Merge branch 'main' into login_page

2026-09-19 02:18:25 +08:00 · 2025-09-16 10:24:05 +08:00
parent 81bd2e6c2c c2d52a9374
commit c3744866fd
18 changed files with 526 additions and 46 deletions
@@ -3,6 +3,13 @@ server {
    server_name localhost;
    client_max_body_size 50M;
    
+    # 安全头配置
+    add_header X-Frame-Options "SAMEORIGIN" always;
+    add_header X-Content-Type-Options "nosniff" always;
+    add_header X-XSS-Protection "1; mode=block" always;
+    add_header Referrer-Policy "strict-origin-when-cross-origin" always;
+    add_header Content-Security-Policy "default-src 'self'; script-src 'self' 'unsafe-inline' 'unsafe-eval'; style-src 'self' 'unsafe-inline'; img-src 'self' data: https: http:; font-src 'self' data:; connect-src 'self' http: https: ws: wss:; frame-ancestors 'self';" always;
+    
    # 错误日志配置
    error_log /var/log/nginx/error.log warn;
    access_log /var/log/nginx/access.log;
@@ -9,7 +9,9 @@
      "version": "0.1.0",
      "dependencies": {
        "@microsoft/fetch-event-source": "^2.0.1",
+        "@types/dompurify": "^3.0.5",
        "axios": "^1.8.4",
+        "dompurify": "^3.2.6",
        "marked": "^5.1.2",
        "pagefind": "^1.1.1",
        "pinia": "^3.0.1",
@@ -1274,6 +1276,15 @@
      "dev": true,
      "license": "MIT"
    },
+    "node_modules/@types/dompurify": {
+      "version": "3.0.5",
+      "resolved": "https://mirrors.tencent.com/npm/@types/dompurify/-/dompurify-3.0.5.tgz",
+      "integrity": "sha512-1Wg0g3BtQF7sSb27fJQAKck1HECM6zV1EB66j8JH9i3LCjYabJa0FSdiSgsD5K/RbrsR0SiraKacLB+T8ZVYAg==",
+      "license": "MIT",
+      "dependencies": {
+        "@types/trusted-types": "*"
+      }
+    },
    "node_modules/@types/eslint": {
      "version": "9.6.1",
      "resolved": "https://mirrors.tencent.com/npm/@types/eslint/-/eslint-9.6.1.tgz",
@@ -1346,6 +1357,12 @@
      "resolved": "https://mirrors.tencent.com/npm/@types/tinycolor2/-/tinycolor2-1.4.6.tgz",
      "integrity": "sha512-iEN8J0BoMnsWBqjVbWH/c0G0Hh7O21lpR2/+PrvAVgWdzL7eexIFm4JN/Wn10PTcmNdtS6U67r499mlWMXOxNw=="
    },
+    "node_modules/@types/trusted-types": {
+      "version": "2.0.7",
+      "resolved": "https://mirrors.tencent.com/npm/@types/trusted-types/-/trusted-types-2.0.7.tgz",
+      "integrity": "sha512-ScaPdn1dQczgbl0QFTeTOmVHFULt394XJgOQNoyVhZ6r2vLnMLJfBPd53SB52T/3G36VI1/g2MZaX0cwDuXsfw==",
+      "license": "MIT"
+    },
    "node_modules/@types/validator": {
      "version": "13.15.2",
      "resolved": "https://mirrors.tencent.com/npm/@types/validator/-/validator-13.15.2.tgz",
@@ -2121,6 +2138,15 @@
        "node": ">=0.4.0"
      }
    },
+    "node_modules/dompurify": {
+      "version": "3.2.6",
+      "resolved": "https://mirrors.tencent.com/npm/dompurify/-/dompurify-3.2.6.tgz",
+      "integrity": "sha512-/2GogDQlohXPZe6D6NOgQvXLPSYBqIWMnZ8zzOhn09REE4eyAzb+Hed3jhoM9OkuaJ8P6ZGTTVWQKAi8ieIzfQ==",
+      "license": "(MPL-2.0 OR Apache-2.0)",
+      "optionalDependencies": {
+        "@types/trusted-types": "^2.0.7"
+      }
+    },
    "node_modules/dunder-proto": {
      "version": "1.0.1",
      "resolved": "https://mirrors.tencent.com/npm/dunder-proto/-/dunder-proto-1.0.1.tgz",
@@ -13,7 +13,9 @@
  },
  "dependencies": {
    "@microsoft/fetch-event-source": "^2.0.1",
+    "@types/dompurify": "^3.0.5",
    "axios": "^1.8.4",
+    "dompurify": "^3.2.6",
    "marked": "^5.1.2",
    "pagefind": "^1.1.1",
    "pinia": "^3.0.1",
@@ -4,6 +4,8 @@ import { onMounted, ref, nextTick, onUnmounted, onUpdated, watch } from "vue";
 import { downKnowledgeDetails } from "@/api/knowledge-base/index";
 import { MessagePlugin } from "tdesign-vue-next";
 import picturePreview from '@/components/picture-preview.vue';
+import { sanitizeHTML, safeMarkdownToHTML, createSafeImage, isValidImageURL } from '@/utils/security';
+
 marked.use({
  mangle: false,
  headerIds: false,
@@ -37,10 +39,16 @@ const checkImage = (url) => {
  });
 };
 renderer.image = function (href, title, text) {
-  // 自定义HTML结构，图片展示带标题
+  // 安全地处理图片链接
+  if (!isValidImageURL(href)) {
+    return `<p>无效的图片链接</p>`;
+  }
+  
+  // 使用安全的图片创建函数
+  const safeImage = createSafeImage(href, text || '', title || '');
  return `<figure>
-                <img class="markdown-image" src="${href}" alt="${title}" title="${text}">
-                <figcaption style="text-align: left;">${text}</figcaption>
+                ${safeImage}
+                <figcaption style="text-align: left;">${text || ''}</figcaption>
            </figure>`;
 };
 const props = defineProps(["visible", "details"]);
@@ -66,14 +74,23 @@ watch(() => props.details.md, (newVal) => {
  deep: true
 })

-// 处理 Markdown 中的图片
+// 安全地处理 Markdown 内容
 const processMarkdown = (markdownText) => {
-  // 自定义渲染器处理图片
+  if (!markdownText || typeof markdownText !== 'string') {
+    return '';
+  }
+  
+  // 首先对 Markdown 内容进行安全处理
+  const safeMarkdown = safeMarkdownToHTML(markdownText);
+  
+  // 使用安全的渲染器
  marked.use({ renderer });
-  let html = marked.parse(markdownText);
-  const parser = new DOMParser();
-  const doc = parser.parseFromString(html, 'text/html');
-  return doc.body.innerHTML;
+  let html = marked.parse(safeMarkdown);
+  
+  // 使用 DOMPurify 进行最终的安全清理
+  const sanitizedHTML = sanitizeHTML(html);
+  
+  return sanitizedHTML;
 };
 const closePreImg = () => {
  reviewImg.value = false
@@ -0,0 +1,207 @@
+/**
+ * 安全工具类 - 防止 XSS 攻击
+ */
+
+import DOMPurify from 'dompurify';
+
+// 配置 DOMPurify 的安全策略
+const DOMPurifyConfig = {
+  // 允许的标签
+  ALLOWED_TAGS: [
+    'p', 'br', 'strong', 'em', 'u', 's', 'del', 'ins',
+    'h1', 'h2', 'h3', 'h4', 'h5', 'h6',
+    'ul', 'ol', 'li', 'blockquote', 'pre', 'code',
+    'a', 'img', 'table', 'thead', 'tbody', 'tr', 'th', 'td',
+    'div', 'span', 'figure', 'figcaption'
+  ],
+  // 允许的属性
+  ALLOWED_ATTR: [
+    'href', 'title', 'alt', 'src', 'class', 'id', 'style',
+    'target', 'rel', 'width', 'height'
+  ],
+  // 允许的协议
+  ALLOWED_URI_REGEXP: /^(?:(?:(?:f|ht)tps?|mailto|tel|callto|cid|xmpp):|[^a-z]|[a-z+.\-]+(?:[^a-z+.\-:]|$))/i,
+  // 禁止的标签和属性
+  FORBID_TAGS: ['script', 'object', 'embed', 'form', 'input', 'button'],
+  FORBID_ATTR: ['onerror', 'onload', 'onclick', 'onmouseover', 'onfocus', 'onblur'],
+  // 其他安全配置
+  KEEP_CONTENT: true,
+  RETURN_DOM: false,
+  RETURN_DOM_FRAGMENT: false,
+  RETURN_DOM_IMPORT: false,
+  SANITIZE_DOM: true,
+  SANITIZE_NAMED_PROPS: true,
+  WHOLE_DOCUMENT: false,
+  // 自定义钩子函数
+  HOOKS: {
+    // 在清理前处理
+    beforeSanitizeElements: (currentNode: Element) => {
+      // 移除所有 script 标签
+      if (currentNode.tagName === 'SCRIPT') {
+        currentNode.remove();
+        return null;
+      }
+      // 移除所有事件处理器
+      const eventAttrs = ['onclick', 'onload', 'onerror', 'onmouseover', 'onfocus', 'onblur'];
+      eventAttrs.forEach(attr => {
+        if (currentNode.hasAttribute(attr)) {
+          currentNode.removeAttribute(attr);
+        }
+      });
+    },
+    // 在清理后处理
+    afterSanitizeElements: (currentNode: Element) => {
+      // 确保所有链接都有 rel="noopener noreferrer"
+      if (currentNode.tagName === 'A') {
+        const href = currentNode.getAttribute('href');
+        if (href && href.startsWith('http')) {
+          currentNode.setAttribute('rel', 'noopener noreferrer');
+          currentNode.setAttribute('target', '_blank');
+        }
+      }
+      // 确保所有图片都有 alt 属性
+      if (currentNode.tagName === 'IMG') {
+        if (!currentNode.getAttribute('alt')) {
+          currentNode.setAttribute('alt', '');
+        }
+      }
+    }
+  }
+};
+
+/**
+ * 安全地清理 HTML 内容
+ * @param html 需要清理的 HTML 字符串
+ * @returns 清理后的安全 HTML 字符串
+ */
+export function sanitizeHTML(html: string): string {
+  if (!html || typeof html !== 'string') {
+    return '';
+  }
+  
+  try {
+    return DOMPurify.sanitize(html, DOMPurifyConfig);
+  } catch (error) {
+    console.error('HTML sanitization failed:', error);
+    // 如果清理失败，返回转义的纯文本
+    return escapeHTML(html);
+  }
+}
+
+/**
+ * 转义 HTML 特殊字符
+ * @param text 需要转义的文本
+ * @returns 转义后的文本
+ */
+export function escapeHTML(text: string): string {
+  if (!text || typeof text !== 'string') {
+    return '';
+  }
+  
+  const map: { [key: string]: string } = {
+    '&': '&amp;',
+    '<': '&lt;',
+    '>': '&gt;',
+    '"': '&quot;',
+    "'": '&#x27;',
+    '/': '&#x2F;',
+    '`': '&#x60;',
+    '=': '&#x3D;'
+  };
+  
+  return text.replace(/[&<>"'`=\/]/g, (s) => map[s]);
+}
+
+/**
+ * 验证 URL 是否安全
+ * @param url 需要验证的 URL
+ * @returns 是否为安全 URL
+ */
+export function isValidURL(url: string): boolean {
+  if (!url || typeof url !== 'string') {
+    return false;
+  }
+  
+  try {
+    const urlObj = new URL(url);
+    // 只允许 http 和 https 协议
+    return ['http:', 'https:'].includes(urlObj.protocol);
+  } catch {
+    return false;
+  }
+}
+
+/**
+ * 安全地处理 Markdown 内容
+ * @param markdown Markdown 文本
+ * @returns 安全的 HTML 字符串
+ */
+export function safeMarkdownToHTML(markdown: string): string {
+  if (!markdown || typeof markdown !== 'string') {
+    return '';
+  }
+  
+  // 首先转义可能的 HTML 标签
+  const escapedMarkdown = markdown
+    .replace(/<script\b[^<]*(?:(?!<\/script>)<[^<]*)*<\/script>/gi, '')
+    .replace(/<iframe\b[^<]*(?:(?!<\/iframe>)<[^<]*)*<\/iframe>/gi, '')
+    .replace(/<object\b[^<]*(?:(?!<\/object>)<[^<]*)*<\/object>/gi, '')
+    .replace(/<embed\b[^<]*(?:(?!<\/embed>)<[^<]*)*<\/embed>/gi, '');
+  
+  return escapedMarkdown;
+}
+
+/**
+ * 清理用户输入
+ * @param input 用户输入
+ * @returns 清理后的安全输入
+ */
+export function sanitizeUserInput(input: string): string {
+  if (!input || typeof input !== 'string') {
+    return '';
+  }
+  
+  // 移除控制字符
+  let cleaned = input.replace(/[\x00-\x1F\x7F-\x9F]/g, '');
+  
+  // 限制长度
+  if (cleaned.length > 10000) {
+    cleaned = cleaned.substring(0, 10000);
+  }
+  
+  return cleaned.trim();
+}
+
+/**
+ * 验证图片 URL 是否安全
+ * @param url 图片 URL
+ * @returns 是否为安全的图片 URL
+ */
+export function isValidImageURL(url: string): boolean {
+  if (!isValidURL(url)) {
+    return false;
+  }
+  
+  // 检查是否为图片文件
+  const imageExtensions = /\.(jpg|jpeg|png|gif|webp|svg|bmp|ico)(\?.*)?$/i;
+  return imageExtensions.test(url);
+}
+
+/**
+ * 创建安全的图片元素
+ * @param src 图片源
+ * @param alt 替代文本
+ * @param title 标题
+ * @returns 安全的图片 HTML
+ */
+export function createSafeImage(src: string, alt: string = '', title: string = ''): string {
+  if (!isValidImageURL(src)) {
+    return '';
+  }
+  
+  const safeSrc = escapeHTML(src);
+  const safeAlt = escapeHTML(alt);
+  const safeTitle = escapeHTML(title);
+  
+  return `<img src="${safeSrc}" alt="${safeAlt}" title="${safeTitle}" class="markdown-image" style="max-width: 100%; height: auto;">`;
+}
@@ -23,6 +23,8 @@ import { marked } from 'marked';
 import docInfo from './docInfo.vue';
 import deepThink from './deepThink.vue';
 import picturePreview from '@/components/picture-preview.vue';
+import { sanitizeHTML, safeMarkdownToHTML, createSafeImage, isValidImageURL } from '@/utils/security';
+
 marked.use({
    mangle: false,
    headerIds: false,
@@ -89,36 +91,36 @@ const checkImage = (url) => {
        img.src = url;
    });
 };
-// 处理 Markdown 中的图片
+// 安全地处理 Markdown 内容
 const processMarkdown = (markdownText) => {
-    // 自定义渲染器处理图片
+    if (!markdownText || typeof markdownText !== 'string') {
+        return '';
+    }
+    
+    // 首先对 Markdown 内容进行安全处理
+    const safeMarkdown = safeMarkdownToHTML(markdownText);
+    
+    // 自定义安全的渲染器处理图片
    const renderer = {
        image(href, title, text) {
-            return `<img src="${href}" alt="${text}" title="${title || ''}"  class="markdown-image" style="max-width: 708px;height: 230px;">`;
+            // 验证图片 URL 是否安全
+            if (!isValidImageURL(href)) {
+                return `<p>无效的图片链接</p>`;
+            }
+            // 使用安全的图片创建函数
+            return createSafeImage(href, text || '', title || '');
        }
    };

    marked.use({ renderer });

-    // 第一次渲染
-    let html = marked.parse(markdownText);
+    // 安全地渲染 Markdown
+    let html = marked.parse(safeMarkdown);

-    // 创建虚拟 DOM 来操作
-    const parser = new DOMParser();
-    const doc = parser.parseFromString(html, 'text/html');
-
-    // 检查所有图片
-    // const images = doc.querySelectorAll('img');
-    // images.forEach(async item => {
-    //     const isValid = await checkImage(item.src);
-    //     if (!isValid) {
-    //         item.remove();
-    //     }
-    // });
-    // if (props.isFirstEnter) { 
-    //     emit('scroll-bottom')
-    // }
-    return doc.body.innerHTML;
+    // 使用 DOMPurify 进行最终的安全清理
+    const sanitizedHTML = sanitizeHTML(html);
+    
+    return sanitizedHTML;
 };
 const handleImg = async (newVal) => {
    let index = newVal.lastIndexOf('![');
@@ -29,6 +29,7 @@
 </template>
 <script setup>
 import { onMounted, watch, computed, ref, reactive, defineProps } from 'vue';
+import { sanitizeHTML } from '@/utils/security';

 const isFold = ref(true)
 const props = defineProps({
@@ -51,7 +52,6 @@ const showHide = () => {
 }
 const handlePanelChange = (val) => {
    isFold.value = !val.length ? true : false;
-
 }
 </script>
 <style lang="less" scoped>
@@ -15,7 +15,7 @@
                    trigger="click">
                    <template #content>
                        <div class="doc_content">
-                            <div v-html="item.content.replace(/\n/g, '<br/>')"></div>
+                            <div v-html="safeProcessContent(item.content)"></div>
                        </div>
                    </template>
                    <span class="doc">
@@ -28,6 +28,7 @@
 </template>
 <script setup>
 import { onMounted, defineProps, computed, ref, reactive } from "vue";
+import { sanitizeHTML } from '@/utils/security';
 const props = defineProps({
    // 必填项
    content: {
@@ -44,6 +45,14 @@ const referBoxSwitch = () => {
    showReferBox.value = !showReferBox.value;
 };

+// 安全地处理内容
+const safeProcessContent = (content) => {
+    if (!content) return '';
+    // 先进行安全清理，然后处理换行
+    const sanitized = sanitizeHTML(content);
+    return sanitized.replace(/\n/g, '<br/>');
+};
+
 </script>
 <style lang="less" scoped>
 .refer {
@@ -12,6 +12,7 @@ import (

 	"github.com/Tencent/WeKnora/internal/logger"
 	"github.com/Tencent/WeKnora/internal/types"
+	secutils "github.com/Tencent/WeKnora/internal/utils"
 )

 // PluginIntoChatMessage handles the transformation of search results into chat messages
@@ -50,9 +51,16 @@ func (p *PluginIntoChatMessage) OnEvent(ctx context.Context,
 	weekdayName := []string{"星期日", "星期一", "星期二", "星期三", "星期四", "星期五", "星期六"}
 	var userContent bytes.Buffer

+	// 验证用户查询的安全性
+	safeQuery, isValid := secutils.ValidateInput(chatManage.Query)
+	if !isValid {
+		logger.Errorf(ctx, "Invalid user query: %s", chatManage.Query)
+		return ErrTemplateExecute.WithError(fmt.Errorf("用户查询包含非法内容"))
+	}
+
 	// Execute template with context data
 	err = tmpl.Execute(&userContent, map[string]interface{}{
-		"Query":       chatManage.Query,                         // User's original query
+		"Query":       safeQuery,                                // User's original query
 		"Contexts":    passages,                                 // Extracted passages from search results
 		"CurrentTime": time.Now().Format("2006-01-02 15:04:05"), // Formatted current time
 		"CurrentWeek": weekdayName[time.Now().Weekday()],        // Current weekday in Chinese
@@ -23,8 +23,8 @@ type cosFileService struct {
 }

 // NewCosFileService creates a new COS file service instance
-func NewCosFileService(appId, region, secretId, secretKey, cosPathPrefix string) (interfaces.FileService, error) {
-	bucketURL := fmt.Sprintf("https://%s.cos.%s.myqcloud.com", appId, region)
+func NewCosFileService(bucketName, region, secretId, secretKey, cosPathPrefix string) (interfaces.FileService, error) {
+	bucketURL := fmt.Sprintf("https://%s.cos.%s.myqcloud.com/", bucketName, region)
 	u, err := url.Parse(bucketURL)
 	if err != nil {
 		return nil, fmt.Errorf("failed to parse bucketURL: %w", err)
@@ -59,7 +59,7 @@ func (s *cosFileService) SaveFile(ctx context.Context,
 	if err != nil {
 		return "", fmt.Errorf("failed to upload file to COS: %w", err)
 	}
-	return fmt.Sprintf("https://%s/%s", s.bucketURL, objectName), nil
+	return fmt.Sprintf("%s%s", s.bucketURL, objectName), nil
 }

 // GetFile retrieves a file from COS storage by its path URL
@@ -25,6 +25,7 @@ import (
 	"github.com/Tencent/WeKnora/internal/tracing"
 	"github.com/Tencent/WeKnora/internal/types"
 	"github.com/Tencent/WeKnora/internal/types/interfaces"
+	secutils "github.com/Tencent/WeKnora/internal/utils"
 	"github.com/Tencent/WeKnora/services/docreader/src/client"
 	"github.com/Tencent/WeKnora/services/docreader/src/proto"
 	"github.com/google/uuid"
@@ -191,15 +192,22 @@ func (s *knowledgeService) CreateKnowledgeFromFile(ctx context.Context,
 		metadataJSON = types.JSON(metadataBytes)
 	}

+	// 验证文件名安全性
+	safeFilename, isValid := secutils.ValidateInput(file.Filename)
+	if !isValid {
+		logger.Errorf(ctx, "Invalid filename: %s", file.Filename)
+		return nil, werrors.NewValidationError("文件名包含非法字符")
+	}
+
 	// Create knowledge record
 	logger.Info(ctx, "Creating knowledge record")
 	knowledge := &types.Knowledge{
 		TenantID:         tenantID,
 		KnowledgeBaseID:  kbID,
 		Type:             "file",
-		Title:            file.Filename,
-		FileName:         file.Filename,
-		FileType:         getFileType(file.Filename),
+		Title:            safeFilename,
+		FileName:         safeFilename,
+		FileType:         getFileType(safeFilename),
 		FileSize:         file.Size,
 		FileHash:         hash,
 		ParseStatus:      "pending",
@@ -258,10 +266,10 @@ func (s *knowledgeService) CreateKnowledgeFromURL(ctx context.Context,
 		return nil, err
 	}

-	// Validate URL format
+	// Validate URL format and security
 	logger.Info(ctx, "Validating URL")
-	if !isValidURL(url) {
-		logger.Error(ctx, "Invalid URL format")
+	if !isValidURL(url) || !secutils.IsValidURL(url) {
+		logger.Error(ctx, "Invalid or unsafe URL format")
 		return nil, ErrInvalidURL
 	}

@@ -339,6 +347,17 @@ func (s *knowledgeService) CreateKnowledgeFromPassage(ctx context.Context,
 	logger.Info(ctx, "Start creating knowledge from passage")
 	logger.Infof(ctx, "Knowledge base ID: %s, passage count: %d", kbID, len(passage))

+	// 验证段落内容安全性
+	safePassages := make([]string, 0, len(passage))
+	for i, p := range passage {
+		safePassage, isValid := secutils.ValidateInput(p)
+		if !isValid {
+			logger.Errorf(ctx, "Invalid passage content at index %d", i)
+			return nil, werrors.NewValidationError(fmt.Sprintf("段落 %d 包含非法内容", i+1))
+		}
+		safePassages = append(safePassages, safePassage)
+	}
+
 	// Get knowledge base configuration
 	logger.Info(ctx, "Getting knowledge base configuration")
 	kb, err := s.kbService.GetKnowledgeBaseByID(ctx, kbID)
@@ -370,7 +389,7 @@ func (s *knowledgeService) CreateKnowledgeFromPassage(ctx context.Context,

 	// Process passages asynchronously
 	logger.Info(ctx, "Starting asynchronous passage processing")
-	go s.processDocumentFromPassage(ctx, kb, knowledge, passage)
+	go s.processDocumentFromPassage(ctx, kb, knowledge, safePassages)

 	logger.Infof(ctx, "Knowledge from passage created successfully, ID: %s", knowledge.ID)
 	return knowledge, nil
@@ -229,7 +229,7 @@ func initFileService(cfg *config.Config) (interfaces.FileService, error) {
 			false,
 		)
 	case "cos":
-		if os.Getenv("COS_APP_ID") == "" ||
+		if os.Getenv("COS_BUCKET_NAME") == "" ||
 			os.Getenv("COS_REGION") == "" ||
 			os.Getenv("COS_SECRET_ID") == "" ||
 			os.Getenv("COS_SECRET_KEY") == "" ||
@@ -237,7 +237,7 @@ func initFileService(cfg *config.Config) (interfaces.FileService, error) {
 			return nil, fmt.Errorf("missing COS configuration")
 		}
 		return file.NewCosFileService(
-			os.Getenv("COS_APP_ID"),
+			os.Getenv("COS_BUCKET_NAME"),
 			os.Getenv("COS_REGION"),
 			os.Getenv("COS_SECRET_ID"),
 			os.Getenv("COS_SECRET_KEY"),
@@ -8,6 +8,7 @@ import (
 	"github.com/Tencent/WeKnora/internal/logger"
 	"github.com/Tencent/WeKnora/internal/types"
 	"github.com/Tencent/WeKnora/internal/types/interfaces"
+	secutils "github.com/Tencent/WeKnora/internal/utils"
 	"github.com/gin-gonic/gin"
 )

@@ -52,6 +53,13 @@ func (h *ChunkHandler) ListKnowledgeChunks(c *gin.Context) {
 		return
 	}

+	// 对 chunk 内容进行安全清理
+	for _, chunk := range result.Data.([]*types.Chunk) {
+		if chunk.Content != "" {
+			chunk.Content = secutils.SanitizeForDisplay(chunk.Content)
+		}
+	}
+
 	logger.Infof(
 		ctx, "Successfully retrieved knowledge chunks list, knowledge ID: %s, total: %d",
 		knowledgeID, result.Total,
@@ -180,6 +180,10 @@ func (h *InitializationHandler) CheckStatus(c *gin.Context) {
 		})
 		return
 	}
+	// ignore api key in response for security
+	for _, model := range models {
+		model.Parameters.APIKey = ""
+	}

 	logger.Info(ctx, "System is already initialized")
 	c.JSON(http.StatusOK, gin.H{
@@ -87,7 +87,7 @@ type CreateSessionRequest struct {
 func (h *SessionHandler) CreateSession(c *gin.Context) {
 	ctx := c.Request.Context()

-	logger.Info(ctx, "Start creating session")
+	logger.Infof(ctx, "Start creating session, config: %+v", h.config.Conversation)

 	// Parse and validate the request body
 	var request CreateSessionRequest
@@ -0,0 +1,170 @@
+package utils
+
+import (
+	"html"
+	"regexp"
+	"strings"
+	"unicode/utf8"
+)
+
+// XSS 防护相关正则表达式
+var (
+	// 匹配潜在的 XSS 攻击模式
+	xssPatterns = []*regexp.Regexp{
+		regexp.MustCompile(`(?i)<script[^>]*>.*?</script>`),
+		regexp.MustCompile(`(?i)<iframe[^>]*>.*?</iframe>`),
+		regexp.MustCompile(`(?i)<object[^>]*>.*?</object>`),
+		regexp.MustCompile(`(?i)<embed[^>]*>.*?</embed>`),
+		regexp.MustCompile(`(?i)<form[^>]*>.*?</form>`),
+		regexp.MustCompile(`(?i)<input[^>]*>`),
+		regexp.MustCompile(`(?i)<button[^>]*>.*?</button>`),
+		regexp.MustCompile(`(?i)javascript:`),
+		regexp.MustCompile(`(?i)vbscript:`),
+		regexp.MustCompile(`(?i)onload\s*=`),
+		regexp.MustCompile(`(?i)onerror\s*=`),
+		regexp.MustCompile(`(?i)onclick\s*=`),
+		regexp.MustCompile(`(?i)onmouseover\s*=`),
+		regexp.MustCompile(`(?i)onfocus\s*=`),
+		regexp.MustCompile(`(?i)onblur\s*=`),
+	}
+)
+
+// SanitizeHTML 清理 HTML 内容，防止 XSS 攻击
+func SanitizeHTML(input string) string {
+	if input == "" {
+		return ""
+	}
+
+	// 检查输入长度
+	if len(input) > 10000 {
+		input = input[:10000]
+	}
+
+	// 检查是否包含潜在的 XSS 攻击
+	for _, pattern := range xssPatterns {
+		if pattern.MatchString(input) {
+			// 如果包含恶意内容，进行 HTML 转义
+			return html.EscapeString(input)
+		}
+	}
+
+	// 如果内容相对安全，返回原内容
+	return input
+}
+
+// EscapeHTML 转义 HTML 特殊字符
+func EscapeHTML(input string) string {
+	if input == "" {
+		return ""
+	}
+	return html.EscapeString(input)
+}
+
+// ValidateInput 验证用户输入
+func ValidateInput(input string) (string, bool) {
+	if input == "" {
+		return "", true
+	}
+
+	// 检查长度
+	if len(input) > 10000 {
+		return "", false
+	}
+
+	// 检查是否包含控制字符
+	for _, r := range input {
+		if r < 32 && r != 9 && r != 10 && r != 13 {
+			return "", false
+		}
+	}
+
+	// 检查 UTF-8 有效性
+	if !utf8.ValidString(input) {
+		return "", false
+	}
+
+	// 检查是否包含潜在的 XSS 攻击
+	for _, pattern := range xssPatterns {
+		if pattern.MatchString(input) {
+			return "", false
+		}
+	}
+
+	return strings.TrimSpace(input), true
+}
+
+// IsValidURL 验证 URL 是否安全
+func IsValidURL(url string) bool {
+	if url == "" {
+		return false
+	}
+
+	// 检查长度
+	if len(url) > 2048 {
+		return false
+	}
+
+	// 检查协议
+	if !strings.HasPrefix(strings.ToLower(url), "http://") &&
+		!strings.HasPrefix(strings.ToLower(url), "https://") {
+		return false
+	}
+
+	// 检查是否包含恶意内容
+	for _, pattern := range xssPatterns {
+		if pattern.MatchString(url) {
+			return false
+		}
+	}
+
+	return true
+}
+
+// IsValidImageURL 验证图片 URL 是否安全
+func IsValidImageURL(url string) bool {
+	if !IsValidURL(url) {
+		return false
+	}
+
+	// 检查是否为图片文件
+	imageExtensions := []string{".jpg", ".jpeg", ".png", ".gif", ".webp", ".svg", ".bmp", ".ico"}
+	lowerURL := strings.ToLower(url)
+
+	for _, ext := range imageExtensions {
+		if strings.Contains(lowerURL, ext) {
+			return true
+		}
+	}
+
+	return false
+}
+
+// CleanMarkdown 清理 Markdown 内容
+func CleanMarkdown(input string) string {
+	if input == "" {
+		return ""
+	}
+
+	// 移除潜在的恶意脚本
+	cleaned := input
+	for _, pattern := range xssPatterns {
+		cleaned = pattern.ReplaceAllString(cleaned, "")
+	}
+
+	return strings.TrimSpace(cleaned)
+}
+
+// SanitizeForDisplay 为显示清理内容
+func SanitizeForDisplay(input string) string {
+	if input == "" {
+		return ""
+	}
+
+	// 首先清理 Markdown
+	cleaned := CleanMarkdown(input)
+
+	// 然后进行 HTML 转义
+	escaped := html.EscapeString(cleaned)
+
+	return escaped
+}
@@ -123,6 +123,7 @@ build_docreader_image() {
    
    docker build \
        --platform $PLATFORM \
+        --build-arg PLATFORM=$PLATFORM \
        --build-arg TARGETARCH=$TARGETARCH \
        -f docker/Dockerfile.docreader \
        -t wechatopenai/weknora-docreader:latest \