diff --git a/backend/internal/app/cloud_agent_storyboard_test.go b/backend/internal/app/cloud_agent_storyboard_test.go
index c5b3883a4..49400403c 100644
--- a/backend/internal/app/cloud_agent_storyboard_test.go
+++ b/backend/internal/app/cloud_agent_storyboard_test.go
@@ -102,6 +102,16 @@ func TestCloudAgentNodeCapabilityCardsExplainStoryboardTradeoffs(t *testing.T) {
 	if !strings.Contains(videoGoodFor, "多图参考视频") {
 		t.Fatalf("video capability card does not explain multi-image video: %s", videoGoodFor)
 	}
+	frame := byType["frame"]
+	frameGoodFor := strings.Join(frame["goodFor"].([]string), "\n")
+	frameNotIdealFor := strings.Join(frame["notIdealFor"].([]string), "\n")
+	if !strings.Contains(frameGoodFor, "按场景或镜头组") || !strings.Contains(frameGoodFor, "制作") || !strings.Contains(frameNotIdealFor, "分镜脚本节点") {
+		t.Fatalf("frame capability card does not explain its use cases and limits: %+v", frame)
+	}
+	capabilityGuide := cloudAgentCapabilityGuide()
+	if !strings.Contains(capabilityGuide, "按场景或镜头组") || !strings.Contains(capabilityGuide, "可移动、可折叠") {
+		t.Fatalf("Agent system guide does not explain when to use a canvas frame: %s", capabilityGuide)
+	}
 	guide := strings.Join(result["selectionGuide"].([]string), "\n")
 	if !strings.Contains(guide, "多镜头") || !strings.Contains(guide, "model_list") || !strings.Contains(guide, "不要为了形式") {
 		t.Fatalf("selection guide does not express soft routing: %s", guide)
diff --git a/backend/internal/canvas/capability/builtin.go b/backend/internal/canvas/capability/builtin.go
index ceb84676f..a633d6d0c 100644
--- a/backend/internal/canvas/capability/builtin.go
+++ b/backend/internal/canvas/capability/builtin.go
@@ -43,9 +43,9 @@ func BuiltinRegistry() *Registry {
 		}),
 		{
 			Type: "frame", Version: "1", Label: "背板", DefaultWidth: 760, DefaultHeight: 520,
-			Purpose:       "组织一组相关节点的画布区域。",
-			GoodFor:       []string{"按场景整理节点", "划分工作区域"},
-			NotIdealFor:   []string{"承载结构化镜头数据", "替代具体业务节点"},
+			Purpose:       "在画布上建立可移动、可折叠的视觉分区，用来归组相关节点；背板本身不承载创作正文或生成结果。",
+			GoodFor:       []string{"按场景或镜头组归拢脚本、参考图和生成结果", "按前期策划、制作、交付等阶段划分工作区", "为大型画布建立清晰分区，便于移动或折叠整组内容"},
+			NotIdealFor:   []string{"承载结构化镜头数据（应使用分镜脚本节点）", "存放需要 Agent 单独读写的正文（应使用文本或 Markdown 节点）", "替代具体业务节点或媒体生成节点"},
 			Tradeoffs:     []string{"改善空间组织但不增加内容结构或生成能力"},
 			SummaryFields: []string{"label"}, DetailFields: []string{"label"},
 			CreateMetadata: func(string) map[string]any {
diff --git a/web/src/components/canvas/nodes/markdown-node.tsx b/web/src/components/canvas/nodes/markdown-node.tsx
index 6b1a94ebd..c4548c275 100644
--- a/web/src/components/canvas/nodes/markdown-node.tsx
+++ b/web/src/components/canvas/nodes/markdown-node.tsx
@@ -39,7 +39,6 @@ export function MarkdownNodeContent({ node, theme }: MarkdownNodeContentProps) {
             data-canvas-no-zoom
             style={{ color: theme.node.text }}
             onWheel={(event) => event.stopPropagation()}
-            onMouseDown={(event) => event.stopPropagation()}
         >
             <AIMessageMarkdown>{source}</AIMessageMarkdown>
         </div>