travelmind-mcp
MCP server for multi-agent travel planning, orchestrating parallel expert calls to generate structured itineraries and persist them to SQLite.
README
TravelMind Agent Lab
一个可以直接运行、适合逐文件学习的多 Agent 旅行规划项目。默认 demo 模式不需要
API Key；主 Agent 会并行调用航班、酒店、活动三个专家，读取本地旅行知识，最后生成
结构化方案并保存到 SQLite。
你会在这里学到什么
- LangGraph 状态图与主 Agent 编排
asyncio.gather并行专家调用- Pydantic 请求/响应模型
- FastAPI、后台任务与 SSE 事件流
- 本地 RAG、SQLite 会话持久化
- DeepSeek V4 Pro 的 OpenAI-compatible API
- 可选 Redis 任务状态、PostgreSQL 报告存储、FastMCP 工具
3 分钟启动
uv sync --extra dev
Copy-Item .env.example .env
uv run travelmind-api
浏览器打开 http://127.0.0.1:8000/docs，调用 POST /api/v1/plan：
{
"query": "帮我规划一次四天的巴黎旅行",
"origin": "Shanghai",
"destination": "Paris",
"days": 4,
"budget": 12000,
"currency": "CNY",
"thread_id": "lesson-001"
}
默认结果是可重复的模拟数据，目的是先学清 Agent 数据流。测试命令：
uv run pytest
uv run ruff check .
切换 DeepSeek
在 .env 中填写：
APP_MODE=deepseek
DEEPSEEK_API_KEY=你的密钥
DEEPSEEK_MODEL=deepseek-v4-pro
DeepSeek 只负责润色基于工具结果的摘要，航班、酒店和活动仍使用演示 Provider，避免
模型虚构价格。以后可在 providers.py 中替换为真实 Amadeus/地图 API。
可选基础设施
docker compose -f docker/compose.yml up -d
uv sync --extra redis --extra postgres --extra mcp
REDIS_URL=redis://localhost:6379/0：任务状态放入 Redis。DATABASE_URL=postgresql://travelmind:travelmind@localhost:5432/travelmind： 报告放入 PostgreSQL。uv run travelmind-mcp：把plan_trip暴露为 MCP 工具。
没有 Redis/PostgreSQL 时会自动使用内存任务表和 SQLite，因此核心项目始终能启动。
学习顺序
models.py：理解输入、专家结果和最终报告。agents.py：看三个专家如何并行工作。graph.py：看主 Agent 如何连接检索、专家和汇总节点。service.py：看图执行和持久化。api.py：看 HTTP、后台任务和 SSE。storage.py、jobs.py：再学习数据库与 Redis 适配器。
项目边界
这是可运行的学习版，不是出票系统。演示价格不应被用于真实预订。真实供应商 API、 认证、限流、退款和支付属于后续工程阶段。
来源与许可证说明见 UPSTREAM.md。
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.