Uses Google Gemini 2.0 Flash as the underlying LLM for processing user requests.
Integrates LangChain v1.0 as the framework for building the agent and tool interactions.
Integrates LangGraph v1.0 for managing the agent's workflow using StateGraph.
Fetches real attraction data for Kyoto from Wikivoyage API.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@kyoto-assistant-mcp plan a 2-day Kyoto itinerary with temples and gardens"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
版本: v4.1 (LangChain v1.0 + LangGraph v1.0)
風格: 完全仿照維也納旅遊助理
✅ Wikivoyage API - 真實景點資料
✅ 營業檢查工具 - 模擬版本（可升級 Google Places）
✅ ReAct 策略 - Think → Act → Observe
✅ Planning 策略 - 動態調整行程
✅ LangChain v1.0 - 最新穩定 API
✅ LangGraph v1.0 - StateGraph 工作流
✅ 7 個 MCP 工具 - 完整協同
✅ Claude Desktop - 可重用 MCP Server
Related MCP server: wanderlog-mcp
# 建立虛擬環境
python -m venv venv
source venv/bin/activate # macOS/Linux
# 安裝套件
pip install -r requirements.txt# 複製環境變數模板
cp .env.example .env
# 編輯 .env，填入你的 API Keys# 測試 MCP Server（稍後實現）
python mcp_server/server.py
# 測試 CLI（稍後實現）
python main.py testkyoto-assistant/
├── src/ # Agent 相關
│ ├── agent/ # LangChain v1.0 Agent
│ ├── workflow/ # LangGraph v1.0 Workflow
│ └── cli/ # CLI 介面
├── mcp_server/ # MCP Server
│ └── tools/ # 7 個 MCP 工具
├── data/ # 資料檔案
├── output/ # 輸出檔案
└── tests/ # 測試腳本LLM: Google Gemini 2.0 Flash (免費)
框架: LangChain 1.0.x + LangGraph 1.0.x
MCP: Model Context Protocol 1.3.2
Python: 3.10+
# 規劃行程
python main.py plan --days 2 --budget 25000 --interests 寺廟
# 詢問問題
python main.py ask "清水寺現在營業嗎？"
# 互動模式
python main.py interactive✅ Wikivoyage API 整合
✅ LangChain v1.0 API
✅ LangGraph v1.0 StateGraph
✅ MCP Server 開發
✅ ReAct + Planning 策略
開發中... 🚧
This server cannot be deployed
Travel tools for AI agents: plan and edit real trips, search stays and tours, import travel videos.
Travel notebook your AI agent writes from your photos and remarks over MCP.
Stateless agentic tools over MCP: concept extraction, long-context, knowledge graph, planning.
Travel planner: create, edit, and explore trip itineraries from your Travolp AI assistant.
