← 返回 Skills 市场
adrianetti

Travel Search

作者 Adrian Ambrosetti · GitHub ↗ · v1.5.0 · MIT-0
cross-platform ✓ 安全检测通过
206
总下载
1
收藏
0
当前安装
10
版本数
在 OpenClaw 中安装
/install travel-search
功能描述
Find the best travel deals by searching and comparing flights, hotels, Airbnb stays, car rentals, and ferries across multiple providers simultaneously. Smart...
安全使用建议
This skill appears coherent with its travel-search description and mostly performs straightforward network calls to public MCP endpoints using curl. Before installing or invoking it: 1) Inspect scripts/mcp-call.sh to confirm it only wraps safe curl/json parsing and does not exfiltrate local files or secrets. 2) Verify the listed MCP endpoints (mcp.kiwi.com, mcp.skiplagged.com, etc.) are legitimate and accessible from your environment and that they accept unauthenticated MCP calls (some provider APIs may require keys or have usage policies). 3) Be aware the agent will make outbound HTTP requests and write temporary header files (examples use /tmp/*.txt); if you run agents with sensitive local credentials, ensure they are not exposed. 4) Optional providers (Airbnb via @openbnb/mcp-server-airbnb, fli) require separate, local installation (Node/Python) that you should install and review yourself. If you need higher assurance, ask the skill author for provenance (source repo/homepage) or perform a minimal test in a sandboxed environment first.
功能分析
Type: OpenClaw Skill Name: travel-search Version: 1.5.0 The travel-search skill is a comprehensive travel planning tool that integrates multiple providers (Kiwi, Skiplagged, Trivago, etc.) via the Model Context Protocol (MCP). It uses a helper script (scripts/mcp-call.sh) to facilitate JSON-RPC calls over HTTP using curl to legitimate-appearing service endpoints. The documentation is extensive and focuses entirely on travel-related tasks like flight searching, hotel comparison, and itinerary generation, with no evidence of malicious intent, data exfiltration, or unauthorized execution.
能力评估
Purpose & Capability
Name/description (search flights/hotels/ferries/etc.) match the declared runtime behavior: the SKILL.md documents calling MCP servers for Kiwi, Skiplagged, Trivago, Ferryhopper, optional local Airbnb/fli MCPs. Required binaries (only curl) and no env vars are proportionate to the stated purpose.
Instruction Scope
Instructions are focused on making JSON-RPC (MCP) calls via curl, parsing SSE and header files in /tmp (examples use curl -D /tmp/...). They do not ask for unrelated local secrets or global system reads. Note: examples show writing/reading temporary header files and using shell parsing (grep/awk), and the skill tells the agent to include booking deep links and explain trade-offs — all within expected scope for a travel aggregator.
Install Mechanism
No install spec — instruction-only skill. The repository includes a small helper shell script (scripts/mcp-call.sh) but there is no automatic download or archive extraction. This is the lower-risk model for skills; still inspect the helper script before executing.
Credentials
The skill declares no required environment variables, no credentials, and no config-path access. That is appropriate for a skill that performs unauthenticated HTTP calls to public MCP endpoints and for optional local MCP services the user must install separately.
Persistence & Privilege
The skill does not request always:true, does not modify other skills, and does not require persistent presence or elevated privileges. It relies on agent-run network calls only.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install travel-search
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /travel-search 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.5.0
Travel Intel: weather forecasts, visa reference, currency/tipping guide, local transport (airport+city) for major destinations, safety notes, power plugs, connectivity. Completes the full v1.0-v1.5 roadmap.
v1.4.0
Multi-city route optimizer: price matrix, geographic flow optimization, smart day allocation, open-jaw detection, hub city awareness, train/bus alternatives for short hops, cluster optimization for 5+ cities.
v1.3.1
Security fix: removed ignore-robots-txt override, removed npx/pipx auto-install patterns, removed ignoreRobotsText params. All optional providers now require explicit user installation. ClawHub-policy compliant.
v1.3.0
Airbnb integration: optional local MCP via npx, hotel-vs-Airbnb comparison workflow, value scoring for apartments/homes, and trip-planner support for families, groups, workcations, and longer stays.
v1.2.1
Guided intake: 7-question conversational questionnaire for trip planning. All in one message, all optional, auto-adapts to user language. Two modes: Quick (direct) and Guided (full planning).
v1.2.0
Smart Price & Value Engine: cross-provider comparison with value scoring (price×quality×convenience), price calendars, flexible date/destination optimization, deal detection, hidden cost flagging, trade-off explanations. The skill now finds THE best deal, not just the cheapest.
v1.1.2
Dimensional profiling: 5 core dimensions (budget/energy/social/structure/focus) replace fixed vibe lists. Infinite trip styles from natural language. 20+ interest categories as anchors. 8 diverse adaptation examples.
v1.1.1
Universal Trip Planner: 20+ travel vibes, smart inference, daily cost benchmarks for every region, travel party detection, pacing control, dietary awareness. Adapts to any traveler on Earth.
v1.1.0
v1.1: Trip Planner — full itinerary generation with flights + hotels + budget breakdown + day-by-day plan. Daily cost estimates by region.
v1.0.0
Initial release
元数据
Slug travel-search
版本 1.5.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 10
常见问题

Travel Search 是什么?

Find the best travel deals by searching and comparing flights, hotels, Airbnb stays, car rentals, and ferries across multiple providers simultaneously. Smart... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 206 次。

如何安装 Travel Search?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install travel-search」即可一键安装,无需额外配置。

Travel Search 是免费的吗?

是的,Travel Search 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Travel Search 支持哪些平台?

Travel Search 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Travel Search?

由 Adrian Ambrosetti(@adrianetti)开发并维护,当前版本 v1.5.0。

💬 留言讨论