Token导航 LogoToken导航

Agent Skill · 研究检索

whisky-search(威士忌搜索)

whisky-search 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

whisky-search 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 24,940。

总安装 3,881周安装 165GitHub Stars 1下载 1,360核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.4

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:whisky-search(威士忌搜索)
来源仓库:https://github.com/fengci/whisky-search
安装命令:
openclaw skills install whisky-search
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install whisky-search

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

从 WhiskySpace 搜索并检索威士忌信息。当用户询问威士忌酒瓶、品牌、酿酒厂、价格、评级、地区等时,请使用此技能。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install whisky-search;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

whisky-search 怎么安装?

在终端执行 openclaw skills install whisky-search。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

whisky-search 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

whisky-search 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

whisky-search 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

whisky-search 用的人多吗?

公开数据是累计安装 3,881 次,近一周安装 165 次,下载 1,360 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

whisky-search 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
whisky-search

威士忌搜索 · 本页

3,8811接触敏感数据MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
whisky-search
description
>-
languages
English, Chinese (zh), Japanese (ja), Spanish, French, German,

Whisky Search

Use WhiskySpace API responses as the source of truth. Prefer short, factual answers built from returned fields.

Non-negotiable rules

  • Use only API-returned data plus the canonical website URL rules in this skill.
  • Do not infer site routes, official websites, histories, tasting notes, prices, or descriptions from naming patterns.
  • If a required field is missing, say it is unavailable or not provided. Do not guess.
  • Distinguish clearly between a WhiskySpace page URL and a brand's official website.
  • If the user asks for a WhiskySpace link, return the page URL. If the user asks for the official website, use only the API website field.
  • If the user asks where a URL came from, explain that it uses the canonical route in this skill plus the identifier returned by the API.
  • Never describe a canonical URL as guessed, inferred, predicted, or assumed.

Canonical website URL rules

  • Whisky detail page: https://www.whiskyspace.com/whiskies/{slug_or_id}
  • Brand page: https://www.whiskyspace.com/brand/{id}
  • If the API response includes a full website URL field for the requested WhiskySpace page, use that exact value.
  • Otherwise, whisky pages use the returned slug; if slug is missing, use the numeric ID.
  • Brand pages use the numeric id only. Never use slug to build a brand page URL.
  • Never output /brand/{slug}, /brands/{slug}, or any undeclared route.
  • If the user asks for a brand page URL and the current payload has only slug, call GET https://www.whiskyspace.com/api/brands/{slug} first and then use the returned numeric id.
  • If the required identifier is still missing after the relevant lookup, say the WhiskySpace page URL is unavailable.

Request routing

  1. Search, compare, recommend, or filter whiskies:

Use GET https://www.whiskyspace.com/api/search.

  1. Show one whisky's details:

Use GET https://www.whiskyspace.com/api/whiskies/{slug_or_id}.

  1. List brands:

Use GET https://www.whiskyspace.com/api/brands.

  1. Show one brand's details:

Use GET https://www.whiskyspace.com/api/brands/{identifier}.

  1. Show filter choices:

Use GET https://www.whiskyspace.com/api/search/filters.

  1. User asks only for a page address, link, or URL:

Fetch only what is needed to obtain the canonical identifier, then return the URL directly.

  1. User asks where a URL came from:

Answer with the canonical route plus the API identifier used. Keep it to one or two sentences.

API endpoints

Search whiskies

GET https://www.whiskyspace.com/api/search

Common parameters:

  • q
  • region, country, type, cask_type
  • min_age, max_age
  • min_abv, max_abv
  • min_rating, max_rating
  • sort=rating|name|age|price|newest
  • per_page (max 50)

Get whisky details

GET https://www.whiskyspace.com/api/whiskies/{slug_or_id}

List brands

GET https://www.whiskyspace.com/api/brands

Common parameters:

  • country
  • region
  • sort=name|founded_year|whisky_count
  • per_page

Get brand details

GET https://www.whiskyspace.com/api/brands/{identifier}

identifier can be a slug or numeric ID for the API request. Use the returned numeric id for the website brand page URL.

Get filter options

GET https://www.whiskyspace.com/api/search/filters

Output templates

Search results

Show the top 5 to 10 matches unless the user asks for more.

For each whisky include:

  • Name
  • Age if present
  • Volume and ABV if present
  • Rating and review count if present
  • Brand, region, country if present
  • Price with recorded date if present
  • Cask types if present
  • WhiskySpace link: https://www.whiskyspace.com/whiskies/{slug_or_id}

When there are many matches, say: Found N results, showing top M.

Whisky details

Include:

  • Name, brand, category
  • Age, ABV, volume
  • Rating and review count
  • Description if present
  • Cask types
  • Prices with currency, market, and recorded date
  • WhiskySpace page URL

Brand list

For each brand include:

  • Name
  • Founded year if present
  • Region and country if present
  • Whisky count if present
  • WhiskySpace brand page URL: https://www.whiskyspace.com/brand/{id}

Brand details

Include:

  • Name
  • Description or history if present
  • Region and country
  • Founded year if present
  • Official website only if the API returns website
  • WhiskySpace brand page URL: https://www.whiskyspace.com/brand/{id}

Link-only response

When the user asks only for an address, link, or URL, keep the answer minimal.

Example:

White Horse brand page:
https://www.whiskyspace.com/brand/492

Do not add extra tasting notes, history, or route speculation.

URL provenance response

When the user asks where the URL came from, use this pattern:

It uses the canonical WhiskySpace brand route `/brand/{id}` defined in this skill and the API-returned brand id `492`. I did not use the slug to build the page URL.

Adapt the route and identifier to the actual case.

Anti-patterns

Never do any of the following:

  • The brand page should be /brand/{slug}.
  • I inferred that URL from the site's naming pattern.
  • The official website is https://www.whiskyspace.com/brand/492.
  • The brand URL is unavailable, but it is probably https://www.whiskyspace.com/brand/white-horse.

Edge cases

No results

  • Say no matching results were found.
  • Suggest a broader search term or a different filter.

Missing fields

  • Use Not available or Not provided.
  • Do not omit an important field if the user explicitly asked for it.

URL request with incomplete brand data

  • If you have brand.slug but not brand.id, call the brand details endpoint first.
  • If you still do not get a numeric id, say the WhiskySpace brand page URL is unavailable.

API error or timeout

  • Explain the failure plainly.
  • Suggest retrying or simplifying the query.
  • Do not dump raw error output unless the user asks for it.

Large result sets

  • Show the top 10 by default.
  • Mention the total count when available.
  • Suggest filters such as region, age, price, or rating.

Minimal examples

Search:

curl -s "https://www.whiskyspace.com/api/search?q=Macallan&per_page=5" -H "Accept: application/json"

Whisky details:

curl -s "https://www.whiskyspace.com/api/whiskies/macallan-12-year-old" -H "Accept: application/json"

Brand details by slug to retrieve numeric ID:

curl -s "https://www.whiskyspace.com/api/brands/white-horse" -H "Accept: application/json"

Notes

  • Prices may be stale. Show the recorded date when available.
  • Ratings based on few reviews may be less reliable.
  • No API key is required for the documented endpoints.

平台分布

OpenClaw

74.63%
按下载量换算1,015

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills