diff --git a/app/integrations/llm/prompt.py b/app/integrations/llm/prompt.py
index 9c00c23..3ee76d4 100644
--- a/app/integrations/llm/prompt.py
+++ b/app/integrations/llm/prompt.py
@@ -1,7 +1,12 @@
 import os
 from pydantic import BaseModel
 from common.utils import get_env
-from integrations.llm.schemas.report import ReportInput, ReportOutput, YouTubeDiagnosisInput, YouTubeDiagnosisOutput, BrandConsistencyInput, BrandConsistencyOutput
+from integrations.llm.schemas.report import (
+    ReportInput, ReportOutput,
+    CriticalIssuesInput, CriticalIssuesOutput,
+    YouTubeDiagnosisInput, YouTubeDiagnosisOutput,
+    BrandConsistencyInput, BrandConsistencyOutput,
+)
 from integrations.llm.schemas.plan import PlanInput, PlanOutput
 from integrations.llm.schemas.market import (
     MarketCompetitorsInput, MarketCompetitorsOutput,
@@ -94,3 +99,10 @@ brand_consistency_prompt = Prompt(
     input_class=BrandConsistencyInput,
     output_class=BrandConsistencyOutput,
 )
+
+critical_issues_prompt = Prompt(
+    file_name="critical_issues_prompt.txt",
+    prompt_model="REPORT_MODEL",
+    input_class=CriticalIssuesInput,
+    output_class=CriticalIssuesOutput,
+)
diff --git a/app/integrations/llm/schemas/report.py b/app/integrations/llm/schemas/report.py
index 5e6ee46..8bbee91 100644
--- a/app/integrations/llm/schemas/report.py
+++ b/app/integrations/llm/schemas/report.py
@@ -357,6 +357,17 @@ class YouTubeDiagnosisOutput(BaseModel):
     diagnosis: list[DiagnosisItem]
 
 
+# --- Diagnosis ---
+
+class CriticalIssuesInput(BaseModel):
+    clinic_name: str | None = None
+    data: str | None = None
+
+
+class CriticalIssuesOutput(BaseModel):
+    diagnosis: list[DiagnosisItem]
+
+
 # --- BrandConsistency ---
 
 class BrandConsistencyInput(BaseModel):
diff --git a/app/integrations/llm/temp-prompt/critical_issues_prompt.txt b/app/integrations/llm/temp-prompt/critical_issues_prompt.txt
new file mode 100644
index 0000000..bcb6f78
--- /dev/null
+++ b/app/integrations/llm/temp-prompt/critical_issues_prompt.txt
@@ -0,0 +1,13 @@
+다음은 성형외과/피부과 {clinic_name} 의 전 채널 수집 데이터입니다.
+
+{data}
+
+위 데이터를 바탕으로 이 병원의 마케팅 전반에 걸친 핵심 문제점과 개선사항을 진단해줘.
+각 항목은 category(진단 카테고리), detail(상세 설명), severity(critical/warning/info) 형식의 JSON 배열로 출력해줘.
+
+현재 주요 진단 카테고리는 3개야.
+브랜드 아이덴티티 파편화
+콘텐츠 전략 부재
+플랫폼 간 유입 단절
+
+출처 번호([1], [2] 등)는 포함하지 마.
diff --git a/app/services/analysis.py b/app/services/analysis.py
index 1229047..48dcea9 100644
--- a/app/services/analysis.py
+++ b/app/services/analysis.py
@@ -8,8 +8,8 @@ from common.db.run import select_run, update_run_report, update_run_plan
 from common.db.source import select_run_raw_data, select_run_mainpage_url
 from common.db.market import select_market
 from integrations.llm.llm_service import LLMService
-from integrations.llm.prompt import report_prompt, plan_prompt, youtube_diagnosis_prompt, brand_consistency_prompt
-from integrations.llm.schemas.report import ReportOutput, ClinicSnapshot, YouTubeAudit, BrandConsistencyOutput
+from integrations.llm.prompt import report_prompt, plan_prompt, youtube_diagnosis_prompt, brand_consistency_prompt, critical_issues_prompt
+from integrations.llm.schemas.report import ReportOutput, ClinicSnapshot, YouTubeAudit, BrandConsistencyOutput, CriticalIssuesOutput, DiagnosisItem
 from integrations.llm.schemas.plan import PlanOutput
 
 logger = logging.getLogger(__name__)
@@ -208,6 +208,17 @@ async def _build_youtube_audit(youtube: dict) -> dict:
     return YouTubeAudit.model_validate(yt_patch).model_dump()
 
 
+async def _build_critical_issues(analysis_run_id: str, raw: dict) -> list[dict]:
+    result: CriticalIssuesOutput = await LLMService(provider="perplexity").generate(
+        critical_issues_prompt,
+        {
+            "clinic_name": (raw.get("mainpage") or {}).get("clinicName"),
+            "data":        json.dumps(raw, ensure_ascii=False),
+        },
+    )
+    return [DiagnosisItem.model_validate(item).model_dump() for item in result.diagnosis]
+
+
 async def _build_overrides(analysis_run_id: str) -> dict:
     raw = await select_run_raw_data(analysis_run_id)
     if not raw:
@@ -243,6 +254,7 @@ async def _build_overrides(analysis_run_id: str) -> dict:
 
     brand = await generate_brand_consistency(analysis_run_id)
     brand_patch = brand.model_dump()["brand_inconsistencies"]
+    critical_issues = await _build_critical_issues(analysis_run_id, raw)
 
     fb_patch: dict = {}
     if fb_pages:
@@ -259,6 +271,8 @@ async def _build_overrides(analysis_run_id: str) -> dict:
         overrides["facebook_audit"] = fb_patch
     if yt_patch:
         overrides["youtube_audit"] = yt_patch
+    if critical_issues:
+        overrides["problem_diagnosis"] = critical_issues
     return overrides