,
+          publicUrl: persisted.publicUrl,
+          mimeType: persisted.contentType,
+          sizeBytes: persisted.bytes,
+        };
+        intent.agentText = `[文件1: ${persisted.filename}]: ${persisted.publicUrl}`;
+        intent.displayText = `文件：${persisted.filename}`;
+      } catch (error) {
+        await persistIntentDetail({ intent, userId: boundUser.userId, rawXmlHash });
+        return {
+          ok: true,
+          status: 200,
+          contentType: 'application/xml; charset=utf-8',
+          body: buildWechatTextReply({
+            toUserName: inbound.fromUserName,
+            fromUserName: inbound.toUserName,
+            content: error instanceof Error ? error.message : '文件处理失败，请稍后重试。',
+          }),
+        };
+      }
+    }
+
     if (
       intent.msgType === 'location' &&
       (intent.location?.latitude == null || intent.location?.longitude == null)

wechat-mp.test.mjs

@@ -3686,6 +3686,7 @@ test('wechat mp service persists image and routes image url into agent prompt',
   const nonce = 'nonce';
   const testUserId = 'test-user-image';
   const prompts = [];
+  const metadataCalls = [];
   const detailCalls = [];
   const service = createWechatMpService({
     config: {
@@ -3701,6 +3702,7 @@ test('wechat mp service persists image and routes image url into agent prompt',
       unboundTextPrefix: '请先绑定',
       progressDelayMs: 0,
       maxImageBytes: 1024 * 1024,
+      mediaAnalysisGrayUsers: [testUserId],
     },
     userAuth: {
       async findWechatUserByOpenid() {
@@ -3747,6 +3749,7 @@ test('wechat mp service persists image and routes image url into agent prompt',
       if (pathname === '/sessions/session-1/reply') {
         const body = JSON.parse(init.body);
         prompts.push(body.user_message.content[0].text);
+        metadataCalls.push(body.user_message.metadata);
         return new Response('{}', { status: 200, headers: { 'Content-Type': 'application/json' } });
       }
       if (pathname === '/agent/harness_remember' || pathname === '/agent/harness_bootstrap') {
@@ -3809,10 +3812,208 @@ test('wechat mp service persists image and routes image url into agent prompt',
     prompts[0],
     /\[图片1\]: https:\/\/example\.com\/MindSpace\/test-user-image\/public\/wechat-mp\//,
   );
+  assert.equal(metadataCalls.length, 1);
+  assert.equal(metadataCalls[0].source, 'wechat_mp');
+  assert.equal(metadataCalls[0].msgType, 'image');
+  assert.equal(metadataCalls[0].imageUrls.length, 1);
+  assert.match(metadataCalls[0].imageUrls[0], /\/public\/wechat-mp\//);
   assert.equal(detailCalls.length, 1);
   assert.match(detailCalls[0].mediaPublicUrl, /\/wechat-mp\//);
 });
 
+test('wechat mp service persists Word and Excel files in user public area and reuses H5 attachment metadata', async () => {
+  const token = 'token';
+  const timestamp = '1710000000';
+  const nonce = 'nonce';
+  const testUserId = 'test-user-wechat-file';
+  const userMessages = [];
+  const service = createBoundWechatService({
+    token,
+    config: {
+      maxFileBytes: 1024 * 1024,
+      acceptFile: true,
+      mediaAnalysisGrayUsers: [testUserId],
+    },
+    userAuth: {
+      async findWechatUserByOpenid() {
+        return { userId: testUserId, status: 'active', nickname: '毕升' };
+      },
+    },
+    sessionApiFetch: async (_sessionId, pathname, init = {}) => {
+      if (pathname === '/sessions/session-1/events') {
+        return new Response(
+          [
+            'data: {"type":"Message","request_id":"req-file","message":{"id":"assistant-1","role":"assistant","metadata":{"userVisible":true},"content":[{"type":"text","text":"文件已解析。"}]}}\n\n',
+            'data: {"type":"Finish","request_id":"req-file","token_state":{"inputTokens":1,"outputTokens":2}}\n\n',
+          ].join(''),
+          { status: 200, headers: { 'Content-Type': 'text/event-stream' } },
+        );
+      }
+      if (pathname === '/sessions/session-1/reply') {
+        const body = JSON.parse(init.body);
+        userMessages.push(body.user_message);
+        return new Response('{}', { status: 200, headers: { 'Content-Type': 'application/json' } });
+      }
+      if (pathname === '/agent/harness_remember' || pathname === '/agent/harness_bootstrap') {
+        return new Response('{}', { status: 200, headers: { 'Content-Type': 'application/json' } });
+      }
+      throw new Error(`unexpected api path: ${pathname}`);
+    },
+    wechatFetch: async (url) => {
+      if (String(url).includes('/cgi-bin/stable_token')) {
+        return new Response(JSON.stringify({ access_token: 'access-1', expires_in: 7200 }), {
+          status: 200,
+          headers: { 'Content-Type': 'application/json' },
+        });
+      }
+      if (String(url).includes('/cgi-bin/media/get')) {
+        return new Response(Buffer.from('fake-docx-content'), {
+          status: 200,
+          headers: { 'Content-Type': 'application/octet-stream' },
+        });
+      }
+      if (String(url).includes('/cgi-bin/message/custom/send')) {
+        return new Response(JSON.stringify({ errcode: 0, errmsg: 'ok' }), {
+          status: 200,
+          headers: { 'Content-Type': 'application/json' },
+        });
+      }
+      throw new Error(`unexpected wechat url: ${url}`);
+    },
+  });
+
+  const originalRandomUuid = crypto.randomUUID;
+  crypto.randomUUID = () => 'req-file';
+  try {
+    for (const [index, filename] of ['季度分析.docx', '销售数据.xlsx'].entries()) {
+      const result = await service.handleInboundMessage(
+        inboundXml({
+          msgType: 'file',
+          content: '',
+          extraFields: {
+            MediaId: `file-media-${index + 1}`,
+            FileName: filename,
+            FileSize: 17,
+          },
+        }),
+        { timestamp, nonce, signature: signatureFor(token, timestamp, nonce) },
+      );
+      assert.equal(result.status, 200);
+      await result.task;
+      const attachment = userMessages.at(-1)?.metadata?.fileAttachments?.[0];
+      const publicFilename = decodeURIComponent(new URL(attachment.downloadUrl).pathname.split('/').at(-1));
+      assert.equal(
+        fs.existsSync(
+          path.join(process.cwd(), 'MindSpace', testUserId, 'public', 'wechat-mp', publicFilename),
+        ),
+        true,
+      );
+    }
+  } finally {
+    crypto.randomUUID = originalRandomUuid;
+    fs.rmSync(path.join(process.cwd(), 'MindSpace', testUserId), {
+      recursive: true,
+      force: true,
+    });
+  }
+
+  assert.equal(userMessages.length, 2);
+  assert.match(userMessages[0].content[0].text, /【微信服务号文件消息】/);
+  assert.match(userMessages[0].content[0].text, /\[文件1: 季度分析\.docx\]:/);
+  assert.equal(userMessages[0].metadata.msgType, 'file');
+  assert.equal(userMessages[0].metadata.fileAttachments.length, 1);
+  assert.equal(userMessages[0].metadata.fileAttachments[0].filename, '季度分析.docx');
+  assert.equal(
+    userMessages[0].metadata.fileAttachments[0].mimeType,
+    'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
+  );
+  assert.match(userMessages[0].metadata.fileAttachments[0].downloadUrl, /\/public\/wechat-mp\//);
+  assert.match(userMessages[1].content[0].text, /\[文件1: 销售数据\.xlsx\]:/);
+  assert.equal(userMessages[1].metadata.fileAttachments[0].filename, '销售数据.xlsx');
+  assert.equal(
+    userMessages[1].metadata.fileAttachments[0].mimeType,
+    'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
+  );
+  assert.match(userMessages[1].metadata.fileAttachments[0].downloadUrl, /\/public\/wechat-mp\//);
+});
+
+test('wechat mp service rejects unsupported public file types without entering the agent flow', async () => {
+  let mediaDownloads = 0;
+  let sessionCalls = 0;
+  const service = createBoundWechatService({
+    config: { acceptFile: true, mediaAnalysisGrayUsers: ['user-1'] },
+    sessionApiFetch: async () => {
+      sessionCalls += 1;
+      throw new Error('session should not be called');
+    },
+    wechatFetch: async (url) => {
+      if (String(url).includes('/cgi-bin/stable_token')) {
+        return new Response(JSON.stringify({ access_token: 'access-1', expires_in: 7200 }), {
+          status: 200,
+          headers: { 'Content-Type': 'application/json' },
+        });
+      }
+      if (String(url).includes('/cgi-bin/media/get')) mediaDownloads += 1;
+      throw new Error(`unexpected wechat url: ${url}`);
+    },
+  });
+
+  const result = await service.handleInboundMessage(
+    inboundXml({
+      msgType: 'file',
+      content: '',
+      extraFields: { MediaId: 'file-media-2', FileName: 'payload.exe' },
+    }),
+    {
+      timestamp: '1710000000',
+      nonce: 'nonce',
+      signature: signatureFor('token', '1710000000', 'nonce'),
+    },
+  );
+
+  assert.equal(result.status, 200);
+  assert.match(result.body, /仅支持 Word/);
+  assert.equal(mediaDownloads, 0);
+  assert.equal(sessionCalls, 0);
+});
+
+test('wechat mp service keeps file analysis disabled outside the media gray allowlist', async () => {
+  let mediaDownloads = 0;
+  let sessionCalls = 0;
+  const service = createBoundWechatService({
+    config: {
+      acceptFile: true,
+      mediaAnalysisGrayUsers: ['唐'],
+    },
+    sessionApiFetch: async () => {
+      sessionCalls += 1;
+      throw new Error('session should not be called');
+    },
+    wechatFetch: async (url) => {
+      if (String(url).includes('/cgi-bin/media/get')) mediaDownloads += 1;
+      throw new Error(`unexpected wechat url: ${url}`);
+    },
+  });
+
+  const result = await service.handleInboundMessage(
+    inboundXml({
+      msgType: 'file',
+      content: '',
+      extraFields: { MediaId: 'file-media-gray', FileName: '测试.docx' },
+    }),
+    {
+      timestamp: '1710000000',
+      nonce: 'nonce',
+      signature: signatureFor('token', '1710000000', 'nonce'),
+    },
+  );
+
+  assert.equal(result.status, 200);
+  assert.match(result.body, /尚未开启服务号文件分析灰度/);
+  assert.equal(mediaDownloads, 0);
+  assert.equal(sessionCalls, 0);
+});
+
 test('wechat mp service accepts full voice xml payload and routes recognition text into agent', async () => {
   const token = 'token';
   const timestamp = '1710000000';

wechat/ack/ack-builders.mjs

@@ -21,6 +21,12 @@ const ImageBuilder = {
   build: (ctx) => buildText(TEMPLATES.image, ctx),
 };
 
+const FileBuilder = {
+  priority: 80,
+  support: (ctx) => ctx.msgType === 'file',
+  build: (ctx) => buildText(TEMPLATES.file, ctx),
+};
+
 const VoiceBuilder = {
   priority: 80,
   support: (ctx) => ctx.msgType === 'voice',
@@ -54,7 +60,7 @@ const DefaultBuilder = {
   build: (ctx) => buildText(TEMPLATES.default, ctx),
 };
 
-const BUILDERS = [ImageBuilder, VoiceBuilder, LocationBuilder, LinkBuilder, IntentBuilder, DefaultBuilder]
+const BUILDERS = [ImageBuilder, FileBuilder, VoiceBuilder, LocationBuilder, LinkBuilder, IntentBuilder, DefaultBuilder]
   .sort((a, b) => b.priority - a.priority);
 
 export function selectBuilder(ctx) {

wechat/ack/ack-intent.mjs

@@ -13,6 +13,7 @@ const INTENT_RULES = [
 
 export function resolveIntent(msgType, text) {
   if (msgType === 'image') return { task: 'image_analysis' };
+  if (msgType === 'file') return { task: 'file_analysis' };
   if (msgType === 'voice') return { task: 'voice_analysis' };
   if (msgType === 'location') return { task: 'location' };
   if (msgType === 'link') return { task: 'link' };

wechat/ack/ack-provider.test.mjs

@@ -26,6 +26,11 @@ describe('buildAckText', () => {
     assert.equal(result, '图片收到，我先看看。');
   });
 
+  it('file → file template', () => {
+    const result = buildAckText({ intent: intent('file'), nickname: '', config: cfg, fallbackText: '' });
+    assert.equal(result, '文件收到，我先读取分析。');
+  });
+
   it('voice → voice template', () => {
     const result = buildAckText({ intent: intent('voice'), nickname: '', config: cfg, fallbackText: '' });
     assert.equal(result, '收到语音，我先听一下。');

wechat/ack/ack-templates.mjs

@@ -12,6 +12,7 @@ export const TEMPLATES = {
     '让我看看这张图片。',
     '图片已收到，我来处理。',
   ],
+  file: ['文件收到，我先读取分析。'],
   voice: [
     '收到语音，我先听一下。',
     '听到啦，我马上处理。',

wechat/prompts/chat-general.mjs

@@ -81,6 +81,20 @@ export function buildWechatAgentPrompt(intent, { grantedSkills = [] } = {}) {
       .filter(Boolean)
       .join('\n');
   }
+  if (msgType === 'file') {
+    const attachment = intent?.attachment ?? {};
+    return [
+      currentTimeHint,
+      '【微信服务号文件消息】用户发送了需要解析的 Office 文件。',
+      attachment.filename ? `文件名：${attachment.filename}` : '',
+      attachment.publicUrl ? `文件链接：${attachment.publicUrl}` : '',
+      '请复用 H5 附件分析结果回答；Excel 在专用工具可用时必须优先使用 Excel 工具读取完整工作簿。',
+      '',
+      String(agentText).trim(),
+    ]
+      .filter(Boolean)
+      .join('\n');
+  }
   if (msgType === 'location') {
     const location = intent?.location ?? {};
     return [