diff --git a/.coderabbit/config.yml b/.coderabbit/config.yml
new file mode 100644
index 0000000..647d280
--- /dev/null
+++ b/.coderabbit/config.yml
@@ -0,0 +1,2 @@
+features:
+  docstrings: false
\ No newline at end of file
diff --git a/.github/workflows/deploy.yml b/.github/workflows/deploy.yml
index afd7075..4ad5e42 100644
--- a/.github/workflows/deploy.yml
+++ b/.github/workflows/deploy.yml
@@ -14,6 +14,7 @@ env:
   AWS_HOST: ${{ secrets.AWS_HOST }}
   AWS_SSH_KEY: ${{ secrets.AWS_SSH_KEY }}
   AWS_USER: ${{ secrets.AWS_USER }}
+  GPT_API_KEY: ${{ secrets.GPT_API_KEY }}
 
 
 jobs:
@@ -55,4 +56,9 @@ jobs:
             fi
 
             docker pull "${{ env.DOCKER_REPO }}:latest"
-            docker run -d --name "Block-Guard-AI" -p 8000:8000 ${{ env.DOCKER_REPO }}:latest
\ No newline at end of file
+
+            docker run -d \
+              --name "Block-Guard-AI" \
+              -p 8000:8000 \
+              -e GPT_API_KEY="${{ env.GPT_API_KEY }}" \
+              ${{ env.DOCKER_REPO }}:latest
\ No newline at end of file
diff --git a/app/api/fraud_analysis.py b/app/api/fraud_analysis.py
new file mode 100644
index 0000000..9bbb7da
--- /dev/null
+++ b/app/api/fraud_analysis.py
@@ -0,0 +1,21 @@
+from fastapi import APIRouter, HTTPException
+from app.services.gpt_service import call_gpt
+from app.models.fraud_request import FraudRequest
+from app.models.fraud_response import FraudResponse
+
+router = APIRouter()
+
+@router.post(
+    path = "",
+    response_model=FraudResponse,
+    summary="사기 유형 분석"
+)
+async def fraud_analysis(request: FraudRequest):
+    try:
+        answer = await call_gpt(request)
+        response = FraudResponse.model_validate_json(answer)
+        return response
+    except RuntimeError as e:
+        raise HTTPException(status_code=500, detail=f"사기분석 실패: {e}") from e
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"응답 파싱 실패: {e}") from e
diff --git a/app/config/setting.py b/app/config/setting.py
new file mode 100644
index 0000000..4d36a7f
--- /dev/null
+++ b/app/config/setting.py
@@ -0,0 +1,9 @@
+from pydantic_settings import BaseSettings
+
+class Settings(BaseSettings):
+    gpt_api_key: str
+
+    class Config:
+        env_file = ".env"
+
+settings = Settings()
diff --git a/app/main.py b/app/main.py
index 7245ae5..e955943 100644
--- a/app/main.py
+++ b/app/main.py
@@ -1,10 +1,10 @@
 from fastapi import FastAPI
-from app.routers import fraud_analysis
+from app.api import fraud_analysis
 
 app = FastAPI()
 
 # router 등록
-app.include_router(fraud_analysis.router, prefix="/api/fraud_analysis", tags=["fraud_analysis"])
+app.include_router(fraud_analysis.router, prefix="/api/fraud-analysis", tags=["fraud-analysis"])
 
 @app.get("/api/hello")
 async def hello():
diff --git a/app/models/fraud_request.py b/app/models/fraud_request.py
new file mode 100644
index 0000000..9a70a3a
--- /dev/null
+++ b/app/models/fraud_request.py
@@ -0,0 +1,9 @@
+from pydantic import BaseModel
+from typing import List
+from typing import Optional
+
+class FraudRequest(BaseModel):
+    messageContent: Optional[str] = None
+    keywords: List[str]
+    additionalDescription: Optional[str] = None
+    imageContent: Optional[str] = None
\ No newline at end of file
diff --git a/app/models/fraud_response.py b/app/models/fraud_response.py
new file mode 100644
index 0000000..4fb0ea3
--- /dev/null
+++ b/app/models/fraud_response.py
@@ -0,0 +1,8 @@
+from pydantic import BaseModel, Field
+from typing import List
+
+class FraudResponse(BaseModel):
+    estimatedFraudType: str                  # 분류된 사기 유형
+    keywords: List[str] = Field(..., min_items=1, max_items=3, description="주요 위험 키워드 (최대 3개)")
+    explanation: str                      # 해당 유형으로 판단한 이유
+    score: float = Field(..., ge=0, le=100, description="위험도(0~100)")
\ No newline at end of file
diff --git a/app/prompts/data/fraud_examples.py b/app/prompts/data/fraud_examples.py
new file mode 100644
index 0000000..7be5e4f
--- /dev/null
+++ b/app/prompts/data/fraud_examples.py
@@ -0,0 +1,28 @@
+from typing import List
+from app.prompts.fraud_prompts import FraudExample
+
+# 여기에 18개 유형 × 3~4개 예시만큼 리스트를 정의
+FRAUD_EXAMPLES: List[FraudExample] = [
+    FraudExample(
+        type_name="피싱",
+        message_content="귀하의 계정이 잠길 수 있습니다. 비밀번호를 확인하려면 여기를 클릭하세요.",
+        additional_description="늦은 밤, 낯선 번호로 온 SMS를 우연히 열어보았습니다.",
+        keywords=["계정 잠금", "비밀번호 확인", "SMS 링크"],
+        image_content="은행알림 Bot: 당신의 계좌가 비정상적으로 활동되어 잠길 예정입니다. https://bit.ly/secure-link"
+    ),
+    FraudExample(
+        type_name="투자 사기",
+        message_content="2주 만에 30% 수익 보장! 지금 투자하세요.",
+        additional_description="점심시간 카페에서 친구가 공유한 링크를 클릭했습니다.",
+        keywords=["30% 수익", "지금 투자", "보장된 수익"],
+        image_content="InvestPro: 한정 시간 30% 수익, 투자하기 ▶ http://invest.example.com"
+    ),
+    FraudExample(
+        type_name="복권 사기",
+        message_content="축하합니다! 당신은 1,000,000달러에 당첨되었습니다. 수수료를 지불하면 수령 가능합니다.",
+        additional_description="출근길 지하철에서 받은 이메일에 포함된 링크를 클릭했습니다.",
+        keywords=["당첨", "수수료", "이메일 링크"],
+        image_content="LotteryKingdom: Congratulations! You've won $1,000,000. Click here https://lotto.fake/claim"
+    ),
+    # … (총 60~70개 항목)
+]
diff --git a/app/prompts/fraud_example.py b/app/prompts/fraud_example.py
new file mode 100644
index 0000000..6d768bf
--- /dev/null
+++ b/app/prompts/fraud_example.py
@@ -0,0 +1,9 @@
+from pydantic import BaseModel
+from typing import List
+
+class FraudExample(BaseModel):
+    type_name: str
+    message_content: str
+    keywords: List[str]
+    additional_description: str
+    image_content: str
\ No newline at end of file
diff --git a/app/prompts/fraud_prompts.py b/app/prompts/fraud_prompts.py
new file mode 100644
index 0000000..4733a15
--- /dev/null
+++ b/app/prompts/fraud_prompts.py
@@ -0,0 +1,67 @@
+from typing import List, Dict
+from app.prompts.fraud_example import FraudExample
+from app.prompts.data.fraud_examples import FRAUD_EXAMPLES
+
+def get_fraud_detection_prompt(
+    message_content: str,
+    additional_description: str,
+    keywords: List[str],
+    image_content: str,
+    examples: List[FraudExample] = FRAUD_EXAMPLES
+) -> List[Dict[str, str]]:
+    
+    system = {
+        "role": "system",
+        "content": (
+            "당신은 사기 탐지 어시스턴트입니다. "
+            "입력된 텍스트를 반드시 미리 정의된 사기 유형 중 하나로 분류하고, "
+            "최소 1개에서 최대 3개의 주요 위험 키워드를 추출하며, 그 이유를 설명하고, "
+            "위험 점수(0–100%)를 제공해야 합니다.\n"
+            "출력은 반드시 valid JSON 객체로만 응답하세요. 아래는 응답 예시입니다:\n"
+            "{\n"
+            "  \"estimatedFraudType\": \"복권 사기\",\n"
+            "  \"keywords\": [\"키워드1\", \"키워드2\"],\n"
+            "  \"explanation\": \"...\",\n"
+            "  \"score\": 92.4\n"
+            "}\n"
+        )
+    }
+
+    example_lines = build_example_lines(examples)
+    assistant_content = "".join(example_lines)
+    assistant_content += (
+        "이제 아래 입력을 같은 형식으로 분류하세요:\n"
+    )
+
+    assistant = {
+        "role": "assistant",
+        "content": assistant_content
+    }
+
+    user = {
+        "role": "user",
+        "content": (
+            f"messageContent: '{message_content}'\n"
+            f"additionalDescription: '{additional_description}'\n"
+            f"keywords: {', '.join(keywords) if keywords else 'None'}\n"
+            f"imageContent: '{image_content}'"
+        )
+    }
+
+    return [system, assistant, user]
+
+def build_example_lines(examples):
+    example_lines = ["사기 유형 및 예시:\n"]
+    for idx, ex in enumerate(examples, start=1):
+        example_lines.append(f"{idx}. {ex.type_name}:\n")
+        example_lines.append(f"   messageContent: '{ex.message_content}'\n")
+        example_lines.append(f"   additionalDescription: '{ex.additional_description}'\n")
+        example_lines.append(f"   keywords: '{ex.keywords}'\n")
+        example_lines.append(f"   imageContent: '{ex.image_content}'\n")
+        example_lines.append("   출력 JSON 예시:\n")
+        example_lines.append(
+            f"   {{\"estimatedFraudType\": \"{ex.type_name}\", "
+            f"\"keywords\": [...], \"explanation\": \"...\", \"score\": ...}}\n\n"
+        )
+        
+    return example_lines
diff --git a/app/routers/fraud_analysis.py b/app/routers/fraud_analysis.py
deleted file mode 100644
index 276d6c6..0000000
--- a/app/routers/fraud_analysis.py
+++ /dev/null
@@ -1,7 +0,0 @@
-from fastapi import APIRouter, HTTPException
-
-router = APIRouter()
-
-@router.get("")
-async def fraud_analysis():
-    return {"message": "사기분석 API"}
\ No newline at end of file
diff --git a/app/services/gpt_service.py b/app/services/gpt_service.py
new file mode 100644
index 0000000..19722b0
--- /dev/null
+++ b/app/services/gpt_service.py
@@ -0,0 +1,37 @@
+from openai import OpenAI, OpenAIError
+from app.config.setting import settings
+from app.models.fraud_request import FraudRequest
+from app.prompts.fraud_prompts import get_fraud_detection_prompt
+import logging
+
+logger = logging.getLogger(__name__)
+
+client = OpenAI(
+    api_key = settings.gpt_api_key
+    )
+
+async def call_gpt(request: FraudRequest):
+    messages = get_fraud_detection_prompt(
+        message_content = request.messageContent, 
+        additional_description = request.additionalDescription,
+        keywords = request.keywords,
+        image_content = request.imageContent
+    )
+    
+    try:
+        response = client.responses.create(
+            model="gpt-4o-mini",
+            input = messages,
+            temperature = 0.5, # 생성된 텍스트의 무작위성을 결정
+            max_output_tokens = 200
+        )
+        logger.info(f"GPT API 호출 성공: {response}")
+        
+    except OpenAIError as e:
+        logger.error(f"GPT API 호출 실패: {e}", exc_info=True)
+        raise RuntimeError(f"GPT API 호출 실패: {e}") from e
+    except Exception as e:
+        logger.error(f"서버 오류 발생: {e}", exc_info=True)
+        raise RuntimeError(f"서버 오류 발생: {e}") from e
+
+    return response.output_text.strip()
diff --git a/requirements.txt b/requirements.txt
index c2ad132..2485208 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,16 +1,22 @@
 annotated-types==0.7.0
 anyio==4.9.0
-click==8.1.3
+distro==1.9.0
 fastapi==0.116.0
 h11==0.16.0
+httpcore==1.0.9
 httptools==0.6.4
+httpx==0.28.1
 idna==3.10
+jiter==0.10.0
+openai==1.97.1
 pydantic==2.11.7
+pydantic-settings==2.10.1
 pydantic_core==2.33.2
 python-dotenv==1.1.1
 PyYAML==6.0.2
 sniffio==1.3.1
 starlette==0.46.2
+tqdm==4.67.1
 typing-inspection==0.4.1
 typing_extensions==4.14.1
 uvicorn==0.35.0