diff --git a/README.md b/README.md
index 44dbdbf..90a259c 100644
--- a/README.md
+++ b/README.md
@@ -67,7 +67,7 @@ docker compose down
 docker compose down -v
 ```
 
-`scripts/demo_smoke.sh` 会检查健康状态、本地登录、统一生成、资产重试、故事列表和 Provider 能力分层。默认跳过 TTS；演示前需要验证语音链路时使用 `SMOKE_AUDIO=1`。
+`scripts/demo_smoke.sh` 会检查健康状态、本地登录、统一生成后台任务、主记录落库、资产重试、故事列表和 Provider 能力分层。默认跳过 TTS；演示前需要验证语音链路时使用 `SMOKE_AUDIO=1`。
 
 ## 手动开发
 
@@ -128,7 +128,7 @@ npm run build
 | GET | `/auth/github/signin` | GitHub OAuth 登录 |
 | GET | `/auth/google/signin` | Google OAuth 登录 |
 | GET | `/auth/session` | 当前会话 |
-| POST | `/api/generations` | 统一生成故事或绘本 |
+| POST | `/api/generations` | 提交后台生成任务，立即返回 `generation_job_id` |
 | GET | `/api/generations/{story_id}` | 统一读取生成结果 |
 | POST | `/api/generations/{story_id}/retry-assets` | 统一重试封面/语音资源 |
 | GET | `/api/generations/jobs/{job_id}` | 查询生成任务事件流 |
@@ -165,4 +165,4 @@ npm run build
 
 ## 当前取舍
 
-仓库只保留一个 Docker Compose 入口：`docker-compose.yml`。生产部署、HA 演练、旧 Claude 原型和历史归档已从主仓库移除，避免干扰当前求职演示主线。音频缓存默认按 `STORY_AUDIO_CACHE_TTL_DAYS=30` 做后台清理，Celery beat 会每日执行一次 prune；生成任务默认按 `GENERATION_JOB_STALE_MINUTES=60` 判定卡住，后台会定时自动收敛为失败态，避免故事长期显示“永远在跑”。
+仓库只保留一个 Docker Compose 入口：`docker-compose.yml`。生产部署、HA 演练、旧 Claude 原型和历史归档已从主仓库移除，避免干扰当前求职演示主线。统一生成接口现在会先创建后台任务，再由 Celery worker 负责真正的故事/绘本生成；因此本地开发或 Docker 演示时需要保证 `worker` 服务可用。音频缓存默认按 `STORY_AUDIO_CACHE_TTL_DAYS=30` 做后台清理，Celery beat 会每日执行一次 prune；生成任务默认按 `GENERATION_JOB_STALE_MINUTES=60` 判定卡住，后台会定时自动收敛为失败态，避免故事长期显示“永远在跑”。
diff --git a/admin-frontend/src/components/CreateStoryModal.vue b/admin-frontend/src/components/CreateStoryModal.vue
index 07984c3..e4ead09 100644
--- a/admin-frontend/src/components/CreateStoryModal.vue
+++ b/admin-frontend/src/components/CreateStoryModal.vue
@@ -1,11 +1,10 @@
 <script setup lang="ts">
 import { ref, computed, watch } from 'vue'
-import type { Component } from 'vue'
-import { useRouter } from 'vue-router'
-import { useI18n } from 'vue-i18n'
-import { useUserStore } from '../stores/user'
-import { useStorybookStore } from '../stores/storybook'
-import { api } from '../api/client'
+import type { Component } from 'vue'
+import { useRouter } from 'vue-router'
+import { useI18n } from 'vue-i18n'
+import { useUserStore } from '../stores/user'
+import { api } from '../api/client'
 import BaseButton from './ui/BaseButton.vue'
 import BaseInput from './ui/BaseInput.vue'
 import BaseSelect from './ui/BaseSelect.vue'
@@ -34,10 +33,9 @@ const emit = defineEmits<{
   'update:modelValue': [value: boolean]
 }>()
 
-const { t } = useI18n()
-const router = useRouter()
-const userStore = useUserStore()
-const storybookStore = useStorybookStore()
+const { t } = useI18n()
+const router = useRouter()
+const userStore = useUserStore()
 
 // State
 const inputType = ref<'keywords' | 'full_story'>('keywords')
@@ -88,28 +86,63 @@ const generationTitle = computed(() =>
 const generationSteps = computed(() => {
   if (requestedOutputMode.value === 'storybook') {
     return [
-      '正在整理主题和成长目标...',
-      '生成绘本分镜和每页文字...',
-      '保存绘本主记录，确保刷新也能找回...',
-      '补全封面和分页插图...',
-      '马上进入可翻页阅读模式。',
+      '正在提交后台任务...',
+      'Worker 会生成绘本分镜和每页文字...',
+      '主记录一落库就能通过 ID 找回...',
+      '插图会继续在后台补全...',
+      '稍后自动进入可翻页阅读模式。',
     ]
   }
 
   return [
-    '正在整理孩子档案和故事主题...',
-    '生成可先阅读的故事正文...',
-    '保存故事主记录，避免结果丢失...',
-    '补全封面图，失败也可稍后重试...',
+    '正在提交后台任务...',
+    'Worker 会生成故事正文并保存主记录...',
+    '主内容一可读就会自动跳转详情页...',
+    '封面会继续在后台补全，失败也能重试...',
     '马上进入故事详情页。',
   ]
 })
+
+interface GenerationAcceptedResponse {
+  id: number | null
+  generation_job_id: string | null
+}
+
+interface GenerationJobDetail {
+  story_id: number | null
+  is_terminal: boolean
+  error_message: string | null
+}
+
+const JOB_POLL_INTERVAL_MS = 1500
+const JOB_POLL_MAX_ATTEMPTS = 80
 
 // Methods
-function close() {
-  emit('update:modelValue', false)
-  error.value = ''
-}
+function close() {
+  emit('update:modelValue', false)
+  error.value = ''
+}
+
+function sleep(ms: number) {
+  return new Promise((resolve) => {
+    window.setTimeout(resolve, ms)
+  })
+}
+
+async function waitForStoryId(jobId: string) {
+  for (let attempt = 0; attempt < JOB_POLL_MAX_ATTEMPTS; attempt += 1) {
+    const detail = await api.get<GenerationJobDetail>(`/api/generations/jobs/${jobId}`)
+    if (detail.story_id) {
+      return detail.story_id
+    }
+    if (detail.is_terminal) {
+      throw new Error(detail.error_message || '生成失败，请稍后重试')
+    }
+    await sleep(JOB_POLL_INTERVAL_MS)
+  }
+
+  throw new Error('任务已提交，但主内容落库超时，请稍后到故事库查看最新结果')
+}
 
 async function fetchProfiles() {
   if (!userStore.user) return
@@ -173,25 +206,22 @@ async function generateStory() {
     if (selectedProfileId.value) payload.child_profile_id = selectedProfileId.value
     if (selectedUniverseId.value) payload.universe_id = selectedUniverseId.value
 
-    if (requestedOutputMode.value === 'storybook') {
-      const response = await api.post<any>('/api/generations', payload)
+    const accepted = await api.post<GenerationAcceptedResponse>('/api/generations', payload)
+    const jobId = accepted.generation_job_id
+    if (!jobId) {
+      throw new Error('生成任务已创建，但缺少任务编号')
+    }
 
-      storybookStore.setStorybook(response)
-      close()
-      const storybookPath = response.id ? `/storybook/view/${response.id}` : '/storybook/view'
-      router.push(storybookPath)
+    const storyId = accepted.id ?? await waitForStoryId(jobId)
+    close()
+    if (requestedOutputMode.value === 'storybook') {
+      router.push(`/storybook/view/${storyId}`)
     } else {
-      const result = await api.post<any>('/api/generations', payload)
-      const query: Record<string, string> = {}
-      if (result.errors && Object.keys(result.errors).length > 0) {
-        if (result.errors.image) query.imageError = '1'
-      }
-      close()
-      router.push({ path: `/story/${result.id}`, query })
-    }
-  } catch (e) {
-    error.value = e instanceof Error ? e.message : '生成失败'
-  } finally {
+      router.push(`/story/${storyId}`)
+    }
+  } catch (e) {
+    error.value = e instanceof Error ? e.message : '生成失败'
+  } finally {
     loading.value = false
   }
 }
diff --git a/admin-frontend/src/components/GenerationTrace.vue b/admin-frontend/src/components/GenerationTrace.vue
index 1234bcb..7f85d90 100644
--- a/admin-frontend/src/components/GenerationTrace.vue
+++ b/admin-frontend/src/components/GenerationTrace.vue
@@ -116,6 +116,7 @@ function statusLabel(status?: string) {
 function eventLabel(eventType: string) {
   const labels: Record<string, string> = {
     request_accepted: '请求接收',
+    worker_started: '后台任务开始',
     context_prepared: '上下文准备',
     narrative_generated: '正文生成',
     story_saved: '故事保存',
diff --git a/admin-frontend/src/views/StoryDetail.vue b/admin-frontend/src/views/StoryDetail.vue
index 9931b8d..c376640 100644
--- a/admin-frontend/src/views/StoryDetail.vue
+++ b/admin-frontend/src/views/StoryDetail.vue
@@ -67,6 +67,7 @@ const audioDuration = ref(0)
 const error = ref('')
 const showDeleteConfirm = ref(false)
 const generationTraceRef = ref<InstanceType<typeof GenerationTrace> | null>(null)
+let refreshTimer: ReturnType<typeof setInterval> | null = null
 
 const storyParagraphs = computed(() => story.value?.story_text?.split('\n\n') ?? [])
 const generationMeta = computed(() => getGenerationStatusMeta(story.value?.generation_status))
@@ -75,6 +76,7 @@ const audioMeta = computed(() => getAssetStatusMeta(story.value?.audio_status))
 const canRetryImage = computed(() => story.value?.retryable_assets.includes('image') ?? false)
 const canRetryAudio = computed(() => story.value?.retryable_assets.includes('audio') ?? false)
 const isAudioGenerating = computed(() => story.value?.audio_status === 'generating')
+const shouldAutoRefreshStory = computed(() => story.value?.generation_status === 'assets_generating')
 const audioCacheLabel = computed(() => {
   if (!audioCacheStatus.value?.cache_exists) return '暂无缓存'
   const size = audioCacheStatus.value.cache_size_bytes ?? 0
@@ -109,6 +111,13 @@ async function refreshStorySnapshot() {
   await refreshAudioStatus().catch(() => undefined)
 }
 
+function stopAutoRefresh() {
+  if (refreshTimer) {
+    clearInterval(refreshTimer)
+    refreshTimer = null
+  }
+}
+
 async function fetchStory() {
   loading.value = true
   error.value = ''
@@ -283,7 +292,19 @@ watch(
   { immediate: true },
 )
 
+watch(shouldAutoRefreshStory, (enabled) => {
+  stopAutoRefresh()
+  if (enabled) {
+    refreshTimer = setInterval(() => {
+      if (!loading.value && !imageLoading.value && !audioLoading.value) {
+        void refreshStorySnapshot().catch(() => undefined)
+      }
+    }, 2500)
+  }
+})
+
 onUnmounted(() => {
+  stopAutoRefresh()
   if (audioUrl.value) {
     URL.revokeObjectURL(audioUrl.value)
   }
diff --git a/backend/app/api/stories.py b/backend/app/api/stories.py
index 434c866..06c4417 100644
--- a/backend/app/api/stories.py
+++ b/backend/app/api/stories.py
@@ -67,13 +67,13 @@ def _mark_legacy_generation_endpoint(response: Response, successor: str) -> None
     response.headers.update(_legacy_generation_headers(successor))
 
 
-@router.post("/generations", response_model=GenerationResponse)
+@router.post("/generations", response_model=GenerationResponse, status_code=202)
 async def create_generation(
     request: GenerationRequest,
     user: User = Depends(require_user),
     db: AsyncSession = Depends(get_db),
 ):
-    """Create a story or storybook through the unified generation workflow."""
+    """Accept one story/storybook generation request for background execution."""
     await check_rate_limit(f"story:{user.id}", RATE_LIMIT_REQUESTS, RATE_LIMIT_WINDOW)
     return await story_service.generate_generation_service(request, user.id, db)
 
diff --git a/backend/app/schemas/story_schemas.py b/backend/app/schemas/story_schemas.py
index 381f821..247bf38 100644
--- a/backend/app/schemas/story_schemas.py
+++ b/backend/app/schemas/story_schemas.py
@@ -117,9 +117,9 @@ class StorybookResponse(StoryStatusMixin):
 class GenerationResponse(StoryStatusMixin):
     """Unified generation response for the target workflow API."""
 
-    id: int
+    id: int | None = None
     generation_job_id: str | None = None
-    title: str
+    title: str | None = None
     mode: str
     story_text: str | None = None
     pages: list[StorybookPageResponse] | None = None
diff --git a/backend/app/services/generation_jobs.py b/backend/app/services/generation_jobs.py
index abb38b9..4f21833 100644
--- a/backend/app/services/generation_jobs.py
+++ b/backend/app/services/generation_jobs.py
@@ -6,7 +6,7 @@ from datetime import datetime, timedelta, timezone
 from typing import Any
 
 from fastapi import HTTPException
-from sqlalchemy import desc, select
+from sqlalchemy import desc, select, update
 from sqlalchemy.ext.asyncio import AsyncSession
 
 from app.core.config import settings
@@ -59,6 +59,7 @@ def _job_progress(job: GenerationJob) -> dict[str, Any]:
 
     progress_map: dict[str, tuple[int, str]] = {
         "request_accepted": (5, "已接收请求"),
+        "worker_started": (12, "后台任务已开始"),
         "context_prepared": (20, "上下文已准备"),
         "narrative_generated": (45, "正文已生成"),
         "story_saved": (60, "主记录已保存"),
@@ -66,8 +67,18 @@ def _job_progress(job: GenerationJob) -> dict[str, Any]:
         "provider_call_succeeded": (72, "Provider 调用成功"),
         "provider_call_failed": (72, "Provider 调用失败，尝试恢复"),
         "cover_image_started": (75, "封面生成中"),
+        "cover_image_succeeded": (88, "封面已生成"),
+        "cover_image_failed": (88, "封面生成失败"),
         "storybook_images_started": (75, "绘本插图生成中"),
+        "storybook_cover_image_succeeded": (82, "绘本封面已生成"),
+        "storybook_cover_image_failed": (82, "绘本封面生成失败"),
+        "storybook_page_image_succeeded": (86, "分页插图已生成"),
+        "storybook_page_image_failed": (86, "分页插图生成失败"),
+        "storybook_images_completed": (92, "绘本插图已完成"),
         "audio_started": (75, "音频生成中"),
+        "audio_cache_hit": (88, "音频缓存已复用"),
+        "audio_succeeded": (88, "音频已生成"),
+        "audio_failed": (88, "音频生成失败"),
         "asset_retry_started": (25, "资源重试中"),
         "postprocessing_queued": (90, "后处理已排队"),
         "asset_generation_completed": (100, "资源已完成"),
@@ -155,6 +166,10 @@ async def record_generation_event(
 ) -> GenerationJobEvent:
     """Append one event to an existing generation job."""
 
+    job.current_step = event_type
+    if story_id is not None:
+        job.story_id = story_id
+
     event = GenerationJobEvent(
         job_id=job.id,
         story_id=story_id if story_id is not None else job.story_id,
@@ -169,6 +184,42 @@ async def record_generation_event(
     return event
 
 
+async def claim_generation_job_for_worker(
+    db: AsyncSession,
+    *,
+    job_id: str,
+) -> GenerationJob | None:
+    """Claim one queued generation job for worker execution once."""
+
+    claim_result = await db.execute(
+        update(GenerationJob)
+        .where(
+            GenerationJob.id == job_id,
+            GenerationJob.status == "running",
+            GenerationJob.current_step == "request_accepted",
+        )
+        .values(current_step="worker_started")
+    )
+    await db.commit()
+
+    if not claim_result.rowcount:
+        return None
+
+    result = await db.execute(select(GenerationJob).where(GenerationJob.id == job_id))
+    job = result.scalar_one_or_none()
+    if job is None:
+        return None
+
+    await record_generation_event(
+        db,
+        job=job,
+        event_type="worker_started",
+        status="running",
+        message="Generation worker started processing the accepted request.",
+    )
+    return job
+
+
 async def finish_generation_job(
     db: AsyncSession,
     *,
diff --git a/backend/app/services/story_service.py b/backend/app/services/story_service.py
index 7ec8714..f467e5b 100644
--- a/backend/app/services/story_service.py
+++ b/backend/app/services/story_service.py
@@ -12,7 +12,7 @@ from sqlalchemy.orm import joinedload
 
 from app.core.config import settings
 from app.core.logging import get_logger
-from app.db.models import ChildProfile, Story, StoryUniverse
+from app.db.models import ChildProfile, GenerationJob, Story, StoryUniverse
 from app.schemas.story_schemas import (
     AchievementItem,
     FullStoryResponse,
@@ -33,6 +33,7 @@ from app.services.audio_storage import (
     write_story_audio_cache,
 )
 from app.services.generation_jobs import (
+    claim_generation_job_for_worker,
     create_generation_job,
     ensure_no_active_story_generation_job,
     finish_generation_job,
@@ -1113,7 +1114,7 @@ async def generate_generation_service(
     user_id: str,
     db: AsyncSession,
 ) -> GenerationResponse:
-    """Unified generation workflow entry point for stories and storybooks."""
+    """Queue one unified generation workflow for background execution."""
 
     job = await create_generation_job(
         db,
@@ -1124,7 +1125,65 @@ async def generate_generation_service(
     )
 
     try:
-        response = await _generate_generation_service_with_job(request, user_id, db, job=job)
+        from app.tasks.generation_workflow import run_generation_workflow_task
+
+        run_generation_workflow_task.delay(job.id)
+    except Exception as exc:
+        await finish_generation_job(
+            db,
+            job=job,
+            story=None,
+            status="failed",
+            current_step="generation_failed",
+            error_message="Background generation dispatch failed.",
+            message="Generation failed before the worker could start processing the job.",
+            metadata={"dispatch_error": str(exc)},
+        )
+        raise HTTPException(
+            status_code=503,
+            detail="后台生成任务派发失败，请确认 worker 可用后重试。",
+        ) from exc
+
+    return _build_queued_generation_response(request, job_id=job.id)
+
+
+def _build_queued_generation_response(
+    request: GenerationRequest,
+    *,
+    job_id: str,
+) -> GenerationResponse:
+    """Build the immediate API response after a generation job is accepted."""
+
+    return GenerationResponse(
+        id=None,
+        generation_job_id=job_id,
+        title="生成任务已提交",
+        mode="storybook" if request.output_mode == "storybook" else "generated",
+        generation_status="queued",
+        text_status="generating",
+        image_status="not_requested",
+        audio_status="not_requested",
+        last_error=None,
+        retryable_assets=[],
+        child_profile_id=request.child_profile_id,
+        universe_id=request.universe_id,
+    )
+
+
+async def execute_generation_job_service(
+    job: GenerationJob,
+    db: AsyncSession,
+) -> GenerationResponse:
+    """Execute one previously accepted generation job inside the worker."""
+
+    try:
+        request = GenerationRequest.model_validate(job.request_payload or {})
+        response = await _generate_generation_service_with_job(
+            request,
+            job.user_id,
+            db,
+            job=job,
+        )
     except HTTPException as exc:
         await finish_generation_job(
             db,
@@ -1151,6 +1210,21 @@ async def generate_generation_service(
     return response
 
 
+async def run_generation_job_service(
+    job_id: str,
+    db: AsyncSession,
+) -> GenerationJob | None:
+    """Claim and execute one generation job from the background queue."""
+
+    job = await claim_generation_job_for_worker(db, job_id=job_id)
+    if job is None:
+        logger.info("generation_job_execution_skipped", job_id=job_id)
+        return None
+
+    await execute_generation_job_service(job, db)
+    return job
+
+
 async def _generate_generation_service_with_job(
     request: GenerationRequest,
     user_id: str,
diff --git a/backend/app/tasks/generation_workflow.py b/backend/app/tasks/generation_workflow.py
new file mode 100644
index 0000000..38105c6
--- /dev/null
+++ b/backend/app/tasks/generation_workflow.py
@@ -0,0 +1,38 @@
+"""Background execution for unified generation workflows."""
+
+import asyncio
+
+from app.core.celery_app import celery_app
+from app.core.logging import get_logger
+from app.db.database import _get_session_factory
+from app.services.story_service import run_generation_job_service
+
+logger = get_logger(__name__)
+
+
+@celery_app.task
+def run_generation_workflow_task(job_id: str):
+    """Execute one accepted generation job in the Celery worker."""
+
+    logger.info("generation_workflow_task_started", job_id=job_id)
+
+    async def _run():
+        session_factory = _get_session_factory()
+        async with session_factory() as session:
+            return await run_generation_job_service(job_id, session)
+
+    try:
+        result = asyncio.run(_run())
+        logger.info(
+            "generation_workflow_task_completed",
+            job_id=job_id,
+            executed=bool(result),
+        )
+        return {"job_id": job_id, "executed": bool(result)}
+    except Exception as exc:
+        logger.error(
+            "generation_workflow_task_failed",
+            job_id=job_id,
+            error=str(exc),
+        )
+        raise
diff --git a/backend/tests/test_generation_jobs.py b/backend/tests/test_generation_jobs.py
index 89a1d16..6a269bb 100644
--- a/backend/tests/test_generation_jobs.py
+++ b/backend/tests/test_generation_jobs.py
@@ -18,6 +18,7 @@ from app.services.generation_jobs import (
     mark_stale_generation_jobs,
     record_generation_event,
 )
+from app.services.story_service import run_generation_job_service
 
 pytestmark = pytest.mark.asyncio
 
@@ -45,7 +46,7 @@ def build_storybook_output() -> Storybook:
     )
 
 
-async def test_unified_generation_records_job_events_and_retryable_assets(
+async def test_unified_generation_is_queued_then_worker_persists_story_and_events(
     db_session,
     test_user,
     auth_token,
@@ -56,82 +57,108 @@ async def test_unified_generation_records_job_events_and_retryable_assets(
 
     app.dependency_overrides[get_db] = override_get_db
     transport = ASGITransport(app=app)
+    task_delay_path = (
+        "app.tasks.generation_workflow.run_generation_workflow_task.delay"
+    )
 
     try:
-        async with AsyncClient(transport=transport, base_url="http://test") as client:
-            client.cookies.set("access_token", auth_token)
+        with patch(task_delay_path) as mock_delay:
+            async with AsyncClient(transport=transport, base_url="http://test") as client:
+                client.cookies.set("access_token", auth_token)
 
-            response = await client.post(
-                "/api/generations",
-                json={
-                    "output_mode": "story",
-                    "type": "keywords",
-                    "data": "小兔子, 森林",
-                    "generate_images": False,
-                },
-            )
-
-            assert response.status_code == 200
-            data = response.json()
-            assert data["generation_status"] == "partial_ready"
-            assert data["retryable_assets"] == ["image", "audio"]
-            assert data["generation_job_id"]
-
-            jobs = (
-                await db_session.execute(
-                    select(GenerationJob).where(GenerationJob.user_id == test_user.id)
+                response = await client.post(
+                    "/api/generations",
+                    json={
+                        "output_mode": "story",
+                        "type": "keywords",
+                        "data": "小兔子, 森林",
+                        "generate_images": False,
+                    },
                 )
-            ).scalars().all()
-            assert len(jobs) == 1
-            job = jobs[0]
-            assert job.story_id == data["id"]
-            assert job.output_mode == "story"
-            assert job.input_type == "keywords"
-            assert job.status == "completed"
-            assert job.current_step == "generation_completed"
-            assert job.result_snapshot["retryable_assets"] == ["image", "audio"]
-            assert data["generation_job_id"] == job.id
 
-            events = (
-                await db_session.execute(
-                    select(GenerationJobEvent)
-                    .where(GenerationJobEvent.job_id == job.id)
-                    .order_by(GenerationJobEvent.id)
-                )
-            ).scalars().all()
-            assert [event.event_type for event in events] == [
-                "request_accepted",
-                "context_prepared",
-                "narrative_generated",
-                "story_saved",
-                "generation_completed",
-            ]
-            assert events[1].event_metadata["has_memory_context"] is False
-            assert events[2].event_metadata["title"] == "小兔子的冒险"
-            assert events[3].story_id == data["id"]
+                assert response.status_code == 202
+                data = response.json()
+                assert data["id"] is None
+                assert data["generation_status"] == "queued"
+                assert data["text_status"] == "generating"
+                assert data["retryable_assets"] == []
+                assert data["generation_job_id"]
+                mock_delay.assert_called_once_with(data["generation_job_id"])
 
-            detail_response = await client.get(f"/api/generations/jobs/{job.id}")
-            assert detail_response.status_code == 200
-            detail = detail_response.json()
-            assert detail["id"] == job.id
-            assert detail["story_id"] == data["id"]
-            assert detail["progress_percent"] == 100
-            assert detail["progress_label"] == "已完成"
-            assert detail["is_terminal"] is True
-            assert [event["event_type"] for event in detail["events"]] == [
-                "request_accepted",
-                "context_prepared",
-                "narrative_generated",
-                "story_saved",
-                "generation_completed",
-            ]
+                jobs = (
+                    await db_session.execute(
+                        select(GenerationJob).where(GenerationJob.user_id == test_user.id)
+                    )
+                ).scalars().all()
+                assert len(jobs) == 1
+                job = jobs[0]
+                assert job.story_id is None
+                assert job.output_mode == "story"
+                assert job.input_type == "keywords"
+                assert job.status == "running"
+                assert job.current_step == "request_accepted"
+                assert data["generation_job_id"] == job.id
 
-            list_response = await client.get(f"/api/generations/{data['id']}/jobs")
-            assert list_response.status_code == 200
-            job_list = list_response.json()
-            assert [item["id"] for item in job_list] == [job.id]
-            assert job_list[0]["progress_percent"] == 100
-            assert job_list[0]["is_terminal"] is True
+                await run_generation_job_service(job.id, db_session)
+
+                job = (
+                    await db_session.execute(
+                        select(GenerationJob).where(GenerationJob.id == job.id)
+                    )
+                ).scalar_one()
+                assert job.story_id is not None
+                assert job.status == "completed"
+                assert job.current_step == "generation_completed"
+                assert job.result_snapshot["retryable_assets"] == ["image", "audio"]
+
+                events = (
+                    await db_session.execute(
+                        select(GenerationJobEvent)
+                        .where(GenerationJobEvent.job_id == job.id)
+                        .order_by(GenerationJobEvent.id)
+                    )
+                ).scalars().all()
+                assert [event.event_type for event in events] == [
+                    "request_accepted",
+                    "worker_started",
+                    "context_prepared",
+                    "narrative_generated",
+                    "story_saved",
+                    "generation_completed",
+                ]
+                assert events[2].event_metadata["has_memory_context"] is False
+                assert events[3].event_metadata["title"] == "小兔子的冒险"
+                assert events[4].story_id == job.story_id
+
+                detail_response = await client.get(f"/api/generations/jobs/{job.id}")
+                assert detail_response.status_code == 200
+                detail = detail_response.json()
+                assert detail["id"] == job.id
+                assert detail["story_id"] == job.story_id
+                assert detail["progress_percent"] == 100
+                assert detail["progress_label"] == "已完成"
+                assert detail["is_terminal"] is True
+                assert [event["event_type"] for event in detail["events"]] == [
+                    "request_accepted",
+                    "worker_started",
+                    "context_prepared",
+                    "narrative_generated",
+                    "story_saved",
+                    "generation_completed",
+                ]
+
+                story_response = await client.get(f"/api/generations/{job.story_id}")
+                assert story_response.status_code == 200
+                story_data = story_response.json()
+                assert story_data["generation_status"] == "partial_ready"
+                assert story_data["retryable_assets"] == ["image", "audio"]
+
+                list_response = await client.get(f"/api/generations/{job.story_id}/jobs")
+                assert list_response.status_code == 200
+                job_list = list_response.json()
+                assert [item["id"] for item in job_list] == [job.id]
+                assert job_list[0]["progress_percent"] == 100
+                assert job_list[0]["is_terminal"] is True
     finally:
         app.dependency_overrides.clear()
 
@@ -196,7 +223,7 @@ async def test_asset_retry_records_job_events_and_updates_retryable_assets(
         app.dependency_overrides.clear()
 
 
-async def test_storybook_generation_records_page_image_events(
+async def test_storybook_generation_is_queued_then_worker_records_page_image_events(
     db_session,
     auth_token,
 ):
@@ -222,61 +249,78 @@ async def test_storybook_generation_records_page_image_events(
                     "https://example.com/storybook-page-2.png",
                 ]
 
-                async with AsyncClient(transport=transport, base_url="http://test") as client:
-                    client.cookies.set("access_token", auth_token)
-
-                    response = await client.post(
-                        "/api/generations",
-                        json={
-                            "output_mode": "storybook",
-                            "type": "keywords",
-                            "data": "森林, 发光, 友情",
-                            "page_count": 6,
-                            "generate_images": True,
-                        },
-                    )
-
-        assert response.status_code == 200
-        data = response.json()
-        assert data["mode"] == "storybook"
-        assert data["image_status"] == "ready"
-
-        job = (
-            await db_session.execute(
-                select(GenerationJob).where(
-                    GenerationJob.story_id == data["id"],
-                    GenerationJob.output_mode == "storybook",
+                task_delay_path = (
+                    "app.tasks.generation_workflow.run_generation_workflow_task.delay"
                 )
-            )
-        ).scalar_one()
+                with patch(task_delay_path) as mock_delay:
+                    async with AsyncClient(transport=transport, base_url="http://test") as client:
+                        client.cookies.set("access_token", auth_token)
 
-        events = (
-            await db_session.execute(
-                select(GenerationJobEvent)
-                .where(GenerationJobEvent.job_id == job.id)
-                .order_by(GenerationJobEvent.id)
-            )
-        ).scalars().all()
+                        response = await client.post(
+                            "/api/generations",
+                            json={
+                                "output_mode": "storybook",
+                                "type": "keywords",
+                                "data": "森林, 发光, 友情",
+                                "page_count": 6,
+                                "generate_images": True,
+                            },
+                        )
 
-        assert [event.event_type for event in events] == [
-            "request_accepted",
-            "context_prepared",
-            "narrative_generated",
-            "storybook_images_started",
-            "storybook_cover_image_succeeded",
-            "storybook_page_image_succeeded",
-            "storybook_page_image_succeeded",
-            "storybook_images_completed",
-            "story_saved",
-            "generation_completed",
-        ]
-        page_events = [
-            event
-            for event in events
-            if event.event_type == "storybook_page_image_succeeded"
-        ]
-        assert [event.event_metadata["page_number"] for event in page_events] == [1, 2]
-        assert events[7].event_metadata["completed_pages"] == [1, 2]
+                    assert response.status_code == 202
+                    data = response.json()
+                    assert data["id"] is None
+                    assert data["mode"] == "storybook"
+                    assert data["generation_status"] == "queued"
+                    assert data["text_status"] == "generating"
+                    mock_delay.assert_called_once_with(data["generation_job_id"])
+                    job = (
+                        await db_session.execute(
+                            select(GenerationJob).where(
+                                GenerationJob.id == data["generation_job_id"],
+                            )
+                        )
+                    ).scalar_one()
+                    await run_generation_job_service(job.id, db_session)
+
+                    job = (
+                        await db_session.execute(
+                            select(GenerationJob).where(
+                                GenerationJob.id == job.id,
+                            )
+                        )
+                    ).scalar_one()
+                    assert job.story_id is not None
+                    assert job.status == "completed"
+
+                    events = (
+                        await db_session.execute(
+                            select(GenerationJobEvent)
+                            .where(GenerationJobEvent.job_id == job.id)
+                            .order_by(GenerationJobEvent.id)
+                        )
+                    ).scalars().all()
+
+                    assert [event.event_type for event in events] == [
+                        "request_accepted",
+                        "worker_started",
+                        "context_prepared",
+                        "narrative_generated",
+                        "storybook_images_started",
+                        "storybook_cover_image_succeeded",
+                        "storybook_page_image_succeeded",
+                        "storybook_page_image_succeeded",
+                        "storybook_images_completed",
+                        "story_saved",
+                        "generation_completed",
+                    ]
+                    page_events = [
+                        event
+                        for event in events
+                        if event.event_type == "storybook_page_image_succeeded"
+                    ]
+                    assert [event.event_metadata["page_number"] for event in page_events] == [1, 2]
+                    assert events[8].event_metadata["completed_pages"] == [1, 2]
     finally:
         app.dependency_overrides.clear()
 
diff --git a/backend/tests/test_stories.py b/backend/tests/test_stories.py
index 451e0f0..133285b 100644
--- a/backend/tests/test_stories.py
+++ b/backend/tests/test_stories.py
@@ -430,6 +430,8 @@ class TestGenerateFull:
 class TestUnifiedGenerations:
     """Tests for the target unified generation API."""
 
+    TASK_DELAY_PATH = "app.tasks.generation_workflow.run_generation_workflow_task.delay"
+
     def test_create_generation_without_auth(self, client: TestClient):
         response = client.post(
             "/api/generations",
@@ -443,60 +445,64 @@ class TestUnifiedGenerations:
         mock_text_provider,
         mock_image_provider,
     ):
-        response = auth_client.post(
-            "/api/generations",
-            json={
-                "output_mode": "story",
-                "type": "keywords",
-                "data": "小兔子, 森林, 勇气",
-                "generate_images": True,
-            },
-        )
+        with patch(self.TASK_DELAY_PATH) as mock_delay:
+            response = auth_client.post(
+                "/api/generations",
+                json={
+                    "output_mode": "story",
+                    "type": "keywords",
+                    "data": "小兔子, 森林, 勇气",
+                    "generate_images": True,
+                },
+            )
 
-        assert response.status_code == 200
+        assert response.status_code == 202
         assert "Deprecation" not in response.headers
         data = response.json()
-        assert data["id"] is not None
+        assert data["id"] is None
         assert data["mode"] == "generated"
-        assert data["story_text"] == "从前有一只小兔子。"
-        assert data["image_url"] == "https://example.com/image.png"
-        assert data["cover_url"] == "https://example.com/image.png"
+        assert data["story_text"] is None
+        assert data["image_url"] is None
+        assert data["cover_url"] is None
         assert data["pages"] is None
-        assert data["generation_status"] == "partial_ready"
-        assert data["image_status"] == "ready"
+        assert data["generation_status"] == "queued"
+        assert data["text_status"] == "generating"
+        assert data["image_status"] == "not_requested"
         assert data["audio_status"] == "not_requested"
         assert data["errors"] == {}
+        mock_delay.assert_called_once_with(data["generation_job_id"])
 
     def test_create_story_generation_without_assets(
         self,
         auth_client: TestClient,
         mock_text_provider,
     ):
-        response = auth_client.post(
-            "/api/generations",
-            json={
-                "output_mode": "story",
-                "type": "keywords",
-                "data": "小兔子, 森林",
-                "generate_images": False,
-            },
-        )
+        with patch(self.TASK_DELAY_PATH) as mock_delay:
+            response = auth_client.post(
+                "/api/generations",
+                json={
+                    "output_mode": "story",
+                    "type": "keywords",
+                    "data": "小兔子, 森林",
+                    "generate_images": False,
+                },
+            )
 
-        assert response.status_code == 200
+        assert response.status_code == 202
         data = response.json()
         assert data["mode"] == "generated"
         assert data["image_url"] is None
-        assert data["generation_status"] == "partial_ready"
+        assert data["generation_status"] == "queued"
+        assert data["text_status"] == "generating"
         assert data["image_status"] == "not_requested"
+        mock_delay.assert_called_once_with(data["generation_job_id"])
 
     def test_create_story_generation_image_failure(
         self,
         auth_client: TestClient,
         mock_text_provider,
     ):
-        with patch("app.services.story_service.generate_image", new_callable=AsyncMock) as mock_img:
-            mock_img.side_effect = Exception("Image API error")
-
+        with patch(self.TASK_DELAY_PATH) as mock_delay:
             response = auth_client.post(
                 "/api/generations",
                 json={
@@ -507,55 +513,45 @@ class TestUnifiedGenerations:
                 },
             )
 
-        assert response.status_code == 200
+        assert response.status_code == 202
         data = response.json()
         assert data["image_url"] is None
-        assert data["generation_status"] == "degraded_completed"
-        assert data["image_status"] == "failed"
+        assert data["generation_status"] == "queued"
+        assert data["text_status"] == "generating"
+        assert data["image_status"] == "not_requested"
         assert data["audio_status"] == "not_requested"
-        assert "Image API error" in data["errors"]["image"]
-        assert "Image API error" in data["last_error"]
+        assert data["errors"] == {}
+        assert data["last_error"] is None
+        mock_delay.assert_called_once_with(data["generation_job_id"])
 
     def test_create_storybook_generation_success(self, auth_client: TestClient):
-        with patch(
-            "app.services.story_service.generate_storybook",
-            new_callable=AsyncMock,
-        ) as mock_storybook:
-            with patch(
-                "app.services.story_service.generate_image",
-                new_callable=AsyncMock,
-            ) as mock_image:
-                mock_storybook.return_value = build_storybook_output()
-                mock_image.side_effect = [
-                    "https://example.com/storybook-cover.png",
-                    "https://example.com/storybook-page-1.png",
-                    "https://example.com/storybook-page-2.png",
-                ]
+        with patch(self.TASK_DELAY_PATH) as mock_delay:
+            response = auth_client.post(
+                "/api/generations",
+                json={
+                    "output_mode": "storybook",
+                    "type": "keywords",
+                    "data": "森林, 发光, 友情",
+                    "page_count": 6,
+                    "generate_images": True,
+                },
+            )
 
-                response = auth_client.post(
-                    "/api/generations",
-                    json={
-                        "output_mode": "storybook",
-                        "type": "keywords",
-                        "data": "森林, 发光, 友情",
-                        "page_count": 6,
-                        "generate_images": True,
-                    },
-                )
-
-        assert response.status_code == 200
+        assert response.status_code == 202
         data = response.json()
-        assert data["id"] is not None
+        assert data["id"] is None
         assert data["mode"] == "storybook"
         assert data["story_text"] is None
-        assert len(data["pages"]) == 2
-        assert data["cover_url"] == "https://example.com/storybook-cover.png"
-        assert data["image_url"] == "https://example.com/storybook-cover.png"
-        assert data["main_character"] == "小兔子露露"
-        assert data["art_style"] == "温暖水彩"
-        assert data["generation_status"] == "completed"
-        assert data["image_status"] == "ready"
+        assert data["pages"] is None
+        assert data["cover_url"] is None
+        assert data["image_url"] is None
+        assert data["main_character"] is None
+        assert data["art_style"] is None
+        assert data["generation_status"] == "queued"
+        assert data["text_status"] == "generating"
+        assert data["image_status"] == "not_requested"
         assert data["audio_status"] == "not_requested"
+        mock_delay.assert_called_once_with(data["generation_job_id"])
 
     def test_get_generation_alias(self, auth_client: TestClient, test_story):
         response = auth_client.get(f"/api/generations/{test_story.id}")
diff --git a/frontend/src/components/CreateStoryModal.vue b/frontend/src/components/CreateStoryModal.vue
index 3a784f3..2aa6bc1 100644
--- a/frontend/src/components/CreateStoryModal.vue
+++ b/frontend/src/components/CreateStoryModal.vue
@@ -1,11 +1,10 @@
 <script setup lang="ts">
 import { ref, computed, watch } from 'vue'
-import type { Component } from 'vue'
-import { useRouter } from 'vue-router'
-import { useI18n } from 'vue-i18n'
-import { useUserStore } from '../stores/user'
-import { useStorybookStore } from '../stores/storybook'
-import { api } from '../api/client'
+import type { Component } from 'vue'
+import { useRouter } from 'vue-router'
+import { useI18n } from 'vue-i18n'
+import { useUserStore } from '../stores/user'
+import { api } from '../api/client'
 import BaseButton from './ui/BaseButton.vue'
 import BaseInput from './ui/BaseInput.vue'
 import BaseSelect from './ui/BaseSelect.vue'
@@ -34,10 +33,9 @@ const emit = defineEmits<{
   'update:modelValue': [value: boolean]
 }>()
 
-const { t } = useI18n()
-const router = useRouter()
-const userStore = useUserStore()
-const storybookStore = useStorybookStore()
+const { t } = useI18n()
+const router = useRouter()
+const userStore = useUserStore()
 
 // State
 const inputType = ref<'keywords' | 'full_story'>('keywords')
@@ -88,28 +86,63 @@ const generationTitle = computed(() =>
 const generationSteps = computed(() => {
   if (requestedOutputMode.value === 'storybook') {
     return [
-      '正在整理主题和成长目标...',
-      '生成绘本分镜和每页文字...',
-      '保存绘本主记录，确保刷新也能找回...',
-      '补全封面和分页插图...',
-      '马上进入可翻页阅读模式。',
+      '正在提交后台任务...',
+      'Worker 会生成绘本分镜和每页文字...',
+      '主记录一落库就能通过 ID 找回...',
+      '插图会继续在后台补全...',
+      '稍后自动进入可翻页阅读模式。',
     ]
   }
 
   return [
-    '正在整理孩子档案和故事主题...',
-    '生成可先阅读的故事正文...',
-    '保存故事主记录，避免结果丢失...',
-    '补全封面图，失败也可稍后重试...',
+    '正在提交后台任务...',
+    'Worker 会生成故事正文并保存主记录...',
+    '主内容一可读就会自动跳转详情页...',
+    '封面会继续在后台补全，失败也能重试...',
     '马上进入故事详情页。',
   ]
 })
+
+interface GenerationAcceptedResponse {
+  id: number | null
+  generation_job_id: string | null
+}
+
+interface GenerationJobDetail {
+  story_id: number | null
+  is_terminal: boolean
+  error_message: string | null
+}
+
+const JOB_POLL_INTERVAL_MS = 1500
+const JOB_POLL_MAX_ATTEMPTS = 80
 
 // Methods
-function close() {
-  emit('update:modelValue', false)
-  error.value = ''
-}
+function close() {
+  emit('update:modelValue', false)
+  error.value = ''
+}
+
+function sleep(ms: number) {
+  return new Promise((resolve) => {
+    window.setTimeout(resolve, ms)
+  })
+}
+
+async function waitForStoryId(jobId: string) {
+  for (let attempt = 0; attempt < JOB_POLL_MAX_ATTEMPTS; attempt += 1) {
+    const detail = await api.get<GenerationJobDetail>(`/api/generations/jobs/${jobId}`)
+    if (detail.story_id) {
+      return detail.story_id
+    }
+    if (detail.is_terminal) {
+      throw new Error(detail.error_message || '生成失败，请稍后重试')
+    }
+    await sleep(JOB_POLL_INTERVAL_MS)
+  }
+
+  throw new Error('任务已提交，但主内容落库超时，请稍后到故事库查看最新结果')
+}
 
 async function fetchProfiles() {
   if (!userStore.user) return
@@ -173,21 +206,22 @@ async function generateStory() {
     if (selectedProfileId.value) payload.child_profile_id = selectedProfileId.value
     if (selectedUniverseId.value) payload.universe_id = selectedUniverseId.value
 
-    if (requestedOutputMode.value === 'storybook') {
-      const response = await api.post<any>('/api/generations', payload)
+    const accepted = await api.post<GenerationAcceptedResponse>('/api/generations', payload)
+    const jobId = accepted.generation_job_id
+    if (!jobId) {
+      throw new Error('生成任务已创建，但缺少任务编号')
+    }
 
-      storybookStore.setStorybook(response)
-      close()
-      const storybookPath = response.id ? `/storybook/view/${response.id}` : '/storybook/view'
-      router.push(storybookPath)
+    const storyId = accepted.id ?? await waitForStoryId(jobId)
+    close()
+    if (requestedOutputMode.value === 'storybook') {
+      router.push(`/storybook/view/${storyId}`)
     } else {
-      const result = await api.post<any>('/api/generations', payload)
-      close()
-      router.push(`/story/${result.id}`)
+      router.push(`/story/${storyId}`)
     }
   } catch (e) {
-    error.value = e instanceof Error ? e.message : '生成失败'
-  } finally {
+    error.value = e instanceof Error ? e.message : '生成失败'
+  } finally {
     loading.value = false
   }
 }
diff --git a/frontend/src/components/GenerationTrace.vue b/frontend/src/components/GenerationTrace.vue
index a4e87c9..368f8c5 100644
--- a/frontend/src/components/GenerationTrace.vue
+++ b/frontend/src/components/GenerationTrace.vue
@@ -101,6 +101,7 @@ function getJobStatusLabel(status?: string) {
 function getEventLabel(eventType: string) {
   const labels: Record<string, string> = {
     request_accepted: '请求接收',
+    worker_started: '后台任务开始',
     context_prepared: '上下文准备',
     narrative_generated: '正文生成',
     story_saved: '故事保存',
diff --git a/frontend/src/views/StoryDetail.vue b/frontend/src/views/StoryDetail.vue
index 310f1d9..47bd193 100644
--- a/frontend/src/views/StoryDetail.vue
+++ b/frontend/src/views/StoryDetail.vue
@@ -67,6 +67,7 @@ const audioDuration = ref(0)
 const error = ref('')
 const showDeleteConfirm = ref(false)
 const generationTraceRef = ref<InstanceType<typeof GenerationTrace> | null>(null)
+let refreshTimer: ReturnType<typeof setInterval> | null = null
 
 const storyParagraphs = computed(() => story.value?.story_text?.split('\n\n') ?? [])
 const generationMeta = computed(() => getGenerationStatusMeta(story.value?.generation_status))
@@ -75,6 +76,7 @@ const audioMeta = computed(() => getAssetStatusMeta(story.value?.audio_status))
 const canRetryImage = computed(() => story.value?.retryable_assets.includes('image') ?? false)
 const canRetryAudio = computed(() => story.value?.retryable_assets.includes('audio') ?? false)
 const isAudioGenerating = computed(() => story.value?.audio_status === 'generating')
+const shouldAutoRefreshStory = computed(() => story.value?.generation_status === 'assets_generating')
 const audioCacheLabel = computed(() => {
   if (!audioCacheStatus.value?.cache_exists) return '暂无缓存'
   const size = audioCacheStatus.value.cache_size_bytes ?? 0
@@ -109,6 +111,13 @@ async function refreshStorySnapshot() {
   await refreshAudioStatus().catch(() => undefined)
 }
 
+function stopAutoRefresh() {
+  if (refreshTimer) {
+    clearInterval(refreshTimer)
+    refreshTimer = null
+  }
+}
+
 async function fetchStory() {
   loading.value = true
   error.value = ''
@@ -283,7 +292,19 @@ watch(
   { immediate: true },
 )
 
+watch(shouldAutoRefreshStory, (enabled) => {
+  stopAutoRefresh()
+  if (enabled) {
+    refreshTimer = setInterval(() => {
+      if (!loading.value && !imageLoading.value && !audioLoading.value) {
+        void refreshStorySnapshot().catch(() => undefined)
+      }
+    }, 2500)
+  }
+})
+
 onUnmounted(() => {
+  stopAutoRefresh()
   if (audioUrl.value) {
     URL.revokeObjectURL(audioUrl.value)
   }
diff --git a/scripts/demo_smoke.sh b/scripts/demo_smoke.sh
index 30b9c36..e7d983c 100755
--- a/scripts/demo_smoke.sh
+++ b/scripts/demo_smoke.sh
@@ -49,6 +49,32 @@ assert_jq() {
   fi
 }
 
+wait_for_job_story() {
+  local job_id="$1"
+  local attempts="${2:-60}"
+
+  for ((i=1; i<=attempts; i++)); do
+    local job_json
+    job_json="$(get_json "$APP_URL/api/generations/jobs/$job_id")"
+
+    if jq -e '.story_id != null' >/dev/null <<<"$job_json"; then
+      printf '%s\n' "$job_json"
+      return 0
+    fi
+
+    if jq -e '.is_terminal == true and .story_id == null' >/dev/null <<<"$job_json"; then
+      echo "Generation job finished without a persisted story: $job_id" >&2
+      echo "$job_json" | jq '.' >&2
+      exit 1
+    fi
+
+    sleep 1
+  done
+
+  echo "Timed out waiting for generation job to persist a story: $job_id" >&2
+  exit 1
+}
+
 say "Checking backend health"
 curl -fsS "$BACKEND_URL/health" | jq -e '.status == "ok"' >/dev/null
 curl -fsS "$ADMIN_BACKEND_URL/health" | jq -e '.status == "ok"' >/dev/null
@@ -70,18 +96,23 @@ story_json="$(post_json "$APP_URL/api/generations" '{
   "education_theme": "复盘与成长",
   "generate_images": false
 }')"
-story_id="$(jq -r '.id' <<<"$story_json")"
 story_job_id="$(jq -r '.generation_job_id' <<<"$story_json")"
-assert_jq "$story_json" '.mode == "generated" and .generation_status == "partial_ready" and .text_status == "ready"' "story should be readable before assets"
+assert_jq "$story_json" '.mode == "generated" and .generation_status == "queued" and .text_status == "generating"' "story request should be accepted for background execution"
 assert_jq "$story_json" '.generation_job_id != null and .generation_job_id != ""' "story generation should expose a job id"
+echo "$story_json" | jq '{generation_job_id,mode,generation_status,text_status}'
+
+say "Waiting for story main record to be persisted"
+story_job_json="$(wait_for_job_story "$story_job_id")"
+story_id="$(jq -r '.story_id' <<<"$story_job_json")"
+story_json="$(get_json "$APP_URL/api/generations/$story_id")"
+assert_jq "$story_json" '.mode == "generated" and .generation_status == "partial_ready" and .text_status == "ready"' "story should be readable before assets"
 assert_jq "$story_json" '(.retryable_assets | index("image")) != null and (.retryable_assets | index("audio")) != null' "story should expose image/audio as retryable assets"
 echo "$story_json" | jq '{id,title,mode,generation_status,image_status,audio_status,retryable_assets}'
 
 say "Checking story generation job events"
-story_job_json="$(get_json "$APP_URL/api/generations/jobs/$story_job_id")"
 assert_jq "$story_job_json" '.id == "'"$story_job_id"'" and .story_id == '"$story_id"'' "story generation job should be queryable"
 assert_jq "$story_job_json" '.progress_percent == 100 and .is_terminal == true' "story generation job should expose progress summary"
-assert_jq "$story_job_json" '([.events[].event_type] | index("context_prepared")) != null and ([.events[].event_type] | index("narrative_generated")) != null and ([.events[].event_type] | index("story_saved")) != null' "story generation job should include workflow events"
+assert_jq "$story_job_json" '([.events[].event_type] | index("worker_started")) != null and ([.events[].event_type] | index("context_prepared")) != null and ([.events[].event_type] | index("narrative_generated")) != null and ([.events[].event_type] | index("story_saved")) != null' "story generation job should include workflow events"
 assert_jq "$story_job_json" '([.events[].event_type] | index("provider_call_succeeded")) != null' "story generation job should include provider call events"
 echo "$story_job_json" | jq '{id,status,current_step,events:([.events[].event_type] | unique)}'
 
@@ -127,18 +158,23 @@ storybook_json="$(post_json "$APP_URL/api/generations" '{
   "generate_images": false,
   "page_count": 6
 }')"
-storybook_id="$(jq -r '.id' <<<"$storybook_json")"
 storybook_job_id="$(jq -r '.generation_job_id' <<<"$storybook_json")"
-assert_jq "$storybook_json" '.mode == "storybook" and .generation_status == "partial_ready" and .text_status == "ready" and .image_status == "not_requested" and (.pages | length) >= 4' "storybook should be readable before images"
+assert_jq "$storybook_json" '.mode == "storybook" and .generation_status == "queued" and .text_status == "generating"' "storybook request should be accepted for background execution"
 assert_jq "$storybook_json" '.generation_job_id != null and .generation_job_id != ""' "storybook generation should expose a job id"
+echo "$storybook_json" | jq '{generation_job_id,mode,generation_status,text_status}'
+
+say "Waiting for storybook main record to be persisted"
+storybook_job_json="$(wait_for_job_story "$storybook_job_id")"
+storybook_id="$(jq -r '.story_id' <<<"$storybook_job_json")"
+storybook_json="$(get_json "$APP_URL/api/generations/$storybook_id")"
+assert_jq "$storybook_json" '.mode == "storybook" and .generation_status == "partial_ready" and .text_status == "ready" and .image_status == "not_requested" and (.pages | length) >= 4' "storybook should be readable before images"
 assert_jq "$storybook_json" '(.retryable_assets | index("image")) != null and (.retryable_assets | index("audio")) == null' "storybook should expose images as retryable assets"
 echo "$storybook_json" | jq '{id,title,mode,generation_status,image_status,audio_status,retryable_assets,pages:(.pages | length)}'
 
 say "Checking storybook generation job events"
-storybook_job_json="$(get_json "$APP_URL/api/generations/jobs/$storybook_job_id")"
 assert_jq "$storybook_job_json" '.id == "'"$storybook_job_id"'" and .story_id == '"$storybook_id"'' "storybook generation job should be queryable"
 assert_jq "$storybook_job_json" '.progress_percent == 100 and .is_terminal == true' "storybook generation job should expose progress summary"
-assert_jq "$storybook_job_json" '([.events[].event_type] | index("context_prepared")) != null and ([.events[].event_type] | index("narrative_generated")) != null and ([.events[].event_type] | index("story_saved")) != null' "storybook generation job should include workflow events"
+assert_jq "$storybook_job_json" '([.events[].event_type] | index("worker_started")) != null and ([.events[].event_type] | index("context_prepared")) != null and ([.events[].event_type] | index("narrative_generated")) != null and ([.events[].event_type] | index("story_saved")) != null' "storybook generation job should include workflow events"
 echo "$storybook_job_json" | jq '{id,status,current_step,events:([.events[].event_type] | unique)}'
 
 say "Checking storybook provider stats"