← Back to Skills Marketplace
travel-skills

全球酒店搜索与推荐

by 旅行技能 · GitHub ↗ · v1.5.0 · MIT-0
cross-platform ⚠ suspicious
43
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install global-hotel-search-recommend
Description
1次调用完成酒店搜索+推荐+退改解读｜零配置即装即用｜自动识别场景｜含预订链接和退改政策。
README (SKILL.md)

name: global-hotel-search-recommend description: | 1次调用完成酒店搜索+推荐+退改解读｜零配置即装即用｜自动识别场景｜含预订链接和退改政策。One-call hotel search + recommendation + cancellation policy — zero config, auto scene detection, booking links included. tags: [酒店推荐, 酒店搜索, 酒店预订, 酒店查询, 订酒店, 亲子酒店, 商务酒店, 度假酒店, 便宜酒店, hotel, travel, booking, recommendation]

全球酒店搜索与推荐

⚡ 1次调用完成搜索→详情→退改解读→推荐理由 · 自动场景识别 · 3档价格推荐 · 极省Token · 零配置即装即用

快速入门

3个开场白示例，复制即用：

  1. "带娃去三亚住哪里好" → 自动识别亲子场景
  2. "出差去上海，住外滩附近" → 自动识别商务场景
  3. "三亚有什么便宜的酒店" → 自动识别背包场景

执行命令格式：

python SKILL_DIR/scripts/hotel_search_recommend.py --destination "目的地" --query "用户原始查询" --check-in "入住日期" --check-out "退房日期"

其中 SKILL_DIR 替换为变量 CODEBUDDY_SKILL_DIR 的值。

【重要约束】

  1. 必须调用真实API — 通过云端代理查询RollingGo实时酒店数据，绝不编造或缓存价格
  2. 场景自动识别 — 从用户原始查询中自动检测商务/亲子/度假/背包/通用场景，智能匹配筛选标签
  3. 3档价格推荐 — 搜索20家→按价格分性价比/品质/豪华3档各4家=12家，一次覆盖全价位
  4. 预算感知 — 用户提预算时自动只返回预算内酒店，不提预算则覆盖全价位
  5. 退改政策解读 — 自动将原始JSON退改规则转为可读中文（如"入住前7天前免费取消"）
  6. 禁止在不确定时编造价格 — 如无数据则明确告知，绝不虚构

核心能力

  1. 1次调用全流程：场景识别→参数补全→搜索→详情→退改解读→推荐理由，极省Token
  2. 5种场景自动识别：商务(💼)/亲子(👨‍👩‍👧)/度假(🌴)/背包(🎒)/通用(🏨)，从自然语言自动检测
  3. 参数智能补全：入住人数/儿童/晚数/星级/价格，从用户一句话推断所有参数
  4. 3档价格分选：性价比之选(💰) / 品质推荐(🏨) / 豪华体验(✨)，每档4家
  5. 退改政策解读：原始JSON→"入住前3天前免费取消；之后取消扣¥553"
  6. 推荐理由生成：基于场景+标签+价格档位自动生成（如"有儿童乐园，有家庭房"）
  7. 74标签库：8大类74个标签，场景×城市双维映射（海滨城市优先推荐"靠近海滩"）
  8. 搜索3级降级：去required标签→去距离→最简搜索，保证有结果

能力边界

能做：

  • 搜索全球酒店（国内+海外），返回实时价格和预订链接
  • 自动识别差旅场景并智能筛选
  • 解读退改政策，生成推荐理由

不能做：

  • 不能查询实时房态（是否有余房以预订页面为准）
  • 搜索结果仅含酒店封面图，不能提供更多实景图片

使用方式

基本用法

python SKILL_DIR/scripts/hotel_search_recommend.py \
  --destination "三亚" \
  --query "带娃去三亚住几天" \
  --check-in 2026-07-01 \
  --check-out 2026-07-03

高级用法（指定场景和筛选条件）

python SKILL_DIR/scripts/hotel_search_recommend.py \
  --destination "上海" \
  --scene 商务 \
  --check-in 2026-07-15 \
  --check-out 2026-07-16 \
  --max-price 800 \
  --preferred-brand "万豪,希尔顿" \
  --required-tag "免费WiFi"

参数说明

参数 必填 说明
--destination ✅ 目的地城市/区域/地标
--query 推荐 用户原始查询，用于场景自动检测和参数补全
--scene ❌ 场景：商务/亲子/度假/背包/通用，默认通用
--check-in ❌ 入住日期 YYYY-MM-DD，默认明天
--check-out ❌ 退房日期，默认根据场景推断
--guests ❌ 入住人数
--max-price ❌ 每晚价格上限（元）
--star-rating ❌ 星级：数字或中文（经济型/舒适型/高档型/豪华型）
--preferred-brand ❌ 偏好品牌，逗号分隔
--required-tag ❌ 必须标签（硬约束），逗号分隔
--preferred-tag ❌ 偏好标签（软约束），逗号分隔
--excluded-tag ❌ 排除标签，逗号分隔
--min-room-size ❌ 最小房间面积（㎡）
--max-distance ❌ 距离上限（米）
--country-code ❌ 国家代码（CN/JP/TH/KR/US等）

场景识别规则

场景 图标 关键词 智能标签策略
商务 💼 出差/商务/办公/会议 必须商务酒店；优选WiFi+健身房
亲子 👨‍👩‍👧 带娃/亲子/家庭/小孩 必须亲子酒店；优选儿童乐园+家庭房
度假 🌴 度假/蜜月/温泉/海边 优选度假酒店+泳池+SPA
背包 🎒 穷游/学生/便宜/经济 优选性价比+免费WiFi
通用 🏨 其他 无标签过滤

输出格式示例

👨‍👩‍👧 亲子推荐 · 三亚 · 2026-07-01(2晚) · ¥250-1903/晚

💰 性价比之选 ¥250-662/晚

1. 君亭酒店（三亚大东海店）⭐4.0 💰¥488/晚
   📍 大东海海花路13号 | 📏300m
   💬 步行至海滩仅3分钟，距离鹿回头景区约5分钟...
   🏷️ 临近商场 支持银联卡 快速办理入住
   💡 性价比之选，有儿童乐园
   🔄 退改：入住前7天前免费取消；之后取消扣¥488
   🔗 https://rollinggo.cn/pages/hotel/detail/index?id=1089166

反模式

  1. ❌ 多次调用不同场景参数尝试获取更多结果 → 1次调用已包含3级降级和智能补全
  2. ❌ 自行组合筛选条件反复搜索 → 直接传用户原始query，工具自动补全
  3. ❌ 忽略退改政策信息 → 退改政策已解读，必须展示给用户
  4. ❌ 编造酒店名称或价格 → 所有数据来自真实API，无结果时明确告知
Usage Guidance
Install only if you are comfortable sending hotel-search text, travel dates, occupancy, and possible family or business-travel details to the skill author's remote proxy. The publisher should rotate/remove the embedded token, clarify the privacy/data-retention behavior, and document the remote service more plainly before this is treated as low-risk.
Capability Assessment
ℹ Purpose & Capability
The script is purpose-aligned: it searches hotel data, interprets cancellation policies, and formats recommendations through a RollingGo/Tencent Cloud proxy.
ℹ Instruction Scope
Runtime instructions are narrow and user-directed via CLI parameters, but the documentation does not clearly explain that original queries, dates, occupancy, and child-related details are sent to the remote proxy.
✓ Install Mechanism
No package installation, startup hooks, or install-time execution were found; the artifact contains a SKILL.md and one Python script.
⚠ Credentials
Network access to a proxy is necessary for real-time hotel search, but the script hardcodes a shared proxy token while claiming the script contains no API keys, creating an avoidable credential and transparency concern.
✓ Persistence & Privilege
No persistence, local file reads or writes, shell execution, privilege changes, background workers, or local credential/profile access were found.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install global-hotel-search-recommend
  3. After installation, invoke the skill by name or use /global-hotel-search-recommend
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.5.0
global-hotel-search-recommend v1.5.0 - Documentation updated, expanding or clarifying usage details in SKILL.md. - No code-level changes to features or logic, only documentation revised.
v1.4.0
global-hotel-search-recommend v1.4.0 - 更新能力边界说明：搜索结果仅含酒店封面图，无法提供更多实景图片 - 精简描述和标签，强调零配置开箱即用、自动场景识别和价格推荐3档组合 - 移除与“酒店比价”技能的对比模块 - 文档内容整体裁剪和优化，更聚焦主要用法和限制，明确不可比价和不可查实时房态
v1.3.0
**Major update: Adds comprehensive auto scene detection, 3-tier price recommendations, and enriched outputs.** - Now supports one-call hotel search, recommendation, and cancellation policy explanation. - Automatically detects business, family, vacation, backpacking, or general travel scenarios from user query. - Recommends 12 hotels in 3 price tiers (4 per tier); includes booking links and readable cancellation policies. - Uses real-time data; never fabricates or caches prices. - Generates scene-specific recommendation reasons and covers budget-aware filtering when specified. - Enforces strict constraints and clear result limits; clearly states if no real data is available.
Metadata
Slug global-hotel-search-recommend
Version 1.5.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is 全球酒店搜索与推荐?

1次调用完成酒店搜索+推荐+退改解读｜零配置即装即用｜自动识别场景｜含预订链接和退改政策。 It is an AI Agent Skill for Claude Code / OpenClaw, with 43 downloads so far.

How do I install 全球酒店搜索与推荐?

Run "/install global-hotel-search-recommend" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 全球酒店搜索与推荐 free?

Yes, 全球酒店搜索与推荐 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 全球酒店搜索与推荐 support?

全球酒店搜索与推荐 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 全球酒店搜索与推荐?

It is built and maintained by 旅行技能 (@travel-skills); the current version is v1.5.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463942 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259883 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200739 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191401 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190375 · by oswalpalash

💬 Comments