/ 目錄 / 演練場 / mcp-server-12306
← 全部伺服器 ↗ GitHub
● 社群 drfccv ⚡ 即開即用

mcp-server-12306

作者 drfccv · drfccv/mcp-server-12306

High-performance Python MCP server for 12306 — real-time tickets, prices, station search, train routes, and transfer planning.

mcp-server-12306 is a FastAPI-based MCP server providing real-time data from China's 12306 railway system. It offers 7 tools for ticket queries, price lookups, station search (Chinese/pinyin/abbreviation), train route stops, transfer planning, and time utilities. Supports both stdio (Claude Desktop) and Streamable HTTP modes, with Docker deployment and sub-second response times.

▶ 觀看演示 📦 安裝 💡 使用場景

為什麼要用

核心特性

即時演示

實際使用效果

mcp-server-12306-2.replay ▶ 就緒
0/0

安裝

選擇你的客戶端

~/Library/Application Support/Claude/claude_desktop_config.json  · Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "mcp-server-12306-2": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/drfccv/mcp-server-12306"
      ]
    }
  }
}

開啟 Claude Desktop → Settings → Developer → Edit Config。儲存後重啟應用。

~/.cursor/mcp.json · .cursor/mcp.json
{
  "mcpServers": {
    "mcp-server-12306-2": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/drfccv/mcp-server-12306"
      ]
    }
  }
}

Cursor 使用與 Claude Desktop 相同的 mcpServers 格式。專案級設定優先於全域。

VS Code → Cline → MCP Servers → Edit
{
  "mcpServers": {
    "mcp-server-12306-2": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/drfccv/mcp-server-12306"
      ]
    }
  }
}

點擊 Cline 側欄中的 MCP Servers 圖示，然後選 "Edit Configuration"。

~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "mcp-server-12306-2": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/drfccv/mcp-server-12306"
      ]
    }
  }
}

格式與 Claude Desktop 相同。重啟 Windsurf 生效。

~/.continue/config.json
{
  "mcpServers": [
    {
      "name": "mcp-server-12306-2",
      "command": "TODO",
      "args": [
        "See README: https://github.com/drfccv/mcp-server-12306"
      ]
    }
  ]
}

Continue 使用伺服器物件陣列，而非映射。

~/.config/zed/settings.json
{
  "context_servers": {
    "mcp-server-12306-2": {
      "command": {
        "path": "TODO",
        "args": [
          "See README: https://github.com/drfccv/mcp-server-12306"
        ]
      }
    }
  }
}

加入 context_servers。Zed 儲存後熱重載。

claude mcp add mcp-server-12306-2 -- TODO 'See README: https://github.com/drfccv/mcp-server-12306'

一行命令搞定。用 claude mcp list 驗證，claude mcp remove 移除。

使用場景

實戰用法： mcp-server-12306

How to compare train ticket prices for a journey with mcp-server-12306

👤 Budget-conscious travelers in China ⏱ ~5 min beginner

何時使用： You want to find the cheapest option for your train journey.

前置條件
  • mcp-server-12306 running — uvx mcp-server-12306 or Docker: docker run -p 8000:8000 drfccv/mcp-server-12306:latest
步驟
  1. Search tickets with prices
    Query tickets from Beijing to Guangzhou for tomorrow. Show all available trains with prices for each seat class.✓ 已複製
    → Table of trains with departure times, durations, and prices per seat class
  2. Get detailed pricing
    Show me the detailed pricing for train G79, including all seat types.✓ 已複製
    → Complete fare breakdown for the specific train

結果： Clear price comparison to choose the best value train.

注意事項
  • Prices may vary by date and availability — Check availability alongside prices — cheap seats sell out first

Plan a multi-city train trip across China

👤 Tourists planning a scenic route through multiple Chinese cities ⏱ ~15 min beginner

何時使用： You want to visit several cities and need to plan the train connections.

前置條件
  • mcp-server-12306 running — uvx mcp-server-12306
步驟
  1. Search stations
    Search for the main train stations in Shanghai, Hangzhou, and Suzhou. Show me the station codes.✓ 已複製
    → Station names and codes for each city
  2. Plan the route
    Find the fastest train connections for: Shanghai → Hangzhou → Suzhou → back to Shanghai, spending one day in each city. Show departure times and durations.✓ 已複製
    → Complete itinerary with train numbers and times

結果： A multi-city itinerary with all train connections planned.

注意事項
  • Small cities may only have one station — Use the station search to confirm available stations before planning

組合

與其他 MCP 搭配，撬動十倍槓桿

mcp-server-12306-2 + fetch

Check weather at your destination alongside train schedules

Find trains from Beijing to Chengdu this weekend and check the weather forecast for Chengdu.✓ 已複製

工具

此 MCP 暴露的能力

工具輸入參數何時呼叫成本
query_tickets from: str, to: str, date: str Search for available train tickets 1 API call
query_ticket_price train_no: str, from: str, to: str, date: str Get detailed pricing for a specific train 1 API call
search_stations keyword: str Search stations by name, pinyin, or abbreviation 1 API call
get_station_info station_code: str Get detailed station metadata 1 API call
query_transfer from: str, to: str, date: str Find optimal transfer routes 1 API call
get_train_route_stations train_no: str Get all stops along a train route 1 API call
get_current_time none Get current time for relative date calculations 0

成本與限制

運行它的成本

API 配額
Subject to 12306 public API rate limits
每次呼叫 Token 數
200–800 tokens per query
費用
Free — uses public 12306 API
提示
Use station search to confirm names before querying tickets to avoid wasted calls.

安全

權限、密鑰、影響範圍

憑證儲存： No credentials required
資料出站： All calls to 12306.cn official API

故障排查

常見錯誤與修復

No tickets found

12306 sells tickets 15 days in advance. Verify your date is within the booking window.

驗證： Try today's date to confirm connectivity
Station not recognized

Use search_stations first with pinyin or Chinese to find the exact station name.

驗證： search_stations with a broader keyword
Timeout during peak hours

12306 API is slow during peak ticket sales (holidays, Spring Festival). Retry after a few minutes.

驗證： Test with a simple station search

替代方案

mcp-server-12306 對比其他方案

替代方案何時用它替代權衡
12306-mcpYou prefer a JavaScript/npx-based server with simpler setupSimpler install but fewer features (no price queries or HTTP streaming)

更多

資源

📖 閱讀 GitHub 上的官方 README

🐙 查看未解決的 issue

🔍 瀏覽全部 400+ MCP 伺服器和 Skills