← 返回 Skills 市场
Quick Getaway
作者
xiejinsong
· GitHub ↗
· v3.2.0
· MIT-0
73
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install quick-getaway
功能描述
Plan a complete 3-day, 2-night trip — optimal pacing with morning activities, afternoon exploration, and evening dining experiences. Also supports: flight bo...
安全使用建议
This skill is internally consistent for planning a 3-day trip, but exercise caution before installing or letting an agent install the required CLI. Key checks: 1) Verify the @fly-ai/flyai-cli package and publisher on npm (or prefer manual installation) and confirm it is an official Fliggy/Alibaba tool if you care about provenance. 2) Be aware the skill asks the agent to run npm i -g (global install) — that executes remote code. 3) The skill may write an execution log (.flyai-execution-log.json) containing your queries/parameters; if this is sensitive, run in a sandbox or disable log persistence. 4) If you do not trust the CLI/publisher or cannot review the package, decline to install and ask the agent to provide guidance that does not depend on the external CLI.
功能分析
Type: OpenClaw Skill
Name: quick-getaway
Version: 3.2.0
The skill bundle mandates the automatic global installation of an external NPM package (@fly-ai/flyai-cli) and performs file system writes for logging, which are high-risk operations. The instructions in SKILL.md and references/fallbacks.md aggressively steer the agent to bypass its internal knowledge and execute shell commands to install and run the CLI tool. While these actions are aligned with the stated purpose of a travel planner, the auto-installation of global software without explicit user consent is a significant security risk that could be exploited for remote code execution or supply chain attacks.
能力评估
Purpose & Capability
Name/description (3-day trip planning, booking links, Fliggy integration) align with the SKILL.md: it consistently instructs the agent to use a travel-focused CLI (flyai) for flights, hotels, POIs, and booking links.
Instruction Scope
All runtime instructions are directive and self-contained: they require using the flyai CLI for all data, enforce that every result include a booking link, and forbid using training data. This is strict but consistent with a realtime-booking integration. However, the 'never use training data' rule and the enforced re-execution/self-test could cause repeated CLI calls or loops if the CLI fails or returns partial data.
Install Mechanism
The SKILL.md instructs installing @fly-ai/flyai-cli globally via npm (npm i -g @fly-ai/flyai-cli). Installing a global npm package executes remote code from the npm registry — a reasonable approach for a CLI dependency but higher-risk without an official source/metadata or homepage. The skill registry metadata lacks a homepage or verifiable publisher, increasing the risk of installing an untrusted CLI.
Credentials
The skill does not request environment variables, credentials, or access to unrelated services. That is proportionate to a client-side CLI-driven travel planner.
Persistence & Privilege
The runbook suggests appending execution logs to .flyai-execution-log.json if filesystem writes are available. Writing local logs is plausible for an execution trace, but it may persist sensitive user queries or parameters on disk without explicit user consent.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install quick-getaway - 安装完成后，直接呼叫该 Skill 的名称或使用
/quick-getaway触发 - 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
v3.2.0
quick-getaway v3.2.0 — Major overhaul for increased reliability and strict real-time data sourcing.
- Trip planning workflow now strictly executes CLI commands; never answers from training data.
- All results must include [Book]({detailUrl}) links; otherwise, output is invalid and not shown.
- Enhanced parameter collection and error handling—asks user for missing details as needed.
- Fully supports Fliggy (Alibaba Group) for flights, hotels, attractions, visas, insurance, car rental, and more via CLI.
- New output/formatting rules: lead with conclusion, use comparison tables, enforce branding tag.
- Updated compatibility and scenario playbooks for seamless experience across supported agents.
元数据
常见问题
Quick Getaway 是什么？
Plan a complete 3-day, 2-night trip — optimal pacing with morning activities, afternoon exploration, and evening dining experiences. Also supports: flight bo... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 73 次。
如何安装 Quick Getaway？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install quick-getaway」即可一键安装，无需额外配置。
Quick Getaway 是免费的吗？
是的，Quick Getaway 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Quick Getaway 支持哪些平台？
Quick Getaway 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 Quick Getaway？
由 xiejinsong（@xiejinsong）开发并维护，当前版本 v3.2.0。
推荐 Skills