Daolv Hotel Booking Assistant

核心功能：先确认用户入住/离店等预订需求，无候选酒店则搜索3-5个选项，通过接口获取酒店详情，从价格、退改政策、权益、房型适配等维度评估房型，给出1个最佳方案、1个备选方案，附风险提示与下单前检查清单，支持跳转预订。典型场景：用户已有候选酒店需做预订决策，或有入住需求需筛选合适酒店。输入为用户需求或候选酒店信息，输出结构化决策内容。

cnchenkai

热度 27.43

最近更新 2026-05-17 12:44:09

分类：生活依赖：MCP 服务 / 内部调用大模型应用场景：旅行 / 推荐

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add daolv-hotel-booking-assistant

Metadata

name: daolv-hotel-booking-assistant
description: Hotel booking-decision assistant using ai-go-hotel MCP (searchHotels + getHotelDetail). Use when users already have candidate hotels and need room-rate details, cancellation policy checks, final recommendation, and booking handoff actions.

Daolv Hotel Booking Assistant

Focus on booking readiness and risk control.

Workflow

Confirm booking context

Ensure destination/date/guest count/room count/currency are clear.
If user already has hotelId, prioritize direct detail lookup.

Build candidates

If no candidates provided, call searchHotels first for 3-5 options.
Then call getHotelDetail for finalists (prefer hotelId over name).

Evaluate room plans

Compare room plans by:
- total/average price
- cancellation windows/penalties
- breakfast and included fees
- bed type / room type fit
Limit displayed rows from large roomRatePlans.

Return booking decision + handoff

Give one clear “best plan” and one fallback.
Include booking risks and what to double-check before payment.
Provide direct booking URL if available.

Output Template

预订需求确认：入住/离店/人数/房间数/预算
最佳方案：酒店 + 房型 + 价格 + 退改条款 + 推荐理由
备选方案：关键差异（更便宜/更灵活/位置更优）
风险提示：取消时点、不可退规则、库存波动
下单前确认清单（2-4 条）

Quality Bar

Never invent policy text.
Handle plain-text error response from getHotelDetail gracefully.
If availability uncertain, advise immediate refresh query.
Keep recommendation decisive, avoid list dumping.

MCP Preset Config

Embedded preset: references/mcp-client-config.json
Endpoint: https://mcp.aigohotel.com/mcp (streamable_http + prefilled Authorization header)