diff --git a/pages/wiki/chat/index.js b/pages/wiki/chat/index.js
index dab7c83..db3448b 100644
--- a/pages/wiki/chat/index.js
+++ b/pages/wiki/chat/index.js
@@ -1,4 +1,5 @@
 // pages/wiki/chat/index.js
+import request from '../../../utils/request';
 
 Page({
   data: {
@@ -7,6 +8,8 @@ Page({
     isTyping: false,
     scrollAnchor: '',
     _counter: 0,
+    quotaRemaining: -1,
+    quotaLimit: 0,
   },
 
   onLoad(options) {
@@ -25,6 +28,19 @@ Page({
     }
   },
 
+  onShow() {
+    this._fetchQuota();
+  },
+
+  _fetchQuota() {
+    request.get('/plant/chat/quota').then(res => {
+      this.setData({
+        quotaRemaining: res.remaining,
+        quotaLimit: res.limit,
+      });
+    }).catch(() => {});
+  },
+
   goToHistory() {
     wx.navigateTo({ url: '/pages/wiki/chat/history/index' });
   },
@@ -42,11 +58,16 @@ Page({
     const query = this.data.inputValue.trim();
     if (!query || this.data.isTyping) return;
 
+    // 额度检查
+    if (this.data.quotaRemaining === 0) {
+      wx.showToast({ title: '今日问答次数已用完，明天再来吧', icon: 'none', duration: 2500 });
+      return;
+    }
+
     const uid = 'u' + (++this.data._counter);
     const aid = 'a' + (++this.data._counter);
     const len = this.data.messages.length;
 
-    // Push user msg + empty AI msg at once
     this.setData({
       [`messages[${len}]`]: { id: uid, role: 'user', content: query },
       [`messages[${len + 1}]`]: { id: aid, role: 'ai', content: '' },
@@ -59,53 +80,58 @@ Page({
   },
 
   _streamRequest(query, aiMsgId) {
-    const token = wx.getStorageSync('token');
-    const baseUrl = 'http://192.168.0.184:8889';
-    const url = `${baseUrl}/plant/chat/stream?query=${encodeURIComponent(query)}`;
     let fullText = '';
 
-    const task = wx.request({
-      url,
-      method: 'GET',
-      enableChunked: true,
-      header: {
-        'Authorization': `Bearer ${token}`,
-        'Accept': 'text/event-stream',
-      },
-      success: () => {
-        this.setData({ isTyping: false });
+    request.stream('/plant/chat/stream', { query }, {
+      onChunk: (res) => {
+        const text = this._decode(res.data);
+
+        // Detect non-SSE JSON error (e.g. quota exceeded returns {code:7, msg:"..."})
+        if (!text.startsWith('data: ')) {
+          try {
+            const json = JSON.parse(text);
+            if (json.code && json.code !== 200 && json.msg) {
+              this._updateAiMsg(aiMsgId, '⚠️ ' + json.msg);
+              this.setData({ isTyping: false });
+              this._fetchQuota();
+              return;
+            }
+          } catch (_) { /* not JSON, continue SSE parsing */ }
+        }
+
+        const lines = text.split('\n');
+
+        for (const line of lines) {
+          if (!line.startsWith('data: ')) continue;
+          const chunk = line.substring(6);
+
+          if (chunk === '[DONE]') {
+            this.setData({ isTyping: false });
+            return;
+          }
+          if (chunk.startsWith('[ERROR]')) {
+            fullText += '\n⚠️ ' + (chunk.substring(7) || '服务异常');
+            this._updateAiMsg(aiMsgId, fullText);
+            this.setData({ isTyping: false });
+            return;
+          }
+
+          fullText += chunk;
+          this._updateAiMsg(aiMsgId, fullText);
+        }
         this.scrollToBottom();
       },
-      fail: () => {
+      onDone: () => {
+        this.setData({ isTyping: false });
+        this.scrollToBottom();
+        // Delay: backend saves history async (go func) after stream ends
+        setTimeout(() => this._fetchQuota(), 800);
+      },
+      onError: () => {
         this._updateAiMsg(aiMsgId, '网络连接失败，请稍后重试');
         this.setData({ isTyping: false });
       },
     });
-
-    task.onChunkReceived((res) => {
-      const text = this._decode(res.data);
-      const lines = text.split('\n');
-
-      for (const line of lines) {
-        if (!line.startsWith('data: ')) continue;
-        const chunk = line.substring(6);
-
-        if (chunk === '[DONE]') {
-          this.setData({ isTyping: false });
-          return;
-        }
-        if (chunk.startsWith('[ERROR]')) {
-          fullText += '\n⚠️ ' + (chunk.substring(7) || '服务异常');
-          this._updateAiMsg(aiMsgId, fullText);
-          this.setData({ isTyping: false });
-          return;
-        }
-
-        fullText += chunk;
-        this._updateAiMsg(aiMsgId, fullText);
-      }
-      this.scrollToBottom();
-    });
   },
 
   _updateAiMsg(id, content) {
diff --git a/pages/wiki/chat/index.wxml b/pages/wiki/chat/index.wxml
index 052cd17..9815b07 100644
--- a/pages/wiki/chat/index.wxml
+++ b/pages/wiki/chat/index.wxml
@@ -85,6 +85,9 @@
         <t-icon name="arrow-up" size="36rpx" color="#fff" />
       </view>
     </view>
+    <view class="quota-bar" wx:if="{{quotaLimit > 0}}">
+      <text class="quota-text {{quotaRemaining <= 2 ? 'warn' : ''}}">今日剩余 {{quotaRemaining}}/{{quotaLimit}} 次</text>
+    </view>
     <view class="safe-bottom"></view>
   </view>
 </view>
diff --git a/pages/wiki/chat/index.wxss b/pages/wiki/chat/index.wxss
index 6f744b4..13f5bb5 100644
--- a/pages/wiki/chat/index.wxss
+++ b/pages/wiki/chat/index.wxss
@@ -263,3 +263,20 @@
 .safe-bottom {
   height: calc(14rpx + env(safe-area-inset-bottom));
 }
+
+/* Quota Bar */
+.quota-bar {
+  text-align: center;
+  padding: 6rpx 0 2rpx;
+}
+
+.quota-text {
+  font-size: 22rpx;
+  color: #90A4AE;
+  font-weight: 500;
+}
+
+.quota-text.warn {
+  color: #EF4444;
+  font-weight: 600;
+}
diff --git a/utils/request.js b/utils/request.js
index ec4f0e2..c085f83 100644
--- a/utils/request.js
+++ b/utils/request.js
@@ -266,12 +266,62 @@ class WxRequest {
     post(url, data = {}, header = {}) {
         return this.request({ url, method: 'POST', data, header });
     }
+
+    /**
+     * SSE stream request with chunked transfer.
+     * Reuses baseUrl + token interceptor.
+     * @param {string} url  API path, e.g. '/plant/chat/stream'
+     * @param {Object} data  Query params (GET) or body (POST)
+     * @param {Object} callbacks  { onChunk, onDone, onError }
+     * @returns {Object} wx.request task (call task.abort() to cancel)
+     */
+    stream(url, data = {}, callbacks = {}) {
+        const fullUrl = url.startsWith('http')
+            ? url
+            : this.baseUrl + url;
+
+        // Build query string for GET
+        const qs = Object.keys(data)
+            .map(k => `${encodeURIComponent(k)}=${encodeURIComponent(data[k])}`)
+            .join('&');
+        const reqUrl = qs ? `${fullUrl}?${qs}` : fullUrl;
+
+        // Resolve token via same logic as interceptor
+        let token = wx.getStorageSync('token');
+
+        const mergedHeader = {
+            ...this.header,
+            'Accept': 'text/event-stream',
+        };
+        if (token) {
+            mergedHeader['Authorization'] = `Bearer ${token}`;
+        }
+
+        const task = wx.request({
+            url: reqUrl,
+            method: 'GET',
+            enableChunked: true,
+            header: mergedHeader,
+            success: () => {
+                if (callbacks.onDone) callbacks.onDone();
+            },
+            fail: (err) => {
+                if (callbacks.onError) callbacks.onError(err);
+            },
+        });
+
+        if (callbacks.onChunk) {
+            task.onChunkReceived(callbacks.onChunk);
+        }
+
+        return task;
+    }
 }
 
 // Initialize with default instance
 const request = new WxRequest({
-    baseUrl: 'http://192.168.0.184:8889',
-    //baseUrl: 'https://go.sundynix.cn/api',
+    //baseUrl: 'http://192.168.0.184:8889',
+    baseUrl: 'https://go.sundynix.cn/api',
     header: {
         'Content-Type': 'application/json'
     }