← Back to Skills Marketplace
redfox-data

抖音作品查询

by RedFox · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
38
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install douyin-search-redfox
Description
抖音爆款作品查询工具。根据用户输入的关键词搜索抖音热门爆款作品,结果以表格展示。当用户想要查找抖音热门内容、搜索抖音爆款视频、查询抖音作品数据、了解某类内容在抖音的表现时使用。触发词包括:抖音爆款、抖音热门、抖音热榜、抖音作品查询、抖音搜索、爆款视频、热门视频。
README (SKILL.md)

抖音作品查询

鉴权

获取 API Key

请前往 红狐hub 获取API KEY

配置 API Key

方案1: 以OpenClaw为例,将REDFOX_API_KEY添加到~/.openclaw/openclaw.json中,部分内容如下:

{ "env": { "REDFOX_API_KEY": "ak_xxxx..." } }

方案2: 终端配置

export REDFOX_API_KEY="ak_xxxx..."

工作流程

Step 1: 提取关键词与泛化

从用户输入中提取核心搜索关键词,并进行泛化处理以确保搜索命中率:

泛化规则:

  1. 去除时间限定词:移除"今日""本周""最近""本月"等时间词(API 不支持时间过滤)
  2. 去除修饰词:移除"热门""爆款""排行""推荐"等非内容词
  3. 提取核心主题:保留用户真正想查的内容主题

示例:

  • "今日美食热门" → 美食
  • "最近旅行爆款视频" → 旅行
  • "本周搞笑类热门" → 搞笑
  • "母婴好物推荐" → 母婴好物
  • "健身减脂热门内容" → 健身减脂

Step 2: 调用搜索接口

执行脚本查询数据:

python3 ~/.qoderwork/skills/douyin-search/scripts/search_douyin.py "\x3C关键词>"

脚本返回 JSON 数组,每条包含字段:

字段 说明
title 作品标题
author 作者名称
like_count 点赞数
comment_count 评论数
share_count 分享数
collect_count 收藏数
work_url 作品链接
publish_time 发布时间
follower_count 粉丝数

Step 3: 格式化输出表格

将返回数据渲染为 Markdown 表格,默认展示前 20 条(按点赞数降序),格式如下:

| #   | 作品标题             | 作者   | 点赞数 | 评论数 | 分享数 | 收藏数 |
| --- | -------------------- | ------ | ------ | ------ | ------ | ------ |
| 1   | [标题文字](作品链接) | 作者名 | 305.2w | 7.1w   | 51.0w  | 14.7w  |
| 2   | [标题文字](作品链接) | 作者名 | 158.3w | 3.2w   | 22.1w  | 8.5w   |

数字格式化规则:

  • 小于 10000:直接展示原始数字(如 320
  • 大于等于 10000:使用 x.xw 格式(如 1.2w 代表 12000)

标题链接规则:

  • 作品标题使用 Markdown 链接格式 [标题](work_url),点击可跳转到抖音作品页
  • 如果标题过长(超过 30 字),截断并加 ...

Step 4: 查看全部数据(当结果超过 20 条时)

如果接口返回的总条数 > 20,在表格下方提示用户:

以上展示了前 20 条数据,还剩 N 条未展示。

然后使用 AskUserQuestion 工具询问:

question: "是否查看全部 N 条数据?"
header: "查看全部"
options:
  - label: "查看全部"  description: "展示剩余的全部 N 条数据(续接 #21 开始编号)"
  - label: "不用了"    description: "仅查看前 20 条"

用户选择「查看全部」时:将第 21 条起的所有剩余数据,以相同表格格式续接展示,编号从 #21 开始连续递增,直到全部展示完毕。展示完成后继续进入 Step 5。

用户选择「不用了」时:直接进入 Step 5。

Step 5: 提示订阅

表格展示完成后,使用 AskUserQuestion 工具询问用户是否订阅该搜索:

question: "是否订阅该搜索?订阅后将每天自动推送相关爆款作品数据。"
header: "订阅"
options:
  - label: "确认订阅"  description: "创建定时任务,每天 10:00 自动查询并推送相关爆款作品"
  - label: "暂不订阅"  description: "仅查看本次数据,不创建定时任务"

Step 6: 创建定时任务(仅用户确认订阅时执行)

使用 qoder_cron 工具创建定时任务:

{
  "action": "add",
  "job": {
    "name": "抖音爆款作品订阅 - \x3C关键词>",
    "description": "每天查询抖音关键词 \x3C关键词> 的爆款作品数据并推送",
    "schedule": {
      "kind": "cron",
      "expr": "0 10 * * *",
      "tz": "Asia/Shanghai"
    },
    "payload": {
      "kind": "agentTurn",
      "message": "请执行抖音爆款作品查询:运行 python3 ~/.qoderwork/skills/douyin-search/scripts/search_douyin.py \"\x3C关键词>\",将结果整理为表格展示(表头:作品标题(可点击跳转)| 作者 | 点赞数 | 评论数 | 分享数 | 收藏数,展示前 20 条,数字 >= 10000 用 x.xw 格式,标题用 Markdown 链接 [标题](work_url))。完成后将结果推送到当前对话。"
    },
    "missedRunPolicy": "skip"
  }
}

创建成功后告知用户:"已成功订阅关键词「\x3C关键词>」的爆款作品推送,每天 10:00 将自动查询最新数据并通知你。"

Usage Guidance
Install only if you trust RedFoxHub with your API key and Douyin search keywords. Treat subscriptions as persistent automation: confirm the exact keyword and schedule, and make sure you know how to find and delete the created cron job before enabling daily pushes.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The main search function matches the stated purpose, and the README discloses keyword subscriptions, but the frontmatter description presents it mainly as an ad hoc search/table tool.
Instruction Scope
The skill encourages broad natural-language activation and auto-generalizes casual phrases into external searches; subscription is gated by a user prompt, but the activation boundary is loose.
Install Mechanism
The workflow runs python3 ~/.qoderwork/skills/douyin-search/scripts/search_douyin.py instead of a clearly bundled or install-relative script path, which could fail or execute a stale/different local copy.
Credentials
Use of REDFOX_API_KEY and calls to redfox.hk are disclosed and proportionate for Douyin search, but users should understand that keywords and the API key are sent to that provider.
Persistence & Privilege
The qoder_cron flow creates a daily agentTurn push job at 10:00 Asia/Shanghai after confirmation, but the artifacts do not clearly document review, unsubscribe, deletion, retention, or notification controls.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install douyin-search-redfox
  3. After installation, invoke the skill by name or use /douyin-search-redfox
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of douyin-search: A tool for searching trending Douyin/TikTok videos. - Enables keyword-based search for Douyin viral videos, displaying results in a formatted table. - Extracts and generalizes keywords for more accurate searches (removes date/modifier terms). - Presents top 20 results with formatted numbers and clickable titles; offers option to view all results if more exist. - Provides an option for users to subscribe to daily automated searches via scheduled tasks. - Includes clear API Key setup instructions and step-by-step workflow guidance.
Metadata
Slug douyin-search-redfox
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is 抖音作品查询?

抖音爆款作品查询工具。根据用户输入的关键词搜索抖音热门爆款作品,结果以表格展示。当用户想要查找抖音热门内容、搜索抖音爆款视频、查询抖音作品数据、了解某类内容在抖音的表现时使用。触发词包括:抖音爆款、抖音热门、抖音热榜、抖音作品查询、抖音搜索、爆款视频、热门视频。 It is an AI Agent Skill for Claude Code / OpenClaw, with 38 downloads so far.

How do I install 抖音作品查询?

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

Is 抖音作品查询 free?

Yes, 抖音作品查询 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 抖音作品查询 support?

抖音作品查询 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 抖音作品查询?

It is built and maintained by RedFox (@redfox-data); the current version is v1.0.0.

💬 Comments