ต้นแบบ · รุ่นทดลอง: มีทุก server ในทะเบียนทางการ อัปเดตทุก 30 นาที รีวิวและบัญชีผู้ใช้จะมาในขั้นถัดไป
PopNx MCPที่รวม MCP สำหรับคนใช้ AI · อธิบายเป็นภาษาคน

Korail MCP

โดย lovelyqualityรุ่น 1.0.1i★ –iรันบนเครื่อง
ชื่อใน registryio.github.lovelyquality/korail-mcp

ติดตั้ง

ภาพรวม

คำอธิบายจากผู้ทำ (ภาษาอังกฤษตามต้นฉบับ)

한국철도공사(KORAIL) 공공데이터 MCP 서버 — 98개 도구, API 키 신청 불필요. 역·열차·화물·통계 데이터 자연어 조회.

สามเสาของ PopNx MCP

สามเรื่องนี้ไม่รวมเป็นคะแนนเดียว เพราะความเห็น การวัด และความนิยม เป็นคนละเรื่องกัน

ใช้ทำอะไรได้

ตัวนี้ยังไม่ได้เรียบเรียงเป็นภาษาไทย ด้านล่างคือข้อเท็จจริงที่เราอ่านจาก registry

  • การเชื่อมต่อ: รันบนเครื่อง
  • ช่องทาง (transport): stdio
  • หมวด: อื่น ๆ

ติดตั้ง

เลือกโปรแกรม AI ที่คุณใช้ แล้วคัดลอกค่าตั้งไปวาง

ก่อนติดตั้ง

ต้องเตรียมก่อน

  • PyPIติดตั้ง uv ของ Python (มีคำสั่ง uvx) ในเครื่อง

คุณใช้แอปไหน

เลือกแอปด้านบน เราจำไว้ให้ ครั้งหน้าปุ่มจะขึ้นว่า “ติดตั้งลง …”

เปิด Claude Desktop → Settings → Developer → Edit Config แล้ววางค่านี้ในไฟล์ claude_desktop_config.json

{
  "mcpServers": {
    "korail": {
      "command": "uvx",
      "args": [
        "korail-mcp"
      ]
    }
  }
}

พิมพ์คำสั่งนี้ใน Terminal

claude mcp add korail -- uvx korail-mcp

เปิดใน Cursor

หรือคัดลอกค่าตั้งเอง

วางในไฟล์ .cursor/mcp.json ของโปรเจกต์ หรือ ~/.cursor/mcp.json

{
  "mcpServers": {
    "korail": {
      "command": "uvx",
      "args": [
        "korail-mcp"
      ]
    }
  }
}

เบราว์เซอร์จะถามก่อนเปิดแอป จากนั้นแอปแสดงค่าตั้งให้ตรวจแล้วกดยืนยัน ช่อง key เว้นว่างไว้ให้คุณใส่เอง

เปิดใน VS Code

หรือคัดลอกค่าตั้งเอง

วางในไฟล์ .vscode/mcp.json ของโปรเจกต์

{
  "servers": {
    "korail": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "korail-mcp"
      ]
    }
  }
}

เบราว์เซอร์จะถามก่อนเปิดแอป จากนั้นแอปแสดงค่าตั้งให้ตรวจแล้วกดยืนยัน ช่อง key เว้นว่างไว้ให้คุณใส่เอง

ตัวนี้รันในเครื่องคุณ แต่ ChatGPT ต่อได้เฉพาะ server บนอินเทอร์เน็ต ลองเลือกแอปอื่น

ตัวเลือกเก็บในคุกกี้บนเครื่องนี้ ไม่ต้องมีบัญชี

ค่าตั้งนี้สร้างจากข้อมูลใน registry ของผู้ทำ ถ้าไม่ตรงกับ README ให้เชื่อ README

ทำอะไรได้บ้าง

รันในเครื่องคุณ รายการเครื่องมือจะขึ้นในแอป AI หลังติดตั้ง

ผลตรวจเบื้องต้นi

เราตรวจจากข้อมูลที่ใครก็เปิดดูได้ ยังไม่ได้อ่านโค้ด ข้อที่ไม่เกี่ยวกับตัวนี้ไม่นับในตัวหาร · ตรวจเมื่อ 7 พ.ย. 2569

  • ✓ ผ่าน

    ตัวตนผู้ทำตรงกับ repo

    ชื่อใน registry ยืนยันด้วยบัญชี GitHub lovelyquality ซึ่งเป็นเจ้าของ repo

  • … ยังไม่ตรวจ

    ยังมีคนดูแล

    ยังไม่ได้อ่านข้อมูล repo จาก GitHub

  • … ยังไม่ตรวจ

    มีสัญญาอนุญาต (license)

    ยังไม่ได้อ่านข้อมูล repo จาก GitHub

  • – ไม่เกี่ยว

    endpoint ตอบ

    ตัวนี้รันบนเครื่องคุณ ไม่มี endpoint บนอินเทอร์เน็ต

  • – ไม่เกี่ยว

    ล็อกอินแบบ OAuth

    ตัวนี้รันบนเครื่อง ใช้ค่าที่ตั้งในเครื่องแทน

  • … ยังไม่ตรวจ

    คำอธิบายเครื่องมือไม่มีคำสั่งแอบแฝง

    รันในเครื่องคุณ เราอ่านรายการเครื่องมือได้เฉพาะตัวที่มี endpoint บนอินเทอร์เน็ต

เทียบกับตัวอื่นในหมวด อื่น ๆi

  • ดาวมากกว่า 0% ของอีก 18348 ตัวในหมวดนี้
  • ส่งรุ่นใหม่เข้า registry ล่าสุดกว่า 29% ของอีก 18348 ตัวในหมวดนี้
  • ผลตรวจเบื้องต้นดีกว่าหรือเท่ากับ 100% ของอีก 18348 ตัวในหมวดนี้

เทียบกับทุก server ที่ยังใช้งานในหมวดนี้ทั้งทะเบียน ดาวนับเฉพาะเมื่อผู้ลงทะเบียนเป็นเจ้าของ repo

คำถามที่พบบ่อย

Korail MCP ปลอดภัยไหม?

จากการตรวจเบื้องต้นของเรา ผ่าน 1 จาก 1 ข้อที่ตรวจได้ เรายังไม่ได้อ่านโค้ด ก่อนใส่กุญแจหรือ token ให้ดูว่า server ขอสิทธิ์อะไร และเลือกกุญแจที่ทำได้เท่าที่จำเป็น

ต้องใช้กุญแจหรือรหัสอะไรบ้าง?

registry ไม่ได้ระบุว่าต้องใช้กุญแจ แต่บางตัวจะขอให้ล็อกอินตอนใช้ครั้งแรก

ตัวนี้รันบนเครื่องหรือบนคลาวด์?

รันบนเครื่องของคุณ โปรแกรม AI จะเปิด server ขึ้นมาเองเมื่อใช้ ข้อมูลไม่ต้องผ่านเซิร์ฟเวอร์ของผู้ทำ (ยกเว้นตัวที่ไปเรียกบริการออนไลน์ต่อ)

ยังมีคนดูแลอยู่ไหม?

repo บน GitHub มีการแก้โค้ดล่าสุด – และส่งรุ่นล่าสุด (1.0.1) เข้า registry 2 เดือนที่ผ่านมา

ใช้กับโปรแกรมไหนได้บ้าง?

โปรแกรมที่รองรับ MCP เช่น Claude Desktop, Claude Code, Cursor และ VS Code ตัวนี้ต้องมี Python (uv) ในเครื่องด้วย

มีตัวไหนใช้แทนได้บ้าง?

ในหมวดเดียวกันมี Spec Driven Development, Remote Desktop Commander, 好縮 Short

ตัวที่ใช้แทนได้

server หมวดเดียวกันจากผู้ทำคนอื่น เรียงตามดาว

อ่านการ์ด:★ ดาว GitHubiสามเสา: รีวิว · ผลตรวจ · การใช้งานiอัปเดต = วันที่ส่งรุ่นล่าสุดเข้า registryi

Spec Driven Development

โดย formulahendry

MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.

รันบนเครื่อง★ 440MITอัปเดต 14 เดือนที่ผ่านมา
ติดตั้ง

Remote Desktop Commander

โดย desktop-commander

Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.

รันบนคลาวด์★ 211อัปเดต เมื่อวานซืน
ติดตั้ง

好縮 Short

โดย gooonamee

Smart URL shortener: create, tag and track short links on s.gooo.name.

รันบนคลาวด์อัปเดต วันนี้
ติดตั้ง

README จากผู้ทำ

เปิดอ่าน README

ส่วนนี้คัดจาก repo ของผู้ทำตามต้นฉบับ PopNx MCP ไม่ได้ตรวจหรือรับรองเนื้อหา ตัด HTML รูปภาพ และสคริปต์ออกแล้ว ลิงก์จะพาออกนอกเว็บ

🇰🇷 한국어 · 🇬🇧 English

KORAIL 공공데이터 MCP

한국철도공사(KORAIL) 공공데이터를 AI에 연결하는 MCP(Model Context Protocol) 서버 모음입니다. 설치 후 Claude Desktop·Claude Code·Cursor·Antigravity·GitHub Copilot(CLI/VS Code) 등에서 자연어로 KORAIL 데이터를 조회할 수 있습니다.

✅ API 키 신청 불필요 — 전용 프록시 서버가 공공데이터 API 호출을 대신 처리합니다.

💻 로컬 설치형 (stdio) — 별도 서버 없이 개인 PC에서 직접 실행됩니다. Claude Desktop·Claude Code·Cursor·Antigravity·GitHub Copilot(CLI/VS Code) 등 로컬 MCP 클라이언트에 연결합니다. ChatGPT·Grok 같은 웹 서비스는 원격 연결이 필요합니다(하단 "그 밖의 방식" 참고).

📦 필요 디스크 공간 — 약 100MB (uv가 관리하는 Python과 패키지 포함)

👉 처음이신가요? 바로 아래 "제일 쉬운 방법"부터 시도해보세요. 98개 도구 전체 목록은 설치를 마친 뒤 필요할 때 참고하세요.


🤖 제일 쉬운 방법 — AI에게 그대로 시키기

Claude Code·Cursor·GitHub Copilot(CLI/VS Code)·Antigravity처럼 터미널 명령을 직접 실행할 수 있는 AI를 쓰고 있다면, 그 채팅창에 아래 문장을 그대로 붙여넣으세요.

https://github.com/lovelyquality/korail-mcp 의 README를 참고해서 이 MCP 서버를 설치하고 내 클라이언트에 연결해줘.

AI가 README를 직접 읽고 uv 설치 → korail-mcp 설치 → 클라이언트 설정 파일 등록까지 알아서 처리합니다. 완료되면 "서울역에 엘리베이터가 있나요?" 같은 질문으로 확인해보세요.

⚠️ ChatGPT·Grok처럼 로컬 명령을 실행하지 못하는 AI에서는 이 방법이 통하지 않습니다(왜인지는 하단 "그 밖의 방식" 참고). 그리고 AI가 중간에 막히거나, 애초에 이런 방식의 AI 도구가 없다면 → 바로 아래 "수동 설치"를 그대로 따라 하시면 됩니다.


⚙️ 수동 설치 (Windows · 2단계)

Python을 따로 설치하거나 저장소를 다운로드할 필요가 없습니다. `uv`가 필요한 것을 알아서 준비합니다.

1단계 — uv 설치 (최초 1회)

PowerShell을 열고 아래를 붙여넣습니다. 관리자 권한이 필요 없습니다.

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

설치 후 PowerShell을 새로 열고 uv --version이 출력되면 성공입니다.

2단계 — KORAIL MCP 설치 (최초 1회)
uv tool install --from git+https://github.com/lovelyquality/korail-mcp.git korail-mcp

마지막에 Installed 1 executable: korail-mcp 가 나오면 성공입니다.

⏳ 첫 설치는 1~3분 걸립니다(Python과 패키지를 받는 시간). 설치 후 실행은 약 5초입니다.

🔄 최신 버전으로 갱신 — uv tool upgrade korail-mcp 실행 후 클라이언트를 재시작하세요.

⚠️ 갱신 전에는 korail-mcp를 쓰는 클라이언트(Claude Desktop·Cursor·Antigravity·VS Code 등)를 먼저 완전히 종료하세요. 실행 중인 상태로 갱신하면 실행파일이 잠겨 있어 내부 패키지만 새 버전으로 바뀌고 실행파일은 그대로 남아 `ModuleNotFoundError: No module named 'gateway'`로 깨질 수 있습니다 (2026-08-24 실측으로 재현·확인). 이미 이 에러가 나면 korail-mcp.exe를 쓰는 모든 프로세스를 작업 관리자에서 종료한 뒤 uv tool install --from git+https://github.com/lovelyquality/korail-mcp.git korail-mcp --force로 재설치하세요. 여러 클라이언트가 같은 설치를 동시에 쓰는 것 자체는 문제 없습니다 — 충돌은 갱신하는 그 순간에만 일어납니다.


🔌 3단계 — 클라이언트 연결

아래 JSON을 클라이언트 설정 파일의 mcpServers 안에 넣고, `<사용자명>` 부분만 본인 윈도우 계정명으로 바꿉니다.

{

"mcpServers": { "korail-mcp": { "command": "C:\\Users\\<사용자명>\\.local\\bin\\korail-mcp.exe" } } }

💡 계정명을 모르면 PowerShell에 echo $env:USERNAME 을 입력하세요. 경로의 역슬래시는 JSON 규칙상 두 개(`\\`) 로 씁니다.

⚠️ 이미 다른 MCP 서버를 쓰고 있다면 `korail-mcp` 항목만 기존 mcpServers 안에 추가하세요(전체를 덮어쓰면 기존 서버가 사라집니다).

설정 파일 위치
클라이언트설정 파일
Claude Desktop%APPDATA%\Claude\claude_desktop_config.json
Claude Codeclaude mcp add 명령으로 등록 (아래 별도 안내)
CursorC:\Users\<사용자명>\.cursor\mcp.json
AntigravityC:\Users\<사용자명>\.gemini\antigravity\mcp_config.json
GitHub Copilot CLIcopilot mcp add 명령으로 등록 (아래 별도 안내)
VS Code (GitHub Copilot Chat)아래 별도 안내 참고 (JSON 형식이 다름)

Claude Desktop — 폴더가 없을 때

  1. 탐색기 주소창에 %APPDATA% 입력 → Enter
  2. Claude 폴더가 없으면 직접 만드세요
  3. 그 안에 claude_desktop_config.json 파일을 만들고 위 JSON을 넣으세요

AppData가 안 보이면 탐색기 → 보기 → 숨긴 항목을 체크하세요.

Claude Code — 실제 계정으로 도구 호출까지 실측 완료

터미널에서 쓰는 Claude Code CLI(claude, VS Code나 Claude Desktop 없이도 동작)로 실제 계정 붙여서 검증했습니다.

# 1) korail-mcp를 MCP 서버로 등록 (최초 1회)

claude mcp add korail-mcp -s user -- "C:\Users\<사용자명>\.local\bin\korail-mcp.exe"

2) 실제 질문 (해당 도구만 허용)

claude -p "korail-mcp 도구를 사용해서 서울역에 엘리베이터가 있는지 알려줘" --allowedTools "mcpkorail-mcp*"

✅ 2026-08-24 실측 결과 — 위 명령을 실제로 실행해서 실데이터 답변까지 확인했습니다:

`

서울역 — 엘리베이터 있음 (18개)

에스컬레이터: 23개 · 일반화장실: 있음 · 수유실: 있음 · 종합안내센터: 있음

(출처: 한국철도공사 공공데이터포털 편의시설정보, 실시간 API 기준)

`

claude mcp list로 등록 확인, claude mcp remove korail-mcp로 제거할 수 있습니다. 대화형으로 쓸 때는 --allowedTools 없이 실행하면 첫 도구 호출 시 승인 여부를 물어봅니다.

Cursor / Antigravity — 파일이나 폴더가 없을 때

.cursor 또는 .gemini\antigravity 폴더나 그 안의 설정 파일이 없다면 직접 만들면 됩니다.

  1. 탐색기 주소창에 %USERPROFILE% 입력 → Enter (본인 계정 폴더로 이동)
  2. 없는 폴더(.cursor 또는 .gemini\antigravity)를 새로 만드세요
  3. 그 안에 mcp.json(Cursor) 또는 mcp_config.json(Antigravity) 파일을 만들고 위 JSON을 넣으세요

Antigravity는 채팅에 위 JSON을 붙여넣고 "이 MCP 서버를 등록해줘"라고 요청하는 방법이 더 쉽습니다. 무료로 설치 가능하며, KORAIL MCP 연결에 별도 구독이 필요 없습니다.

GitHub Copilot CLI — 실제 계정으로 도구 호출까지 실측 완료

터미널에서 쓰는 공식 GitHub Copilot CLI(@github/copilot, VS Code 없이도 동작)로 실제 계정 붙여서 검증했습니다. 무료 플랜에서도 MCP 서버 연결이 됩니다 (다만 월 채팅 50회, 모델은 Claude Haiku 4.5 / GPT-5 mini로 제한).

⚠️ 사전 준비 — Node.js 22 이상이 필요합니다. node --version으로 확인하고, 없으면 nodejs.org에서 LTS 버전을 설치하세요(uv와 달리 자동으로 준비되지 않습니다).

# 1) 설치 (최초 1회)

npm install -g @github/copilot

2) korail-mcp를 MCP 서버로 등록 (최초 1회)

copilot mcp add korail-mcp -- "C:\Users\<사용자명>\.local\bin\korail-mcp.exe"

3) 실제 질문 (도구 자동 승인)

copilot -p "서울역에 엘리베이터가 있는지 korail-mcp 도구로 조회해줘" --allow-all-tools

✅ 2026-08-24 실측 결과 — 위 명령을 실제로 실행해서 GitHub Copilot이 get_urban_accessibility 도구를 스스로 호출하고 실데이터로 답변하는 것까지 확인했습니다:

`

● get_urban_accessibility (MCP: korail-mcp) · station_name: "서울역", facility_type: "elevator"

네, 서울역에는 엘리베이터가 있습니다! 총 17개의 엘리베이터가 설치되어 있습니다.

- 공항철도(AR): 9개 · 경의중앙선(KR): 1개 · 1호선(S1): 4개 · 4호선(S1): 3개

`

설정 파일은 ~/.copilot/mcp-config.json에 mcpServers.korail-mcp로 저장되며, copilot mcp list로 등록 확인, copilot mcp remove korail-mcp로 제거할 수 있습니다.

VS Code (GitHub Copilot Chat) — JSON 형식이 다릅니다 (서버 프로토콜은 실측, VS Code GUI 자체는 미확인)

VS Code 확장 형태의 Copilot Chat은 위 CLI와 별개 제품이라 설정 파일이 다릅니다. 최상위 키가 mcpServers가 아니라 `servers` 라서 위 JSON을 그대로 쓸 수 없습니다.

  1. 저장소(작업 폴더) 루트에 .vscode\mcp.json 파일을 만들고 아래 내용을 넣습니다. (모든 작업 폴더에서 쓰려면 명령 팔레트(Ctrl+Shift+P) → MCP: Open User Configuration 으로 열리는 %APPDATA%\Code\User\mcp.json 에 넣으세요.)
{

"servers": { "korail-mcp": { "command": "C:\\Users\\<사용자명>\\.local\\bin\\korail-mcp.exe" } } }

  1. 파일을 저장하면 상단에 나타나는 Start 버튼을 클릭합니다.
  2. Copilot Chat을 열고 Agent 모드를 선택 → 도구 아이콘에서 korail-mcp 98개 도구가 보이면 연결 완료입니다.

⚠️ 위 CLI 테스트로 서버 쪽(stdio 프로토콜·98개 도구·실데이터 응답)은 이미 검증됐지만, VS Code 화면에서 Start를 눌러 실제로 붙는지는 GUI 조작 도구가 없어 확인하지 못했습니다. 안 되면 알려주세요.

연결 후 반드시 — 클라이언트를 완전히 종료했다 다시 실행

ℹ️ Claude Code·GitHub Copilot CLI는 이 단계가 필요 없습니다 — 명령을 실행할 때마다 설정을 새로 읽습니다. 아래는 Claude Desktop·Cursor·Antigravity·VS Code처럼 창을 띄워두는 클라이언트에만 해당합니다.

창의 X를 눌러 닫아도 트레이(작업표시줄 오른쪽 `^` 안)에 계속 실행 중이라 설정이 적용되지 않습니다. → 트레이 아이콘 우클릭 → Quit / 종료 후 다시 실행하세요.

정상 연결되면 설정의 MCP 서버 목록에 korail-mcp가 running 으로 표시되고, 98개 도구를 쓸 수 있습니다.

💬 설치 확인 — 이렇게 물어보세요
서울역에 엘리베이터가 있나요?

2024년 간선철도 수송실적을 알려주세요. KTX 101 열차의 운행 계획을 알려주세요.

정상 응답이 오면 설치 완료입니다. 더 많은 사용 예시는 문서 하단의 "사용 예시" 섹션을 참고하세요.


📦 제공 서버 (총 11개 · 98개 도구)

서버도구 수제공 데이터
m-convenience6역사 편의시설·접근성·엘리베이터·위치 정보
m-stats15수송실적·발권 통계·이용유형·KTX 장기 통계
m-train-ops4열차 운행계획·운행이력
m-codebook4역코드·노선코드 조회
m-freight11화물·컨테이너·물류시설·품목·위험물
m-network8노선·역간거리·운임·역 선로제원
m-rolling-stock6차량 보유현황·형별제원·차종별 운행실적
m-voc-cs10고객서비스·정보공개
m-internal-svc14임대매장·사회공헌·인사 정보
m-procurement4자재그룹·G2B 품명·자재속성·대상장비
m-urban-rail16전국 도시철도 역사·노선·차량 시설·접근성·안전·환경·시각표
서버별 도구 상세 (클릭하여 펼치기)

m-convenience · 6개 도구 — 역사 편의시설

도구설명
get_station_facilities역 이름으로 편의시설 정보 조회
get_accessible_facilities역 이름으로 교통약자 편의시설 조회
list_stations_with_elevator엘리베이터가 설치된 역 목록 조회
get_station_facilities_detail역사 내외부 시설현황 조회
get_station_transfer_info역별 타 교통수단 환승현황 조회
get_station_location역 위치(좌표) 정보 조회

m-stats · 15개 도구 — 여객·화물 수송통계

도구설명
get_mainline_station_per간선열차 역별 승하차 통계
get_mainline_route_per간선열차 노선별 이용인원 통계
get_wide_rail_station_per광역철도 역별 승하차 통계
get_wide_rail_route_per광역철도 노선별 이용인원 통계
get_mainline_distance_per간선열차 거리별 이용인원 통계
get_mainline_model_per간선열차 차량별 이용인원 통계
get_mainline_day_of_week_per간선열차 요일별 이용인원 통계
get_mainline_grade_per간선열차 객실별 이용인원 통계
get_mainline_ticketing_stat간선열차 발권유형 통계
get_mainline_person_distance간선열차 노선별 인거리 통계
get_ktx_long_term_statsKTX 장기 통계
get_mainline_carriage간선 여객열차 수송실적 조회
get_wide_area_carriage광역 여객열차 수송실적 조회
get_freight_carriage화물열차 수송실적 조회
get_transport_stat_codes수송실적 통계 코드정보 조회

m-train-ops · 4개 도구 — 열차 운행

도구설명
get_train_codes열차운행 코드정보 조회
get_train_run_plan여객열차 운행계획 조회
get_train_run_info여객열차 실제 운행정보 조회
get_train_run_history차세대예약발매 열차 운행내역 조회

m-codebook · 4개 도구 — 역·노선 코드

도구설명
search_station역명으로 역코드·영문명·지역본부 통합 조회
decode_station_code역코드로 역명 조회
search_route노선명으로 노선코드 조회
list_stations_by_region지역본부명으로 관할 역 목록 조회

m-freight · 11개 도구 — 화물·물류

도구설명
search_freight_code내적화물코드 검색
decode_freight_code내적화물분류코드 단건 디코딩
search_container_record컨테이너 적재 이력 조회
list_freight_work_lines화물적하작업 전용 작업선 정보
list_standard_loading_time표준 적하시간 마스터 조회
search_loading_time_adjustment적하시간 조정 이력 조회
search_consignment_change수탁변경요금 검색
search_consignment_change_per_wagon수탁변경요금 화차별 조회
get_logistics_facility물류시설 정보 통합 조회
get_freight_items화물 품목정보 조회
get_hazardous_cargo위험물 정보 조회

m-network · 8개 도구 — 노선·거리·운임

도구설명
search_operation_patterns전국 철도 운행계통 검색
get_station_distance두 역 간 최단 운행거리 조회
get_freight_minimum_fare화물 운송 최저운임 기준 조회
get_freight_rate철도 화물 임율 정보 조회
get_segment_info철도 전동차 세그먼트 정보 조회
get_operation_distance노선별 역간 운행거리 조회
get_ktx_stationsKTX 노선별 역 정보 조회
get_station_track_info역별 선로·시설 상세 정보 조회

m-rolling-stock · 6개 도구 — 철도차량

도구설명
get_train_type_specs동력차 형별제원 조회
get_rolling_stock_by_year연도별 차량보유현황 조회
get_wagon_by_weight_class화차 자중별 보유현황 조회
get_wagon_by_load_capacity화차 적재하중별 보유현황 조회
get_maintenance_equipment철도차량 검수용 기계 보유현황 조회
get_train_operation_by_type차종별 연간 운행실적 조회

m-voc-cs · 10개 도구 — 고객서비스·정보공개

도구설명
get_customer_satisfaction_stats고객의소리 만족도 일별 통계
get_consultation_types철도 고객센터 상담유형 코드 조회
get_consultation_departments철도 고객센터 담당 부서 목록 조회
get_advance_disclosure홈페이지 사전정보공표 목록 조회
get_advance_disclosure_detail사전정보공표 세부 내역 조회
get_advance_disclosure_files사전정보공표 첨부파일 목록 조회
get_info_disclosure_dept정보공개 담당 부서 목록 조회
get_info_disclosure_codes정보공개 시스템 공통코드 조회
get_homepage_deptKORAIL 홈페이지 부서 정보 조회
get_homepage_positionKORAIL 홈페이지 직책 코드 조회

m-internal-svc · 14개 도구 — 임대·사회공헌·인사

도구설명
get_lease_stores역사 내 임대매장 운영정보 조회
get_lease_codes임대 시스템 코드 조회
get_leased_assets임대자산 현황 조회
get_dormitory_longterm_codes직원숙사 장기예약 사유 코드 조회
get_social_funds사회공헌 펀드 종류 조회
get_social_volunteer_fields사회공헌 봉사 분야 코드 조회
get_social_donations사랑의 성금 사용 내역 조회
get_social_volunteer_matching봉사활동 매칭 지출 내역 조회
get_social_org사회공헌 포털 조직정보 조회
get_support_facilities사옥 내 부대시설 목록 조회

| g

ตัดมาแค่ช่วงต้น อ่านต่อใน repo

ถาม PopNx