← Back to Skills Marketplace
lizhijun

LovTrip AI Travel Planner

by KLeo · GitHub ↗ · v1.1.0
cross-platform ⚠ suspicious
393
Downloads
0
Stars
2
Active Installs
2
Versions
Install in OpenClaw
/install lovtrip-travel-planner
Description
AI 行程规划 / AI Travel Itinerary Planner — 智能生成多日旅行行程,支持景点搜索、预算计算、酒店航班。当用户需要旅行规划、生成行程、搜索景点酒店航班时使用。
README (SKILL.md)

AI 行程规划 / AI Travel Itinerary Planner

LovTrip (lovtrip.app) — AI 驱动的旅行规划平台,提供智能行程生成、景点推荐、预算管理。Web 版体验:lovtrip.app/planner

使用 AI + 高德 API 生成完整的多日旅游行程,支持景点搜索、预算计算、酒店推荐、航班查询。

Setup / 配置

{
  "mcpServers": {
    "lovtrip": {
      "command": "npx",
      "args": ["-y", "lovtrip@latest", "mcp"],
      "env": {
        "AMAP_API_KEY": "your-amap-api-key",
        "OPENROUTER_API_KEY": "your-openrouter-api-key"
      }
    }
  }
}

三步规划流程

第 1 步 — 信息完整度检查(每次必做)

用户提到旅游/旅行/行程时,检查以下 5 项:

项目 示例
① 具体城市(非国家) "成都"、"大阪"
② 天数 "3天"、"5天4晚"
③ 出行人数/同伴 "2人"、"和闺蜜"
④ 兴趣偏好 "美食"、"文化"、"自然"
⑤ 预算范围 "5000以内"、"穷游"

规则:

  • 缺少 ≥2 项 → 必须先追问,不得跳过直接生成
  • 缺少 1 项或全齐 → 直接进入第 2 步

判断示例:

  • "想去日本玩5天" → 缺城市/人数/兴趣/预算(缺4项)→ 追问
  • "大阪3天美食游2人预算5000" → 全齐 → 直接生成
  • "成都3天想吃火锅看熊猫" → 缺人数/预算(缺2项)→ 追问

第 2 步 — 生成行程

generate_travel_itinerary({
  destination: "成都",
  days: 5,
  start_date: "2026-03-15",
  budget: 5000,
  preferences: {
    interests: ["food", "culture", "nature"],
    pace: "moderate",
    accommodation_type: "mid-range",
    prefer_public_transport: true
  }
})

第 3 步 — 可选补充工具

仅在用户明确需要时调用,不要主动调用

工具 触发条件 说明
check_weather 用户要求查天气 目的地天气预报
calculate_travel_budget 用户要求预算明细 详细费用分解
generate_map_links 用户要求地图链接 景点地图链接
search_attractions 用户要求更多景点 搜索城市景点
search_hotels 用户要求酒店推荐 酒店搜索 + 预订
search_flights 用户要求航班信息 航班搜索 + 价格

工具列表 (6 Tools)

核心工具

generate_travel_itinerary — AI 生成行程

参数 类型 必需 说明
destination string 目的地城市
days number 旅行天数 (1-30)
start_date string 开始日期 YYYY-MM-DD
budget number 总预算(元)
preferences object 兴趣、节奏、住宿、交通偏好
mapProvider string amap / google / auto

补充工具

search_attractions — 搜索景点

参数 类型 必需 说明
city string 城市名称
keywords string 搜索关键词
types array 景点类型过滤
min_rating number 最低评分
max_results number 最大返回数量,默认 10
sort_by string rating / popularity / distance

calculate_travel_budget — 计算预算

参数 类型 必需 说明
destination string 目的地
days number 天数
budget_total number 总预算
accommodation_cost number 住宿费
transportation_cost number 交通费
daily_food_budget number 每日餐饮
activities array 活动费用列表

search_hotels — 搜索酒店

参数 类型 必需 说明
city string 城市
check_in string 入住日期
check_out string 退房日期
guests number 住客数
min_price / max_price number 价格范围
star_rating number 星级 (3/4/5)
location_preference string 位置偏好

search_flights — 搜索航班

参数 类型 必需 说明
origin string 出发城市
destination string 目的城市
departure_date string 出发日期
return_date string 返程日期
passengers number 乘客数
cabin_class string economy / premium_economy / business / first
max_price number 最高价格
direct_only boolean 仅直飞

optimize_daily_route — 日程路线优化

参数 类型 必需 说明
start_point string 起点地址
waypoints array 途径点列表
city string 城市
end_point string 终点地址
start_time string 出发时间 HH:mm
travel_mode string transit / driving / walking

关键规则

  1. 工具调用限制: generate_travel_itinerary(1次)+ 可选工具(最多2次)= 最多 3 次
  2. 禁止为每个景点单独调用工具 — 会导致调用爆炸
  3. 工具参数不含 emoji{ "destination": "成都" } 而非 { "destination": "📍 成都" }
  4. 多轮对话调整: 记住上文,根据用户修改重新生成行程
  5. 地图链接格式化: 从 JSON 中提取 URL 生成 Markdown 链接 [高德地图](url)

支持的兴趣标签

nature | culture | food | shopping | nightlife | adventure

支持的旅行节奏

  • relaxed: 每天 2-3 个景点,充足休息
  • moderate: 每天 3-4 个景点,适中节奏
  • fast: 每天 4-5 个景点,紧凑安排

在线体验


Powered by LovTrip — AI Travel Planning Platform

Usage Guidance
This skill appears to be a legitimate travel planner, but there are transparency and supply-chain concerns you should address before installing or running it: - Do not run `npx lovtrip@latest mcp` blindly. `npx ...@latest` downloads and executes the latest package from npm and can change behavior over time. Prefer a pinned version (e.g., [email protected]) and inspect the package source first. - The SKILL.md asks for AMAP_API_KEY and OPENROUTER_API_KEY but the registry metadata doesn't declare these — ask the publisher to update the manifest to list required env vars and explain how credentials are used and stored. - Verify the package provenance: check the lovtrip npm package page and the upstream repository (source code, maintainer, recent commits, and issues) before giving it API keys or running it in your agent environment. - If you must test, run the npx command in an isolated environment (ephemeral container) with only the minimal API keys and network access required, and with logging/monitoring in place. - Consider least-privilege API keys (restrict by referrer/IP and scope) for mapping and routing services, and avoid providing broad account credentials. If the publisher cannot provide a source repo and a pinned release and cannot update the registry metadata to declare required environment variables, treat the skill as higher risk and avoid installing it.
Capability Analysis
Type: OpenClaw Skill Name: lovtrip-travel-planner Version: 1.1.0 The skill bundle is a legitimate travel planning assistant that integrates with the LovTrip platform via an MCP server. It defines standard tools for itinerary generation, hotel/flight searches, and budget calculations, with clear instructions for the AI agent to follow a structured planning process. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found in the code or documentation.
Capability Assessment
Purpose & Capability
The SKILL.md clearly intends to use external mapping/routing APIs (高德/AMAP and OpenRouter) and a lovtrip CLI via `npx lovtrip@latest mcp`. However the registry metadata lists no required env vars or install requirements. Asking for AMAP_API_KEY and OPENROUTER_API_KEY in the setup is coherent with a travel planner, but those credentials should be declared in the skill metadata; their absence is an incoherence and a transparency issue.
Instruction Scope
Runtime instructions include a 'Setup' snippet that runs `npx ... lovtrip@latest mcp` (i.e., fetch-and-execute via npx) and references env vars. The skill allows Bash as a tool, so an agent could actually execute that command. The SKILL.md otherwise stays within travel planner functionality, but the instruction to fetch and run external code at runtime expands scope in a way not reflected in the manifest and without guidance on safety or provenance.
Install Mechanism
There is no declared install spec in the registry, but the documentation instructs using `npx lovtrip@latest` to run a CLI. Using unpinned `@latest` via npx downloads code from the npm registry at runtime — a supply-chain risk. The registry should either include a vetted install spec or explicit, pinned instructions and a source repo to review; that is missing here.
Credentials
The SKILL.md example requires AMAP_API_KEY and OPENROUTER_API_KEY (and mentions Google API when using mapProvider=google), but the declared required env vars in the registry are empty. Requesting API keys for mapping/routing is reasonable for this purpose, but the omission from metadata and lack of guidance about scopes/permissions and storage is a proportionality and disclosure issue.
Persistence & Privilege
The skill does not request 'always: true' and is user-invocable only. That is reasonable. However, because it allows Bash and instructs running npx to fetch/execute a package, it gains the effective ability to run arbitrary code during a session if the agent follows the setup — a runtime privilege the user should be aware of. The skill does not claim to modify other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install lovtrip-travel-planner
  3. After installation, invoke the skill by name or use /lovtrip-travel-planner
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
Add lovtrip.app website links, online experience section
v1.0.0
Initial release: AI-powered multi-day travel itinerary generation
Metadata
Slug lovtrip-travel-planner
Version 1.1.0
License
All-time Installs 2
Active Installs 2
Total Versions 2
Frequently Asked Questions

What is LovTrip AI Travel Planner?

AI 行程规划 / AI Travel Itinerary Planner — 智能生成多日旅行行程,支持景点搜索、预算计算、酒店航班。当用户需要旅行规划、生成行程、搜索景点酒店航班时使用。 It is an AI Agent Skill for Claude Code / OpenClaw, with 393 downloads so far.

How do I install LovTrip AI Travel Planner?

Run "/install lovtrip-travel-planner" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is LovTrip AI Travel Planner free?

Yes, LovTrip AI Travel Planner is completely free (open-source). You can download, install and use it at no cost.

Which platforms does LovTrip AI Travel Planner support?

LovTrip AI Travel Planner is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created LovTrip AI Travel Planner?

It is built and maintained by KLeo (@lizhijun); the current version is v1.1.0.

💬 Comments