diff --git a/docs/technical/【前端架构】图片画布编辑器MVP接入方案-2026-06-11.md b/docs/technical/【前端架构】图片画布编辑器MVP接入方案-2026-06-11.md
index 441a1dddd..9d3af0f74 100644
--- a/docs/technical/【前端架构】图片画布编辑器MVP接入方案-2026-06-11.md
+++ b/docs/technical/【前端架构】图片画布编辑器MVP接入方案-2026-06-11.md
@@ -111,6 +111,7 @@
 - 素材文件夹可以新建、折叠、重命名和删除；删除普通文件夹后，其素材移动到“项目素材”。普通上传默认落入“上传素材”文件夹；素材库缺少该文件夹时，前端在首次普通上传前创建一次并复用，拖到指定文件夹或点击指定文件夹上传时仍进入目标文件夹。
 - 上传按钮和拖拽上传都支持多文件；底部工具栏的上传入口选择文件后直接进入“上传素材”并在当前画布视口中心创建画布图层，素材栏文件夹内的上传入口只写入对应素材文件夹、不自动入画布；拖到文件夹或该文件夹内素材时进入目标文件夹；拖到画布时进入“上传素材”并在投放点创建画布图层。上传图片必须在创建占位素材、画布图层和账号级素材记录前先读取原图 Resolution，图层宽高、`originalWidth/originalHeight` 和素材库 `width/height` 都使用图片本身尺寸；仅在无法解析尺寸时才使用上传兜底尺寸。
 - 音频 / 视频素材卡和画布媒体图层必须提供稳定的非文字视觉预览：优先使用 `thumbnailSrc` / 视频 `poster`，没有真实首帧或音频封面时使用由媒体类型、素材名和地址派生的确定性视觉底图。视频图层使用原生 `<video controls preload="metadata" playsInline>` 播放，外层图层仍承接选择和拖拽语义；音频图层播放前继续通过 `/api/assets/read-url` 换签。画布素材导出按 `mediaType` 保留真实媒体格式：图片进入 `images/`，音频 / 视频进入 `media/`，文件扩展名从响应 MIME、`objectKey` 或源 URL 推断，不得把音频 / 视频导出成 PNG。
+- 画布素材 ZIP 的 `metadata.json` 只保存前端信息弹窗和导出文件列表可见的展示快照：项目标题、导出时间、图层标题、文件路径、类型、生成输入、模型显示名、Task 短 ID、Object 显示值、Resolution / 时长和导出错误。导出的生成输入只保留用户实际填写或选择的内容；系统默认兜底提示词、固定工作流提示词、内置图标描述、UI 提取素材固定提示词等内置提示词即使存在于历史 `generationInputs`，也不得写入导出元数据。不得把 `projectId`、`layerId`、`resourceId`、`sourceAssetId`、`sourceResourceId`、原始 `prompt` / `actualPrompt` / `provider` 或画布坐标、锁定、隐藏等布局状态写入导出元数据；Object 字段仅沿用信息弹窗当前可见值。
 - `生成角色动作` 的完成结果同样按视频素材处理：图层主 `src` 使用后端返回的 `previewVideoPath`，`mediaType` 固定为 `video`，`assetKind` 固定为 `character-animation`，首个透明帧只写入 `thumbnailSrc` 供 poster 和后续动作源使用。下载和 ZIP 导出必须因此得到 mp4/webm/mov 等真实视频文件，不能回退为首帧 PNG；后端抽帧、绿幕去背和帧素材落盘流程仍完整保留。
 - 素材面板支持按素材名、文件夹名、生成信息、模型、任务和媒体类型搜索，并支持选择模式框选，一次选中多个素材，并可批量移动或删除上传素材。
 - 图层面板支持按图层名、生成信息、模型、任务和媒体类型搜索；支持选择多个图层后创建图层组，组名和 groupId 随画布布局保存。
diff --git a/src/components/image-editor/ImageCanvasEditorTypes.ts b/src/components/image-editor/ImageCanvasEditorTypes.ts
index 98fc276f5..3153562df 100644
--- a/src/components/image-editor/ImageCanvasEditorTypes.ts
+++ b/src/components/image-editor/ImageCanvasEditorTypes.ts
@@ -453,44 +453,24 @@ export type CanvasAssetExportImage = {
 };
 
 export type CanvasAssetExportMetadata = {
-  projectId: string | null;
   projectTitle: string;
   exportedAt: string;
   layers: Array<{
-    layerId: string;
     title: string;
     file: string | null;
-    mediaType?: CanvasMediaType | null;
-    sourceType: CanvasSourceType;
-    prompt?: string | null;
-    actualPrompt?: string | null;
-    model?: string | null;
-    provider?: string | null;
-    taskId?: string | null;
-    objectKey?: string | null;
-    assetObjectId?: string | null;
-    sourceResourceId?: string | null;
-    sourceAssetId?: string | null;
-    exportError?: string;
-    canvas: {
-      x: number;
-      y: number;
-      width: number;
-      height: number;
-      originalWidth: number;
-      originalHeight: number;
-      zIndex: number;
-      groupId?: string | null;
-      hidden?: boolean;
-      locked?: boolean;
-      flipX?: boolean;
-      flipY?: boolean;
+    visible: {
+      type: string;
+      generationInputs: CanvasGenerationInputs | null;
+      model: string;
+      task: string;
+      object: string;
+      resolution?: string;
+      duration?: string;
     };
+    exportError?: string;
   }>;
   failedImages: Array<{
-    key: string;
     title: string;
-    src: string;
     error: string;
   }>;
 };
diff --git a/src/components/image-editor/ImageCanvasEditorView.test.tsx b/src/components/image-editor/ImageCanvasEditorView.test.tsx
index 54bc91db3..cdf7b3bba 100644
--- a/src/components/image-editor/ImageCanvasEditorView.test.tsx
+++ b/src/components/image-editor/ImageCanvasEditorView.test.tsx
@@ -362,9 +362,15 @@ describe('ImageCanvasEditorView', () => {
           sourceType: 'generated',
           sourceResourceId: 'resource-generated',
           prompt: '明亮主视觉',
+          actualPrompt: '后端拼接提示词',
           model: 'gpt-image-2',
           provider: 'VectorEngine',
           taskId: 'task-1',
+          assetObjectId: 'asset-object-visible',
+          generationInputs: {
+            fields: [{ title: '生成提示词', value: '明亮主视觉' }],
+            references: [],
+          },
         },
         {
           layerId: 'layer-failed',
@@ -460,19 +466,33 @@ describe('ImageCanvasEditorView', () => {
       const metadata = JSON.parse(
         await readZipText(zip, '导出项目-画布素材/metadata.json'),
       );
-      expect(metadata.projectId).toBe('editor-project-export');
+      expect(metadata.projectId).toBeUndefined();
       expect(metadata.layers).toHaveLength(4);
       expect(metadata.layers[0].file).toBe('images/001-素材 A.png');
       expect(metadata.layers[1].file).toBe('images/001-素材 A.png');
-      expect(metadata.layers[0].canvas.hidden).toBe(true);
-      expect(metadata.layers[0].canvas.locked).toBe(true);
-      expect(metadata.layers[0].canvas.flipX).toBe(true);
-      expect(metadata.layers[0].canvas.groupId).toBe('group-a');
-      expect(metadata.layers[2].sourceType).toBe('generated');
-      expect(metadata.layers[2].prompt).toBe('明亮主视觉');
+      expect(metadata.layers[0].canvas).toBeUndefined();
+      expect(metadata.layers[0].visible.type).toBe('上传图片');
+      expect(metadata.layers[2].visible).toEqual({
+        type: '生成图片',
+        generationInputs: {
+          fields: [{ title: '生成提示词', value: '明亮主视觉' }],
+          references: [],
+        },
+        model: 'gpt-image-2',
+        task: '1',
+        object: 'asset-object-visible',
+        resolution: '1024 x 1024 px',
+      });
+      expect(metadata.layers[2].sourceType).toBeUndefined();
+      expect(metadata.layers[2].prompt).toBeUndefined();
+      expect(metadata.layers[2].actualPrompt).toBeUndefined();
+      expect(metadata.layers[2].provider).toBeUndefined();
+      expect(metadata.layers[2].sourceResourceId).toBeUndefined();
       expect(metadata.layers[3].file).toBeNull();
       expect(metadata.layers[3].exportError).toContain('404');
-      expect(metadata.failedImages).toHaveLength(1);
+      expect(metadata.failedImages).toEqual([
+        { title: '失败图', error: expect.stringContaining('404') },
+      ]);
       expect(
         await readZipText(zip, '导出项目-画布素材/manifest.txt'),
       ).toContain('失败素材数量：1');
diff --git a/src/components/image-editor/ImageCanvasExportModel.test.ts b/src/components/image-editor/ImageCanvasExportModel.test.ts
index a52aea87b..86351dc8a 100644
--- a/src/components/image-editor/ImageCanvasExportModel.test.ts
+++ b/src/components/image-editor/ImageCanvasExportModel.test.ts
@@ -11,6 +11,10 @@ import {
   sanitizeExportFilePart,
 } from './ImageCanvasExportModel';
 import type { CanvasLayer } from './ImageCanvasEditorTypes';
+import {
+  DEFAULT_ICON_DESCRIPTIONS,
+  UI_DESIGN_ASSET_EXTRACTION_PROMPT,
+} from './ImageCanvasGenerationModel';
 
 describe('ImageCanvasExportModel', () => {
   it('normalizes export file names and dates', () => {
@@ -100,35 +104,67 @@ describe('ImageCanvasExportModel', () => {
     expect(await blob.text()).toBe('Hello');
 
     expect(buildLayerExportMetadata(buildLayer(), 'images/001-layer.png')).toEqual({
-      layerId: 'layer-1',
       title: '导出图层',
       file: 'images/001-layer.png',
-      mediaType: 'image',
-      sourceType: 'generated',
-      prompt: '生成提示',
-      actualPrompt: '实际提示',
-      model: 'gpt-image-2',
-      provider: 'VectorEngine',
-      taskId: 'task-1',
-      objectKey: 'generated/layer.png',
-      assetObjectId: undefined,
-      sourceResourceId: 'source-resource',
-      sourceAssetId: 'asset-1',
-      exportError: undefined,
-      canvas: {
-        x: 10,
-        y: 20,
-        width: 512,
-        height: 512,
-        originalWidth: 1024,
-        originalHeight: 1024,
-        zIndex: 3,
-        groupId: 'group-1',
-        hidden: true,
-        locked: false,
-        flipX: true,
-        flipY: false,
+      visible: {
+        type: '生成图片',
+        generationInputs: {
+          fields: [{ title: '生成提示词', value: '清爽游戏按钮' }],
+          references: [
+            {
+              title: '参考图',
+              label: '角色立绘',
+              refType: 'project-resource',
+              refId: 'resource-reference',
+            },
+          ],
+        },
+        model: 'gpt-image-2',
+        task: '1',
+        object: 'generated/layer.png',
+        resolution: '1024 x 1024 px',
       },
+      exportError: undefined,
+    });
+  });
+
+  it('filters built-in prompts from exported visible generation inputs', () => {
+    const metadata = buildLayerExportMetadata(
+      buildLayer({
+        generationInputs: {
+          fields: [
+            { title: '生成提示词', value: 'AI 生成图片' },
+            { title: '修改要求', value: '修改当前图片' },
+            { title: 'prompt', value: '游戏音效' },
+            {
+              title: 'gpt_description_prompt',
+              value: '游戏背景音乐',
+            },
+            {
+              title: '提取提示词',
+              value: UI_DESIGN_ASSET_EXTRACTION_PROMPT,
+            },
+            {
+              title: '素材描述',
+              value: DEFAULT_ICON_DESCRIPTIONS.join('\n'),
+            },
+            { title: '用户输入', value: '用户填写的视觉要求' },
+            { title: '游戏分类', value: '游戏音效' },
+            { title: 'duration', value: '5秒' },
+          ],
+          references: [],
+        },
+      }),
+      'images/001-layer.png',
+    );
+
+    expect(metadata.visible.generationInputs).toEqual({
+      fields: [
+        { title: '用户输入', value: '用户填写的视觉要求' },
+        { title: '游戏分类', value: '游戏音效' },
+        { title: 'duration', value: '5秒' },
+      ],
+      references: [],
     });
   });
 
@@ -203,6 +239,17 @@ function buildLayer(overrides: Partial<CanvasLayer> = {}): CanvasLayer {
     objectKey: 'generated/layer.png',
     sourceResourceId: 'source-resource',
     sourceAssetId: 'asset-1',
+    generationInputs: {
+      fields: [{ title: '生成提示词', value: '清爽游戏按钮' }],
+      references: [
+        {
+          title: '参考图',
+          label: '角色立绘',
+          refType: 'project-resource',
+          refId: 'resource-reference',
+        },
+      ],
+    },
     groupId: 'group-1',
     hidden: true,
     locked: false,
diff --git a/src/components/image-editor/ImageCanvasExportModel.ts b/src/components/image-editor/ImageCanvasExportModel.ts
index d8e7030f6..19105cf80 100644
--- a/src/components/image-editor/ImageCanvasExportModel.ts
+++ b/src/components/image-editor/ImageCanvasExportModel.ts
@@ -1,8 +1,16 @@
 import type {
   CanvasAssetExportMetadata,
+  CanvasGenerationInputField,
   CanvasMediaType,
   CanvasLayer,
 } from './ImageCanvasEditorTypes';
+import {
+  DEFAULT_ICON_DESCRIPTIONS,
+  formatLayerImageType,
+  getEditorImageModelDisplayName,
+  UI_DESIGN_ASSET_EXTRACTION_PROMPT,
+} from './ImageCanvasGenerationModel';
+import { formatCanvasDurationMetric } from './ImageCanvasMediaModel';
 import {
   getSignedAssetReadUrl,
   resolveAssetReadUrl,
@@ -232,40 +240,91 @@ export async function blobToUint8Array(blob: Blob) {
   });
 }
 
+export function formatTaskIdForDisplay(taskId?: string | null) {
+  const numericParts = taskId?.match(/\d+/gu);
+  return numericParts?.length
+    ? (numericParts[numericParts.length - 1] ?? '-')
+    : '-';
+}
+
+const BUILT_IN_GENERATION_INPUT_TITLES = new Set(['提取提示词']);
+const BUILT_IN_GENERATION_INPUT_VALUES_BY_TITLE = new Map(
+  Object.entries({
+    生成提示词: new Set(['AI 生成图片']),
+    修改要求: new Set(['修改当前图片']),
+    快速编辑提示词: new Set(['修改当前图片']),
+    重绘提示词: new Set(['修改当前图片']),
+    prompt: new Set(['游戏音效']),
+    gpt_description_prompt: new Set(['游戏背景音乐']),
+    素材描述: new Set([DEFAULT_ICON_DESCRIPTIONS.join('\n')]),
+  }),
+);
+
+function isBuiltInGenerationInputField(field: CanvasGenerationInputField) {
+  const normalizedTitle = field.title.trim();
+  const normalizedValue = field.value.trim();
+  const builtInValues =
+    BUILT_IN_GENERATION_INPUT_VALUES_BY_TITLE.get(normalizedTitle);
+  return (
+    BUILT_IN_GENERATION_INPUT_TITLES.has(normalizedTitle) ||
+    normalizedValue === UI_DESIGN_ASSET_EXTRACTION_PROMPT ||
+    Boolean(builtInValues?.has(normalizedValue))
+  );
+}
+
+function buildVisibleGenerationInputs(layer: CanvasLayer) {
+  const fields =
+    layer.generationInputs?.fields
+      .filter((field) => !isBuiltInGenerationInputField(field))
+      .map((field) => ({
+        title: field.title,
+        value: field.value,
+      })) ?? [];
+  const references =
+    layer.generationInputs?.references.map((reference) => ({
+      title: reference.title,
+      label: reference.label,
+      refType: reference.refType,
+      refId: reference.refId,
+    })) ?? [];
+
+  return fields.length || references.length ? { fields, references } : null;
+}
+
+export function buildLayerVisibleExportMetadata(layer: CanvasLayer) {
+  const base = {
+    type: formatLayerImageType(layer),
+    generationInputs: buildVisibleGenerationInputs(layer),
+    model: layer.model ? getEditorImageModelDisplayName(layer.model) : '-',
+    task: formatTaskIdForDisplay(layer.taskId),
+    object: layer.objectKey ?? layer.assetObjectId ?? '-',
+  };
+
+  if (layer.mediaType === 'audio') {
+    return {
+      ...base,
+      duration: formatCanvasDurationMetric(layer.durationSeconds).replace(
+        /^时长\s*/u,
+        '',
+      ),
+    };
+  }
+
+  return {
+    ...base,
+    resolution: `${layer.originalWidth} x ${layer.originalHeight} px`,
+  };
+}
+
 export function buildLayerExportMetadata(
   layer: CanvasLayer,
   file: string | null,
   exportError?: string,
 ): CanvasAssetExportMetadata['layers'][number] {
   return {
-    layerId: layer.id,
     title: layer.title,
     file,
-    mediaType: layer.mediaType ?? 'image',
-    sourceType: layer.sourceType,
-    prompt: layer.prompt,
-    actualPrompt: layer.actualPrompt,
-    model: layer.model,
-    provider: layer.provider,
-    taskId: layer.taskId,
-    objectKey: layer.objectKey,
-    assetObjectId: layer.assetObjectId,
-    sourceResourceId: layer.sourceResourceId,
-    sourceAssetId: layer.sourceAssetId,
+    visible: buildLayerVisibleExportMetadata(layer),
     exportError,
-    canvas: {
-      x: layer.x,
-      y: layer.y,
-      width: layer.width,
-      height: layer.height,
-      originalWidth: layer.originalWidth,
-      originalHeight: layer.originalHeight,
-      zIndex: layer.zIndex,
-      groupId: layer.groupId,
-      hidden: layer.hidden,
-      locked: layer.locked,
-      flipX: layer.flipX,
-      flipY: layer.flipY,
-    },
   };
 }
diff --git a/src/components/image-editor/ImageCanvasMetadataModalView.tsx b/src/components/image-editor/ImageCanvasMetadataModalView.tsx
index ca7749893..4cda682d0 100644
--- a/src/components/image-editor/ImageCanvasMetadataModalView.tsx
+++ b/src/components/image-editor/ImageCanvasMetadataModalView.tsx
@@ -1,5 +1,6 @@
 import { UnifiedModal } from '../common/UnifiedModal';
 import type { CanvasLayer } from './ImageCanvasEditorTypes';
+import { formatTaskIdForDisplay } from './ImageCanvasExportModel';
 import {
   formatLayerImageType,
   getEditorImageModelDisplayName,
@@ -11,11 +12,6 @@ export type ImageCanvasMetadataModalViewProps = {
   onClose: () => void;
 };
 
-function formatTaskIdForDisplay(taskId?: string | null) {
-  const numericParts = taskId?.match(/\d+/gu);
-  return numericParts?.length ? numericParts[numericParts.length - 1] : '-';
-}
-
 function formatReferenceType(refType: 'project-resource' | 'asset') {
   return refType === 'project-resource' ? '项目资源' : '素材';
 }
diff --git a/src/components/image-editor/useImageCanvasAssetExportWorkflow.test.tsx b/src/components/image-editor/useImageCanvasAssetExportWorkflow.test.tsx
index 694ffde12..d69956dd6 100644
--- a/src/components/image-editor/useImageCanvasAssetExportWorkflow.test.tsx
+++ b/src/components/image-editor/useImageCanvasAssetExportWorkflow.test.tsx
@@ -128,6 +128,26 @@ describe('useImageCanvasAssetExportWorkflow', () => {
               src: '/generated-ok.png',
               sourceType: 'generated',
               prompt: '明亮主视觉',
+              actualPrompt: '后端拼接提示词',
+              provider: 'VectorEngine',
+              taskId: 'task-987',
+              assetObjectId: 'asset-object-visible',
+              sourceResourceId: 'resource-generated',
+              sourceAssetId: 'asset-generated',
+              generationInputs: {
+                fields: [
+                  { title: '生成提示词', value: '明亮主视觉' },
+                  { title: '提取提示词', value: '系统固定提取提示词' },
+                ],
+                references: [
+                  {
+                    title: '参考图',
+                    label: '角色立绘',
+                    refType: 'project-resource',
+                    refId: 'resource-reference',
+                  },
+                ],
+              },
               zIndex: 3,
             }),
             createLayer('video', {
@@ -209,27 +229,63 @@ describe('useImageCanvasAssetExportWorkflow', () => {
       const metadata = JSON.parse(
         await readZipText(zip, '导出项目-画布素材/metadata.json'),
       );
-      expect(metadata.projectId).toBe('project-export');
+      expect(metadata.projectId).toBeUndefined();
       expect(metadata.layers).toHaveLength(7);
       expect(metadata.layers[0].file).toBe('images/001-素材 A.png');
       expect(metadata.layers[1].file).toBe('images/001-素材 A.png');
-      expect(metadata.layers[2].mediaType).toBe('image');
-      expect(metadata.layers[0].canvas.hidden).toBe(true);
-      expect(metadata.layers[0].canvas.locked).toBe(true);
-      expect(metadata.layers[0].canvas.flipX).toBe(true);
-      expect(metadata.layers[2].sourceType).toBe('generated');
-      expect(metadata.layers[2].prompt).toBe('明亮主视觉');
+      expect(metadata.layers[0].canvas).toBeUndefined();
+      expect(metadata.layers[0].visible.type).toBe('上传图片');
+      expect(metadata.layers[0].visible.resolution).toBe('100 x 80 px');
+      expect(metadata.layers[2].visible.type).toBe('生成图片');
+      expect(metadata.layers[2].visible.model).toBe('-');
+      expect(metadata.layers[2].visible.task).toBe('987');
+      expect(metadata.layers[2].visible.object).toBe('asset-object-visible');
+      expect(metadata.layers[2].visible.generationInputs).toEqual({
+        fields: [{ title: '生成提示词', value: '明亮主视觉' }],
+        references: [
+          {
+            title: '参考图',
+            label: '角色立绘',
+            refType: 'project-resource',
+            refId: 'resource-reference',
+          },
+        ],
+      });
+      expect(JSON.stringify(metadata)).not.toContain('系统固定提取提示词');
+      expect(metadata.layers[2].sourceType).toBeUndefined();
+      expect(metadata.layers[2].prompt).toBeUndefined();
+      expect(metadata.layers[2].actualPrompt).toBeUndefined();
+      expect(metadata.layers[2].provider).toBeUndefined();
+      expect(metadata.layers[2].sourceResourceId).toBeUndefined();
+      expect(metadata.layers[2].sourceAssetId).toBeUndefined();
       expect(metadata.layers[3].file).toBe('media/003-生成视频.mp4');
-      expect(metadata.layers[3].mediaType).toBe('video');
+      expect(metadata.layers[3].visible.type).toBe('生成视频');
       expect(metadata.layers[4].file).toBe('media/004-胜利音效.wav');
-      expect(metadata.layers[4].mediaType).toBe('audio');
+      expect(metadata.layers[4].visible.type).toBe('生成音频');
+      expect(metadata.layers[4].visible.duration).toBe('--:--');
       expect(metadata.layers[5].file).toBe(
         'sequences/005-角色动作/manifest.txt',
       );
-      expect(metadata.layers[5].mediaType).toBe('image-sequence');
+      expect(metadata.layers[5].visible.type).toBe('角色动作序列帧');
+      const sequenceMetadata = JSON.parse(
+        await readZipText(
+          zip,
+          '导出项目-画布素材/sequences/005-角色动作/metadata.json',
+        ),
+      );
+      expect(sequenceMetadata.visible.type).toBe('角色动作序列帧');
+      expect(sequenceMetadata.taskId).toBeUndefined();
+      expect(sequenceMetadata.previewVideoPath).toBeUndefined();
+      expect(sequenceMetadata.failedFrames).toEqual([]);
       expect(metadata.layers[6].file).toBeNull();
       expect(metadata.layers[6].exportError).toContain('404');
       expect(metadata.failedImages).toHaveLength(1);
+      expect(metadata.failedImages[0]).toEqual({
+        title: '失败图',
+        error: expect.stringContaining('404'),
+      });
+      expect(metadata.failedImages[0].key).toBeUndefined();
+      expect(metadata.failedImages[0].src).toBeUndefined();
       expect(
         await readZipText(zip, '导出项目-画布素材/manifest.txt'),
       ).toContain('失败素材数量：1');
@@ -348,6 +404,13 @@ describe('useImageCanvasAssetExportWorkflow', () => {
       expect(zip.file('frames/frame-01.png')).toBeTruthy();
       expect(zip.file('frames/frame-02.png')).toBeTruthy();
       expect(await readZipText(zip, 'manifest.txt')).toContain('类型：序列帧');
+      const metadata = JSON.parse(await readZipText(zip, 'metadata.json'));
+      expect(metadata.title).toBe('角色动作/跑步');
+      expect(metadata.visible.type).toBe('角色动作序列帧');
+      expect(metadata.visible.resolution).toBe('100 x 80 px');
+      expect(metadata.taskId).toBeUndefined();
+      expect(metadata.previewVideoPath).toBeUndefined();
+      expect(metadata.failedFrames).toEqual([]);
     } finally {
       delete (URL as unknown as { createObjectURL?: unknown }).createObjectURL;
       delete (URL as unknown as { revokeObjectURL?: unknown }).revokeObjectURL;
diff --git a/src/components/image-editor/useImageCanvasAssetExportWorkflow.ts b/src/components/image-editor/useImageCanvasAssetExportWorkflow.ts
index a44b58060..0d2699452 100644
--- a/src/components/image-editor/useImageCanvasAssetExportWorkflow.ts
+++ b/src/components/image-editor/useImageCanvasAssetExportWorkflow.ts
@@ -9,6 +9,7 @@ import type {
 import {
   blobToUint8Array,
   buildLayerExportMetadata,
+  buildLayerVisibleExportMetadata,
   formatExportDate,
   getImageSequenceFrameFileName,
   getLayerAssetExtensionFromTypeOrSrc,
@@ -57,7 +58,6 @@ async function buildImageSequenceZip(layer: CanvasLayer) {
   const framesFolder = zip.folder('frames') ?? zip;
   const failedFrames: Array<{
     frameIndex: number;
-    imageSrc: string;
     error: string;
   }> = [];
   let successCount = 0;
@@ -73,7 +73,6 @@ async function buildImageSequenceZip(layer: CanvasLayer) {
     } catch (error) {
       failedFrames.push({
         frameIndex: frame.frameIndex || index + 1,
-        imageSrc: frame.imageSrc,
         error: error instanceof Error ? error.message : '序列帧读取失败',
       });
     }
@@ -88,12 +87,9 @@ async function buildImageSequenceZip(layer: CanvasLayer) {
     JSON.stringify(
       {
         title: layer.title,
-        taskId: layer.taskId ?? null,
-        mediaType: 'image-sequence',
+        visible: buildLayerVisibleExportMetadata(layer),
         frameCount: frames.length,
         exportedFrameCount: successCount,
-        durationSeconds: layer.durationSeconds ?? null,
-        previewVideoPath: layer.previewVideoPath ?? null,
         failedFrames,
       },
       null,
@@ -118,7 +114,6 @@ async function buildImageSequenceZip(layer: CanvasLayer) {
 
 export function useImageCanvasAssetExportWorkflow({
   layers,
-  projectId,
   projectTitle,
 }: UseImageCanvasAssetExportWorkflowOptions) {
   const [assetExportStatus, setAssetExportStatus] =
@@ -224,7 +219,6 @@ export function useImageCanvasAssetExportWorkflow({
               sequenceFolder.folder('frames') ?? sequenceFolder;
             const failedFrames: Array<{
               frameIndex: number;
-              imageSrc: string;
               error: string;
             }> = [];
             let successCount = 0;
@@ -247,7 +241,6 @@ export function useImageCanvasAssetExportWorkflow({
               } catch (error) {
                 failedFrames.push({
                   frameIndex: frame.frameIndex || frameArrayIndex + 1,
-                  imageSrc: frame.imageSrc,
                   error:
                     error instanceof Error ? error.message : '序列帧读取失败',
                 });
@@ -262,14 +255,10 @@ export function useImageCanvasAssetExportWorkflow({
               'metadata.json',
               JSON.stringify(
                 {
-                  layerId: layer.id,
                   title: layer.title,
-                  mediaType: 'image-sequence',
-                  taskId: layer.taskId ?? null,
+                  visible: buildLayerVisibleExportMetadata(layer),
                   frameCount: frames.length,
                   exportedFrameCount: successCount,
-                  durationSeconds: layer.durationSeconds ?? null,
-                  previewVideoPath: layer.previewVideoPath ?? null,
                   failedFrames,
                 },
                 null,
@@ -352,7 +341,6 @@ export function useImageCanvasAssetExportWorkflow({
       }
 
       const metadata: CanvasAssetExportMetadata = {
-        projectId,
         projectTitle,
         exportedAt: exportedAt.toISOString(),
         layers: exportableLayers.map((layer) => {
@@ -364,9 +352,7 @@ export function useImageCanvasAssetExportWorkflow({
           );
         }),
         failedImages: failedImages.map((image) => ({
-          key: image.key,
           title: image.layer.title,
-          src: image.layer.src,
           error: image.error ?? '素材读取失败',
         })),
       };
@@ -415,7 +401,7 @@ export function useImageCanvasAssetExportWorkflow({
     } finally {
       setIsExportingAssets(false);
     }
-  }, [isExportingAssets, layers, projectId, projectTitle]);
+  }, [isExportingAssets, layers, projectTitle]);
 
   return {
     assetExportStatus,