diff --git a/api/models.py b/api/models.py
index 375d2e9..3f403f0 100644
--- a/api/models.py
+++ b/api/models.py
@@ -146,9 +146,14 @@ class SearchRequest(BaseModel):
     debug: bool = Field(False, description="是否返回调试信息")
     
     # SKU筛选参数
-    sku_filter_dimension: Optional[str] = Field(
+    sku_filter_dimension: Optional[List[str]] = Field(
         None,
-        description="子SKU筛选维度（店铺配置）。指定后，每个SPU下的SKU将按该维度分组，每组选择第一个SKU返回。例如：'color'表示按颜色分组，每种颜色选一款。支持的值：'option1'、'option2'、'option3'或specifications中的name（如'color'、'size'）"
+        description=(
+            "子SKU筛选维度（店铺配置），为字符串列表。"
+            "指定后，每个SPU下的SKU将按这些维度的组合进行分组，每个维度组合只保留一个SKU返回。"
+            "例如：['color'] 表示按颜色分组，每种颜色选一款；['color', 'size'] 表示按颜色+尺码组合分组。"
+            "支持的值：'option1'、'option2'、'option3' 或选项名称（如 'color'、'size'，将通过 option1_name/2_name/3_name 匹配）。"
+        )
     )
     
     # 个性化参数（预留）
diff --git a/api/result_formatter.py b/api/result_formatter.py
index 936bb0a..6fdf947 100644
--- a/api/result_formatter.py
+++ b/api/result_formatter.py
@@ -14,7 +14,7 @@ class ResultFormatter:
         es_hits: List[Dict[str, Any]],
         max_score: float = 1.0,
         language: str = "zh",
-        sku_filter_dimension: Optional[str] = None
+        sku_filter_dimension: Optional[List[str]] = None
     ) -> List[SpuResult]:
         """
         Convert ES hits to SpuResult list.
@@ -85,10 +85,10 @@ class ResultFormatter:
                     )
                     skus.append(sku)
 
-            # Apply SKU filtering if dimension is specified
+            # Apply SKU filtering if dimension list is specified
             if sku_filter_dimension and skus:
-                skus = ResultFormatter._filter_skus_by_dimension(
-                    skus, 
+                skus = ResultFormatter._filter_skus_by_dimensions(
+                    skus,
                     sku_filter_dimension,
                     source.get('option1_name'),
                     source.get('option2_name'),
@@ -138,22 +138,22 @@ class ResultFormatter:
         return results
 
     @staticmethod
-    def _filter_skus_by_dimension(
+    def _filter_skus_by_dimensions(
         skus: List[SkuResult],
-        dimension: str,
+        dimensions: List[str],
         option1_name: Optional[str] = None,
         option2_name: Optional[str] = None,
         option3_name: Optional[str] = None,
         specifications: Optional[List[Dict[str, Any]]] = None
     ) -> List[SkuResult]:
         """
-        Filter SKUs by dimension, keeping only one SKU per dimension value.
+        Filter SKUs by one or more dimensions, keeping only one SKU per dimension value combination.
         
         Args:
             skus: List of SKU results to filter
-            dimension: Filter dimension, can be:
+            dimensions: Filter dimensions, each dimension can be:
                 - 'option1', 'option2', 'option3': Direct option field
-                - A specification name (e.g., 'color', 'size'): Match by option name
+                - A specification/option name (e.g., 'color', 'size'): Match by option name
             option1_name: Name of option1 (e.g., 'color')
             option2_name: Name of option2 (e.g., 'size')
             option3_name: Name of option3
@@ -162,54 +162,59 @@ class ResultFormatter:
         Returns:
             Filtered list of SKUs (one per dimension value)
         """
-        if not skus:
+        if not skus or not dimensions:
             return skus
-        
-        # Determine which field to use for filtering
-        filter_field = None
-        
-        # Direct option field (option1, option2, option3)
-        if dimension.lower() == 'option1':
-            filter_field = 'option1_value'
-        elif dimension.lower() == 'option2':
-            filter_field = 'option2_value'
-        elif dimension.lower() == 'option3':
-            filter_field = 'option3_value'
-        else:
-            # Try to match by option name
-            dimension_lower = dimension.lower()
-            if option1_name and option1_name.lower() == dimension_lower:
-                filter_field = 'option1_value'
-            elif option2_name and option2_name.lower() == dimension_lower:
-                filter_field = 'option2_value'
-            elif option3_name and option3_name.lower() == dimension_lower:
-                filter_field = 'option3_value'
-                
-        # If no matching field found, return all SKUs (no filtering)
-        if not filter_field:
-            return skus
-        
-        # Group SKUs by dimension value and select first one from each group
-        dimension_groups: Dict[str, SkuResult] = {}
-        
+
+        # Resolve each dimension to an underlying SKU field (option1_value / option2_value / option3_value)
+        filter_fields: List[str] = []
+
+        for dim in dimensions:
+            if not dim:
+                continue
+            dim_lower = dim.lower()
+
+            field_name: Optional[str] = None
+            # Direct option field (option1, option2, option3)
+            if dim_lower == 'option1':
+                field_name = 'option1_value'
+            elif dim_lower == 'option2':
+                field_name = 'option2_value'
+            elif dim_lower == 'option3':
+                field_name = 'option3_value'
+            else:
+                # Try to match by option name
+                if option1_name and option1_name.lower() == dim_lower:
+                    field_name = 'option1_value'
+                elif option2_name and option2_name.lower() == dim_lower:
+                    field_name = 'option2_value'
+                elif option3_name and option3_name.lower() == dim_lower:
+                    field_name = 'option3_value'
+
+            if field_name and field_name not in filter_fields:
+                filter_fields.append(field_name)
+
+        # If no matching field found for all dimensions, do not return any child SKUs
+        if not filter_fields:
+            return []
+
+        # Group SKUs by dimension value combination and select first one from each group
+        dimension_groups: Dict[tuple, SkuResult] = {}
+
         for sku in skus:
-            # Get dimension value from the determined field
-            dimension_value = None
-            if filter_field == 'option1_value':
-                dimension_value = sku.option1_value
-            elif filter_field == 'option2_value':
-                dimension_value = sku.option2_value
-            elif filter_field == 'option3_value':
-                dimension_value = sku.option3_value
-            
-            # Use empty string as key for None values
-            key = str(dimension_value) if dimension_value is not None else ''
-            
-            # Keep first SKU for each dimension value
+            # Build key as combination of all dimension values
+            key_values: List[str] = []
+            for field in filter_fields:
+                dimension_value = getattr(sku, field, None)
+                # Use empty string as key part for None values
+                key_values.append(str(dimension_value) if dimension_value is not None else '')
+
+            key = tuple(key_values)
+
+            # Keep first SKU for each dimension combination
             if key not in dimension_groups:
                 dimension_groups[key] = sku
-        
-        # Return filtered SKUs (one per dimension value)
+
+        # Return filtered SKUs (one per dimension combination)
         return list(dimension_groups.values())
 
     @staticmethod
diff --git a/docs/搜索API对接指南.md b/docs/搜索API对接指南.md
index f0ddca5..545a953 100644
--- a/docs/搜索API对接指南.md
+++ b/docs/搜索API对接指南.md
@@ -104,7 +104,7 @@ curl -X POST "http://120.76.41.98:6002/search/" \
   "sort_by": "string",
   "sort_order": "desc",
   "min_score": 0.0,
-  "sku_filter_dimension": "string",
+  "sku_filter_dimension": ["string"],
   "debug": false,
   "user_id": "string",
   "session_id": "string"
@@ -127,7 +127,7 @@ curl -X POST "http://120.76.41.98:6002/search/" \
 | `sort_by` | string | N | null | 排序字段名（如 `min_price`, `max_price`） |
 | `sort_order` | string | N | "desc" | 排序方向：`asc`（升序）或 `desc`（降序） |
 | `min_score` | float | N | null | 最小相关性分数阈值 |
-| `sku_filter_dimension` | string | N | null | 子SKU筛选维度（店铺配置）。指定后，每个SPU下的SKU将按该维度分组，每组选择第一个SKU返回。支持的值：`option1`、`option2`、`option3` 或 specifications 中的 name（如 `color`、`size`）。详见下文说明 |
+| `sku_filter_dimension` | array[string] | N | null | 子SKU筛选维度列表（店铺配置）。指定后，每个SPU下的SKU将按这些维度的组合进行分组，每个组合只返回第一个SKU。支持的值：`option1`、`option2`、`option3` 或选项名称（如 `color`、`size`）。详见下文说明 |
 | `debug` | boolean | N | false | 是否返回调试信息 |
 | `user_id` | string | N | null | 用户ID（用于个性化，预留） |
 | `session_id` | string | N | null | 会话ID（用于分析，预留） |
@@ -349,7 +349,7 @@ curl -X POST "http://120.76.41.98:6002/search/" \
 ### SKU筛选维度 (sku_filter_dimension)
 
 **功能说明**:
-`sku_filter_dimension` 用于控制每个SPU下返回的SKU数量。当指定此参数后，系统会按指定维度对SKU进行分组，每个分组只返回第一个SKU（从简实现，选择该维度下的第一款）。
+`sku_filter_dimension` 用于控制每个SPU下返回的SKU数量，为字符串列表。当指定此参数后，系统会按指定维度**组合**对SKU进行分组，每个维度组合只返回第一个SKU（从简实现，选择该组合下的第一款）。
 
 **使用场景**:
 - 店铺配置了SKU筛选维度（如 `color`），希望每个SPU下每种颜色只显示一个SKU
@@ -360,8 +360,8 @@ curl -X POST "http://120.76.41.98:6002/search/" \
 1. **直接选项字段**: `option1`、`option2`、`option3`
    - 直接使用对应的 `option1_value`、`option2_value`、`option3_value` 字段进行分组
    
-2. **规格名称**: 通过 `option1_name`、`option2_name`、`option3_name` 匹配
-   - 例如：如果 `option1_name` 为 `"color"`，则可以使用 `sku_filter_dimension: "color"` 来按颜色分组
+2. **规格/选项名称**: 通过 `option1_name`、`option2_name`、`option3_name` 匹配
+   - 例如：如果 `option1_name` 为 `"color"`，则可以使用 `sku_filter_dimension: ["color"]` 来按颜色分组
 
 **示例**:
 
@@ -369,7 +369,7 @@ curl -X POST "http://120.76.41.98:6002/search/" \
 ```json
 {
   "query": "芭比娃娃",
-  "sku_filter_dimension": "color"
+  "sku_filter_dimension": ["color"]
 }
 ```
 
@@ -377,7 +377,7 @@ curl -X POST "http://120.76.41.98:6002/search/" \
 ```json
 {
   "query": "芭比娃娃",
-  "sku_filter_dimension": "option1"
+  "sku_filter_dimension": ["option1"]
 }
 ```
 
@@ -385,7 +385,15 @@ curl -X POST "http://120.76.41.98:6002/search/" \
 ```json
 {
   "query": "芭比娃娃",
-  "sku_filter_dimension": "option2"
+  "sku_filter_dimension": ["option2"]
+}
+```
+
+**按颜色 + 尺寸组合筛选（假设 option1_name = "color", option2_name = "size"）**:
+```json
+{
+  "query": "芭比娃娃",
+  "sku_filter_dimension": ["color", "size"]
 }
 ```
 
diff --git a/frontend/index.html b/frontend/index.html
index 5d579f7..ac0e3c7 100644
--- a/frontend/index.html
+++ b/frontend/index.html
@@ -135,6 +135,6 @@
         <p>SearchEngine © 2025 | API: <span id="apiUrl">Loading...</span></p>
     </footer>
 
-    <script src="/static/js/app.js?v=3.3"></script>
+    <script src="/static/js/app.js?v=3.4"></script>
 </body>
 </html>
diff --git a/frontend/static/js/app.js b/frontend/static/js/app.js
index de9e667..9ad3058 100644
--- a/frontend/static/js/app.js
+++ b/frontend/static/js/app.js
@@ -14,13 +14,17 @@ function getTenantId() {
     return '1'; // Default fallback
 }
 
-// Get sku_filter_dimension from input
+// Get sku_filter_dimension (as list) from input
 function getSkuFilterDimension() {
     const skuFilterInput = document.getElementById('skuFilterDimension');
     if (skuFilterInput) {
         const value = skuFilterInput.value.trim();
-        // Return the value if not empty, otherwise return null
-        return value.length > 0 ? value : null;
+        if (!value.length) {
+            return null;
+        }
+        // 支持用逗号分隔多个维度，例如：color,size 或 option1,color
+        const parts = value.split(',').map(v => v.trim()).filter(v => v.length > 0);
+        return parts.length > 0 ? parts : null;
     }
     return null;
 }
diff --git a/search/searcher.py b/search/searcher.py
index af8faba..b8902c5 100644
--- a/search/searcher.py
+++ b/search/searcher.py
@@ -130,7 +130,7 @@ class Searcher:
         sort_order: Optional[str] = "desc",
         debug: bool = False,
         language: str = "zh",
-        sku_filter_dimension: Optional[str] = None,
+        sku_filter_dimension: Optional[List[str]] = None,
     ) -> SearchResult:
         """
         Execute search query (外部友好格式).
--
libgit2 0.21.2