diff --git a/app/api/plans.py b/app/api/plans.py
index d9cb440..c918a75 100644
--- a/app/api/plans.py
+++ b/app/api/plans.py
@@ -4,18 +4,33 @@ from fastapi import APIRouter, Depends, HTTPException, Response
 from common.db import fetchone
 from common.deps import verify_api_key
 from integrations.llm.schemas.plan import PlanOutput
+from models.plan import PlanApiResponse
 
 router = APIRouter(prefix="/api/plans", tags=["plans"], dependencies=[Depends(verify_api_key)])
 logger = logging.getLogger(__name__)
 
 
-@router.get("/{run_id}", response_model=PlanOutput | None)
+@router.get("/{run_id}", response_model=PlanApiResponse)
 async def get_plan(run_id: str):
     logger.info("GET /api/plans/%s", run_id)
-    row = await fetchone("SELECT plan_data FROM analysis_runs WHERE analysis_run_id = %s", (run_id,))
+    row = await fetchone(
+        "SELECT ar.plan_data, ar.created_at, h.hospital_name, h.hospital_name_en, h.url"
+        " FROM analysis_runs ar"
+        " JOIN hospital_baseinfo h ON ar.hospital_id = h.hospital_id"
+        " WHERE ar.analysis_run_id = %s",
+        (run_id,),
+    )
     if row is None:
         raise HTTPException(status_code=404, detail="Run not found")
     if row["plan_data"] is None:
         return Response(status_code=204)
     data = json.loads(row["plan_data"]) if isinstance(row["plan_data"], str) else row["plan_data"]
-    return PlanOutput(**data)
+    plan = PlanOutput(**data)
+    return PlanApiResponse(
+        id=run_id,
+        clinicName=row["hospital_name"],
+        clinicNameEn=row["hospital_name_en"],
+        createdAt=str(row["created_at"]),
+        targetUrl=row["url"],
+        **plan.model_dump(),
+    )
diff --git a/app/api/reports.py b/app/api/reports.py
index 1c1ff96..6fca3da 100644
--- a/app/api/reports.py
+++ b/app/api/reports.py
@@ -4,16 +4,20 @@ from fastapi import APIRouter, Depends, HTTPException, Response
 from common.db import fetchone
 from common.deps import verify_api_key
 from integrations.llm.schemas.report import ReportOutput
+from models.report import MarketingReportResponse
 
 router = APIRouter(prefix="/api/reports", tags=["reports"], dependencies=[Depends(verify_api_key)])
 logger = logging.getLogger(__name__)
 
 
-@router.get("/{run_id}", response_model=ReportOutput | None)
+@router.get("/{run_id}", response_model=MarketingReportResponse, response_model_by_alias=True)
 async def get_report(run_id: str):
     logger.info("GET /api/reports/%s", run_id)
     row = await fetchone(
-        "SELECT report_data FROM analysis_runs WHERE analysis_run_id = %s",
+        "SELECT ar.report_data, ar.created_at, h.hospital_name, h.hospital_name_en, h.url"
+        " FROM analysis_runs ar"
+        " JOIN hospital_baseinfo h ON ar.hospital_id = h.hospital_id"
+        " WHERE ar.analysis_run_id = %s",
         (run_id,),
     )
     if row is None:
@@ -21,4 +25,11 @@ async def get_report(run_id: str):
     if row["report_data"] is None:
         return Response(status_code=204)
     data = json.loads(row["report_data"]) if isinstance(row["report_data"], str) else row["report_data"]
-    return ReportOutput(**data)
+    llm_output = ReportOutput(**data)
+    response = MarketingReportResponse.model_validate(llm_output.model_dump())
+    response.id = run_id
+    response.clinic_name = row["hospital_name"]
+    response.clinic_name_en = row["hospital_name_en"]
+    response.created_at = str(row["created_at"])
+    response.target_url = row["url"]
+    return response
diff --git a/app/common/utils.py b/app/common/utils.py
index e2d6be2..d7c4b21 100644
--- a/app/common/utils.py
+++ b/app/common/utils.py
@@ -7,6 +7,7 @@ import httpx
 REQUEST_TIMEOUT = 60
 
 
+
 def get_env(key: str) -> str:
     v = os.environ.get(key, "")
     if not v:
diff --git a/app/integrations/llm/llm_service.py b/app/integrations/llm/llm_service.py
index 8272d54..c35928f 100644
--- a/app/integrations/llm/llm_service.py
+++ b/app/integrations/llm/llm_service.py
@@ -45,6 +45,7 @@ class LLMService:
                 response = await self.client.chat.completions.create(
                     model=prompt.model,
                     messages=[{"role": "user", "content": prompt_text}],
+                    max_tokens=16000,
                     response_format={
                         "type": "json_schema",
                         "json_schema": {"name": prompt.output_class.__name__, "schema": prompt.output_class.model_json_schema()},
diff --git a/app/integrations/llm/schemas/report.py b/app/integrations/llm/schemas/report.py
index 0757e14..4dae1c4 100644
--- a/app/integrations/llm/schemas/report.py
+++ b/app/integrations/llm/schemas/report.py
@@ -1,8 +1,316 @@
+from __future__ import annotations
+from typing import Literal
 from pydantic import BaseModel
 
+Severity = Literal["critical", "warning", "good", "excellent", "unknown"]
+ChannelStatus = Literal["active", "inactive", "unknown", "not_found"]
+DataSource = Literal["registry", "scrape"]
+Language = Literal["KR", "EN"]
+VideoType = Literal["Short", "Long"]
+AnnotationType = Literal["highlight", "arrow", "text"]
+
+
+class ScreenshotAnnotation(BaseModel):
+    type: AnnotationType
+    x: float
+    y: float
+    width: float | None = None
+    height: float | None = None
+    label: str | None = None
+    color: str | None = None
+
+
+class ScreenshotEvidence(BaseModel):
+    id: str
+    url: str
+    channel: str
+    captured_at: str
+    caption: str
+    source_url: str | None = None
+    annotations: list[ScreenshotAnnotation] | None = None
+
+
+class DiagnosisItem(BaseModel):
+    category: str
+    detail: str
+    severity: Severity
+    evidence_ids: list[str] | None = None
+
+
+# --- ClinicSnapshot ---
+
+class LeadDoctor(BaseModel):
+    name: str
+    credentials: str
+    rating: float
+    review_count: int
+
+
+class PriceRange(BaseModel):
+    min: str
+    max: str
+    currency: str
+
+
+class LogoImages(BaseModel):
+    circle: str | None = None
+    horizontal: str | None = None
+    korean: str | None = None
+
+
+class BrandColors(BaseModel):
+    primary: str
+    accent: str
+    text: str
+
+
+class RegistryData(BaseModel):
+    district: str | None = None
+    branches: str | None = None
+    brand_group: str | None = None
+    website_en: str | None = None
+    naver_place_url: str | None = None
+    gangnam_unni_url: str | None = None
+    google_maps_url: str | None = None
+
+
+class ClinicSnapshot(BaseModel):
+    name: str
+    name_en: str
+    established: str
+    years_in_business: int
+    staff_count: int
+    lead_doctor: LeadDoctor
+    overall_rating: float
+    total_reviews: int
+    price_range: PriceRange
+    certifications: list[str]
+    media_appearances: list[str]
+    medical_tourism: list[str]
+    location: str
+    nearest_station: str
+    phone: str
+    domain: str
+    logo_images: LogoImages | None = None
+    brand_colors: BrandColors | None = None
+    source: DataSource | None = None
+    registry_data: RegistryData | None = None
+
+
+# --- ChannelScore ---
+
+class ChannelScore(BaseModel):
+    channel: str
+    icon: str
+    score: int
+    max_score: int
+    status: Severity
+    headline: str
+
+
+# --- YouTube ---
+
+class WeeklyViewGrowth(BaseModel):
+    absolute: int
+    percentage: float
+
+
+class EstimatedRevenue(BaseModel):
+    min: int
+    max: int
+
+
+class LinkedUrl(BaseModel):
+    label: str
+    url: str
+
+
+class TopVideo(BaseModel):
+    title: str
+    views: int
+    uploaded_ago: str
+    type: VideoType
+    duration: str | None = None
+
+
+class YouTubeAudit(BaseModel):
+    channel_name: str
+    handle: str
+    subscribers: int
+    total_videos: int
+    total_views: int
+    weekly_view_growth: WeeklyViewGrowth
+    estimated_monthly_revenue: EstimatedRevenue
+    avg_video_length: str
+    upload_frequency: str
+    channel_created_date: str
+    subscriber_rank: str
+    channel_description: str
+    linked_urls: list[LinkedUrl]
+    playlists: list[str]
+    top_videos: list[TopVideo]
+    diagnosis: list[DiagnosisItem]
+
+
+# --- Instagram ---
+
+class InstagramAccount(BaseModel):
+    handle: str
+    language: Language
+    label: str
+    posts: int
+    followers: int
+    following: int
+    category: str
+    profile_link: str
+    highlights: list[str]
+    reels_count: int
+    content_format: str
+    profile_photo: str
+    bio: str
+
+
+class InstagramAudit(BaseModel):
+    accounts: list[InstagramAccount]
+    diagnosis: list[DiagnosisItem]
+
+
+# --- Facebook ---
+
+class BrandInconsistencyValue(BaseModel):
+    channel: str
+    value: str
+    is_correct: bool
+
+
+class BrandInconsistency(BaseModel):
+    field: str
+    values: list[BrandInconsistencyValue]
+    impact: str
+    recommendation: str
+
+
+class FacebookPage(BaseModel):
+    url: str
+    page_name: str
+    language: Language
+    label: str
+    followers: int
+    following: int
+    category: str
+    bio: str
+    logo: str
+    logo_description: str
+    link: str
+    linked_domain: str
+    reviews: int
+    recent_post_age: str
+    has_whatsapp: bool
+    post_frequency: str | None = None
+    top_content_type: str | None = None
+    engagement: str | None = None
+
+
+class FacebookAudit(BaseModel):
+    pages: list[FacebookPage]
+    diagnosis: list[DiagnosisItem]
+    brand_inconsistencies: list[BrandInconsistency]
+    consolidation_recommendation: str
+
+
+# --- 기타 채널 / 웹사이트 ---
+
+class OtherChannel(BaseModel):
+    name: str
+    status: ChannelStatus
+    details: str
+    url: str | None = None
+
+
+class TrackingPixel(BaseModel):
+    name: str
+    installed: bool
+    details: str | None = None
+
+
+class SnsLink(BaseModel):
+    platform: str
+    url: str
+    location: str
+
+
+class AdditionalDomain(BaseModel):
+    domain: str
+    purpose: str
+
+
+class WebsiteAudit(BaseModel):
+    primary_domain: str
+    additional_domains: list[AdditionalDomain]
+    sns_links_on_site: bool
+    sns_links_detail: list[SnsLink] | None = None
+    tracking_pixels: list[TrackingPixel]
+    main_cta: str
+
+
+# --- Transformation ---
+
+class AsIsToBeItem(BaseModel):
+    area: str
+    as_is: str
+    to_be: str
+
+
+class StrategyDetail(BaseModel):
+    strategy: str
+    detail: str
+
+
+class PlatformStrategy(BaseModel):
+    platform: str
+    icon: str
+    current_metric: str
+    target_metric: str
+    strategies: list[StrategyDetail]
+
+
+class NewChannelProposal(BaseModel):
+    channel: str
+    priority: str
+    rationale: str
+
+
+class TransformationProposal(BaseModel):
+    brand_identity: list[AsIsToBeItem]
+    content_strategy: list[AsIsToBeItem]
+    platform_strategies: list[PlatformStrategy]
+    website_improvements: list[AsIsToBeItem]
+    new_channel_proposals: list[NewChannelProposal]
+
+
+# --- Roadmap / KPI ---
+
+class RoadmapTask(BaseModel):
+    task: str
+    completed: bool
+
+
+class RoadmapMonth(BaseModel):
+    month: int
+    title: str
+    subtitle: str
+    tasks: list[RoadmapTask]
+
+
+class KPIMetric(BaseModel):
+    metric: str
+    current: str
+    target_3_month: str
+    target_12_month: str
+
+
+# --- ReportInput (prompt 템플릿 변수) ---
 
-# template.format(**model_dump()) 에 삽입될 변수들
-# 각 채널 raw_data를 호출부에서 json.dumps()로 직렬화해서 넘겨야 함
 class ReportInput(BaseModel):
     clinic_name: str | None = None
     clinic_name_en: str | None = None
@@ -18,26 +326,25 @@ class ReportInput(BaseModel):
     gangnam_unni: str | None = None
 
 
-class ChannelScore(BaseModel):
-    score: int
-    summary: str
-    strengths: list[str]
-    weaknesses: list[str]
+# --- MarketingReport ---
 
-
-class ConversionStrategy(BaseModel):
-    summary: str
-    actions: list[str]
-
-
-# response_format으로 OpenAI structured output에 전달 — dict 필드 사용 불가
-class ReportOutput(BaseModel):
+class MarketingReport(BaseModel):
+    id: str
+    created_at: str
+    target_url: str
     overall_score: int
-    instagram: ChannelScore | None = None
-    facebook: ChannelScore | None = None
-    naver_blog: ChannelScore | None = None
-    youtube: ChannelScore | None = None
-    gangnam_unni: ChannelScore | None = None
-    conversion_strategy: ConversionStrategy
-    roadmap: list[str]
-    kpis: list[str]
+    clinic_snapshot: ClinicSnapshot
+    channel_scores: list[ChannelScore]
+    youtube_audit: YouTubeAudit
+    instagram_audit: InstagramAudit
+    facebook_audit: FacebookAudit
+    other_channels: list[OtherChannel]
+    website_audit: WebsiteAudit
+    problem_diagnosis: list[DiagnosisItem]
+    transformation: TransformationProposal
+    roadmap: list[RoadmapMonth]
+    kpi_dashboard: list[KPIMetric]
+    screenshots: list[ScreenshotEvidence] | None = None
+
+
+ReportOutput = MarketingReport
diff --git a/app/models/plan.py b/app/models/plan.py
index 724dda9..9e41b1a 100644
--- a/app/models/plan.py
+++ b/app/models/plan.py
@@ -1,16 +1,19 @@
 from pydantic import BaseModel
+from integrations.llm.schemas.plan import (
+    BrandGuide, ChannelStrategyCard, ContentStrategyData,
+    CalendarData, AssetCollectionData, RepurposingProposalItem,
+)
 
 
-class PlanCreate(BaseModel):
-    report_id: str
-    regenerate: bool = False
-
-
-class PlanResponse(BaseModel):
+class PlanApiResponse(BaseModel):
     id: str
-    analysis_run_id: str
-    brand_guide: dict
-    channel_strategies: list
-    content_strategy: dict
-    calendar: list
-    created_at: str
+    clinicName: str | None = None
+    clinicNameEn: str | None = None
+    createdAt: str
+    targetUrl: str
+    brandGuide: BrandGuide
+    channelStrategies: list[ChannelStrategyCard]
+    contentStrategy: ContentStrategyData
+    calendar: CalendarData
+    assetCollection: AssetCollectionData
+    repurposingProposals: list[RepurposingProposalItem] | None = None
diff --git a/app/models/report.py b/app/models/report.py
index 043e9b6..b719769 100644
--- a/app/models/report.py
+++ b/app/models/report.py
@@ -1,22 +1,318 @@
-from pydantic import BaseModel
+from __future__ import annotations
+from pydantic import BaseModel, ConfigDict
+from integrations.llm.schemas.report import (
+    Severity, ChannelStatus, DataSource, Language, VideoType, AnnotationType,
+)
 
 
-class ClinicInfo(BaseModel):
+def _to_camel(s: str) -> str:
+    head, *tail = s.split("_")
+    return head + "".join(w.capitalize() for w in tail)
+
+
+class CamelModel(BaseModel):
+    model_config = ConfigDict(populate_by_name=True, alias_generator=_to_camel)
+
+
+class ScreenshotAnnotation(CamelModel):
+    type: AnnotationType
+    x: float
+    y: float
+    width: float | None = None
+    height: float | None = None
+    label: str | None = None
+    color: str | None = None
+
+
+class ScreenshotEvidence(CamelModel):
+    id: str
+    url: str
+    channel: str
+    captured_at: str
+    caption: str
+    source_url: str | None = None
+    annotations: list[ScreenshotAnnotation] | None = None
+
+
+class DiagnosisItem(CamelModel):
+    category: str
+    detail: str
+    severity: Severity
+    evidence_ids: list[str] | None = None
+
+
+class LeadDoctor(CamelModel):
     name: str
+    credentials: str
+    rating: float
+    review_count: int
+
+
+class PriceRange(CamelModel):
+    min: str
+    max: str
+    currency: str
+
+
+class LogoImages(CamelModel):
+    circle: str | None = None
+    horizontal: str | None = None
+    korean: str | None = None
+
+
+class BrandColors(CamelModel):
+    primary: str
+    accent: str
+    text: str
+
+
+class RegistryData(CamelModel):
+    district: str | None = None
+    branches: str | None = None
+    brand_group: str | None = None
+    website_en: str | None = None
+    naver_place_url: str | None = None
+    gangnam_unni_url: str | None = None
+    google_maps_url: str | None = None
+
+
+class ClinicSnapshot(CamelModel):
+    name: str
+    name_en: str
+    established: str
+    years_in_business: int
+    staff_count: int
+    lead_doctor: LeadDoctor
+    overall_rating: float
+    total_reviews: int
+    price_range: PriceRange
+    certifications: list[str]
+    media_appearances: list[str]
+    medical_tourism: list[str]
+    location: str
+    nearest_station: str
+    phone: str
+    domain: str
+    logo_images: LogoImages | None = None
+    brand_colors: BrandColors | None = None
+    source: DataSource | None = None
+    registry_data: RegistryData | None = None
+
+
+class ChannelScore(CamelModel):
+    channel: str
+    icon: str
+    score: int
+    max_score: int
+    status: Severity
+    headline: str
+
+
+class WeeklyViewGrowth(CamelModel):
+    absolute: int
+    percentage: float
+
+
+class EstimatedRevenue(CamelModel):
+    min: int
+    max: int
+
+
+class LinkedUrl(CamelModel):
+    label: str
     url: str
 
 
-class ReportResponse(BaseModel):
+class TopVideo(CamelModel):
+    title: str
+    views: int
+    uploaded_ago: str
+    type: VideoType
+    duration: str | None = None
+
+
+class YouTubeAudit(CamelModel):
+    channel_name: str
+    handle: str
+    subscribers: int
+    total_videos: int
+    total_views: int
+    weekly_view_growth: WeeklyViewGrowth
+    estimated_monthly_revenue: EstimatedRevenue
+    avg_video_length: str
+    upload_frequency: str
+    channel_created_date: str
+    subscriber_rank: str
+    channel_description: str
+    linked_urls: list[LinkedUrl]
+    playlists: list[str]
+    top_videos: list[TopVideo]
+    diagnosis: list[DiagnosisItem]
+
+
+class InstagramAccount(CamelModel):
+    handle: str
+    language: Language
+    label: str
+    posts: int
+    followers: int
+    following: int
+    category: str
+    profile_link: str
+    highlights: list[str]
+    reels_count: int
+    content_format: str
+    profile_photo: str
+    bio: str
+
+
+class InstagramAudit(CamelModel):
+    accounts: list[InstagramAccount]
+    diagnosis: list[DiagnosisItem]
+
+
+class BrandInconsistencyValue(CamelModel):
+    channel: str
+    value: str
+    is_correct: bool
+
+
+class BrandInconsistency(CamelModel):
+    field: str
+    values: list[BrandInconsistencyValue]
+    impact: str
+    recommendation: str
+
+
+class FacebookPage(CamelModel):
+    url: str
+    page_name: str
+    language: Language
+    label: str
+    followers: int
+    following: int
+    category: str
+    bio: str
+    logo: str
+    logo_description: str
+    link: str
+    linked_domain: str
+    reviews: int
+    recent_post_age: str
+    has_whatsapp: bool
+    post_frequency: str | None = None
+    top_content_type: str | None = None
+    engagement: str | None = None
+
+
+class FacebookAudit(CamelModel):
+    pages: list[FacebookPage]
+    diagnosis: list[DiagnosisItem]
+    brand_inconsistencies: list[BrandInconsistency]
+    consolidation_recommendation: str
+
+
+class OtherChannel(CamelModel):
+    name: str
+    status: ChannelStatus
+    details: str
+    url: str | None = None
+
+
+class TrackingPixel(CamelModel):
+    name: str
+    installed: bool
+    details: str | None = None
+
+
+class SnsLink(CamelModel):
+    platform: str
+    url: str
+    location: str
+
+
+class AdditionalDomain(CamelModel):
+    domain: str
+    purpose: str
+
+
+class WebsiteAudit(CamelModel):
+    primary_domain: str
+    additional_domains: list[AdditionalDomain]
+    sns_links_on_site: bool
+    sns_links_detail: list[SnsLink] | None = None
+    tracking_pixels: list[TrackingPixel]
+    main_cta: str
+
+
+class AsIsToBeItem(CamelModel):
+    area: str
+    as_is: str
+    to_be: str
+
+
+class StrategyDetail(CamelModel):
+    strategy: str
+    detail: str
+
+
+class PlatformStrategy(CamelModel):
+    platform: str
+    icon: str
+    current_metric: str
+    target_metric: str
+    strategies: list[StrategyDetail]
+
+
+class NewChannelProposal(CamelModel):
+    channel: str
+    priority: str
+    rationale: str
+
+
+class TransformationProposal(CamelModel):
+    brand_identity: list[AsIsToBeItem]
+    content_strategy: list[AsIsToBeItem]
+    platform_strategies: list[PlatformStrategy]
+    website_improvements: list[AsIsToBeItem]
+    new_channel_proposals: list[NewChannelProposal]
+
+
+class RoadmapTask(CamelModel):
+    task: str
+    completed: bool
+
+
+class RoadmapMonth(CamelModel):
+    month: int
+    title: str
+    subtitle: str
+    tasks: list[RoadmapTask]
+
+
+class KPIMetric(CamelModel):
+    metric: str
+    current: str
+    target_3_month: str
+    target_12_month: str
+
+
+class MarketingReportResponse(CamelModel):
     id: str
-    clinic: ClinicInfo
+    clinic_name: str | None = None
+    clinic_name_en: str | None = None
+    created_at: str
+    target_url: str
     overall_score: int
-    youtube: dict
-    instagram: dict
-    facebook: dict
-    naver_place: dict
-    naver_blog: dict
-    gangnam_unni: dict
-    conversion_strategy: dict
-    roadmap: list
-    kpis: list
-    generated_at: str
+    clinic_snapshot: ClinicSnapshot
+    channel_scores: list[ChannelScore]
+    youtube_audit: YouTubeAudit
+    instagram_audit: InstagramAudit
+    facebook_audit: FacebookAudit
+    other_channels: list[OtherChannel]
+    website_audit: WebsiteAudit
+    problem_diagnosis: list[DiagnosisItem]
+    transformation: TransformationProposal
+    roadmap: list[RoadmapMonth]
+    kpi_dashboard: list[KPIMetric]
+    screenshots: list[ScreenshotEvidence] | None = None