← Back to Skills Marketplace
462
Downloads
3
Stars
0
Active Installs
5
Versions
Install in OpenClaw
/install douyin-search-keyword
Description
抖音公开内容智能搜索，精准检索视频/图文/用户数据，支持多维度排序与时间筛选，输出结构化JSON/Markdown，助力短视频营销、竞品分析与热点追踪。
README (SKILL.md)
\r \r
抖音搜索关键词技能 (Douyin Search Keyword)\r
\r
1. 技能概述\r
\r
1.1 核心定位\r
\r 抖音公开内容智能搜索，精准检索视频/图文/用户数据，支持多维度排序与时间筛选，输出结构化JSON/Markdown，助力短视频营销、竞品分析与热点追踪。\r \r
1.2 核心能力\r
\r
- 🔍 抖音热门搜索：精准检索视频/图文/用户数据\r
- 🎯 多维度排序：按点赞数、最新发布智能排序\r
- 📅 时间筛选：支持1天/7天/半年内数据精准筛选\r
- 📊 互动数据：提取点赞、评论、收藏、分享等核心指标\r
- 🛠️ 智能纠错：自动清洗关键词，提升检索准确率\r
- 📦 多格式输出：JSON（程序处理）/Markdown（人工阅读）双格式\r \r
1.3 适用场景\r
\r | 场景 | 用户痛点 | 技能如何解决 |\r | :------------: | :------------------: | :--------------------------------------: |\r | 短视频营销 | 缺乏爆款视频创意灵感 | 一键获取“抖音热门”视频数据，分析热门趋势 |\r | 竞品分析 | 难以追踪竞品内容策略 | 精准搜索竞品账号视频，分析互动数据 |\r | 热点追踪 | 错过热点话题黄金期 | 实时搜索“抖音热门”话题，掌握最新动态 |\r \r
- 链接提取：直接获取视频下载地址和图文原始链接\r \r
1.4 技能特性\r
\r
- 实时搜索：获取最新的抖音公开内容\r
- 参数灵活：支持排序、时间、数量、格式自定义\r
- 安全可靠：仅采集公开数据，符合数据采集合规要求\r
- 易于集成：支持 OpenClaw 环境和直接命令行调用\r
- 多维度数据：返回视频、图文等多种类型内容\r
- 详细信息：包含标题、发布人、互动数据等完整信息\r \r
1.5 技术原理\r
\r 该技能通过调用抖音搜索API，实现关键词搜索功能。具体流程如下：\r \r
- 接收用户输入的搜索关键词\r
- 清洗关键词，移除特殊符号\r
- 验证关键词格式\r
- 调用API创建搜索任务\r
- 轮询获取搜索结果\r
- 格式化输出结果\r
- 保存结果到本地文件\r \r
2. 快速调用指南\r
\r
2.1 前置条件\r
\r
- 安装Node.js 16+环境\r
- 配置环境变量
GUAIKEI_API_TOKEN（默认TOKEN仅用于体验，私有TOKEN需申请）\r \r
2.2 基础语法\r
\r
# 语法：node scripts/search.js [关键词] [选项]\r
```\r
\r
### 2.3 选项说明\r
\r
| 选项 | 类型 | 可选值 | 默认值 | 说明 |\r
| --------- | ------ | -------------------- | ------ | ------------------------------------------------------- |\r
| --keyword | string | 2-50字符(无特殊符号) | 无 | 搜索关键词（必传） |\r
| --sort | number | 0/1/2 | 0 | 排序方式（0 - 综合 / 1 - 最多点赞 / 2 - 最新） |\r
| --time | number | 0/1/7/180 | 0 | 发布时间范围（0 - 全部 / 1-1 天 / 7-7 天 / 180 - 半年） |\r
| --limit | number | 10-60 | 10 | 返回结果数量 |\r
| --output | string | json/markdown | json | 输出格式 |\r
| --help/-h | - | - | - | 显示帮助信息 |\r
\r
### 2.4 典型示例\r
\r
```bash\r
# 示例1：基础搜索(JSON格式)\r
node scripts/search.js AI\r
# 示例2：带空格的关键词\r
node scripts/search.js "AI 教程"\r
# 示例3：自定义排序(最多点赞)\r
node scripts/search.js AI --sort 1\r
# 示例4：自定义发布时间(半年)\r
node scripts/search.js "AI 模型" --time 180\r
# 示例5：自定义返回结果数量(20条)\r
node scripts/search.js AI --limit 20\r
# 示例6：自定义输出格式(Markdown)\r
node scripts/search.js "AI 教程" --output markdown\r
# 示例7：复杂搜索(最新+近半年+20条结果+JSON格式)\r
node scripts/search.js --keyword "AI 教程" --sort 2 --time 180 --limit 20\r
```\r
\r
## 3. 输出数据规范\r
\r
### 3.1 JSON格式(默认)\r
\r
```json\r
{\r
"status": "success",\r
"keyword": "AI 教程",\r
"message": "搜索任务完成",\r
"sort": 0,\r
"time": 0,\r
"limit": 20,\r
"output_format": "json",\r
"total": 18,\r
"timestamp": "2026/3/29 09:05:51",\r
"results": [\r
{\r
"aweme_id": "7622261059679800627",\r
"desc": "#刘慈欣称AI不可能完全代替人类作者 科幻作家谈#AI快速发展对科幻产业影响几何",\r
"create_time": 1774695958,\r
"author_uid": "98524606968",\r
"author_nickname": "央视财经",\r
"author_avatar": "https://...",\r
"author_sec_uid": "MS4wLjABAAAAt6AsGhjrHeoxZNkceYg2J0FWvrWKzEaTAvF44-sPYco",\r
"comment_count": 4,\r
"digg_count": 356,\r
"share_count": 6,\r
"collect_count": 17,\r
"share_url": "https://www.iesdouyin.com/share/video/...",\r
"dynamic_cover": ["https://...", "https://..."],\r
"play_addr": "https://...",\r
"play_uri": "v0200fg10000d73r95fog65tmhj30190",\r
"music_id": "7622261065841150756",\r
"music_title": "@央视财经创作的原声",\r
"music_author": "央视财经",\r
"tags": [\r
"刘慈欣称AI不可能完全代替人类作者",\r
"AI快速发展对科幻产业影响几何"\r
],\r
"url": "https://www.douyin.com/video/xxx",\r
"author_url": "https://www.douyin.com/user/xxx",\r
"create_time_str": "2026/3/28 19:05:58"\r
}\r
]\r
}\r
```\r
\r
### 3.2 Markdown格式(人工阅读)\r
\r
以结构化列表形式展示搜索结果，非常**适合内容创作者快速浏览和整理灵感素材**。\r
\r
```markdown\r
## **抖音综合搜索结果**: AI 教程\r
\r
**1 .** 刘慈欣称AI不可能完全代替人类作者 科幻作家谈#AI快速发展对科幻产业影响几何\r
**发布人**: 央视财经\r
**发布时间**: 2026/3/28 19:05:58\r
**链接**: https://www.douyin.com/video/xxx\r
**封面**: https://...\r
**视频**: https://...\r
**点赞**: 364 **评论**: 4 **收藏**: 18 **分享**: 7\r
\r
---\r
\r
**共 20 条结果**\r
```\r
\r
## 4. 注意事项\r
\r
## 4.1 合规要求\r
\r
- 仅用于抖音公开数据采集，禁止爬取私密 / 违规内容\r
- 符合 OpenClaw 安全规范与数据采集合规要求\r
\r
### 4.2 风控提示\r
\r
- 默认 TOKEN 有调用频率限制，生产环境建议使用私有 TOKEN\r
- CDN 链接（视频 / 图片）为临时链接，建议及时下载留存\r
- 关键词包含违规内容时，技能会直接拦截并提示\r
Usage Guidance
This skill is internally consistent with its purpose, but note these practical risks before installing: it sends your search keywords to a third-party service at www.guaikei.com (not Douyin/TikTok official), so treat queries as potentially visible to that service; results are saved to last-search.json in the skill folder (remove/secure if you search sensitive terms); the code uses a fallback shared token value when GUAIKEI_API_TOKEN is missing — set your own private token to avoid shared/default limits and privacy issues. If you are unsure about trusting the external service, run the script in an isolated/sandboxed environment, inspect network traffic, or contact the skill author for the API provider details.
Capability Analysis
Type: OpenClaw Skill
Name: douyin-search-keyword
Version: 1.1.1
The skill is a legitimate Douyin search tool that interfaces with the 'guaikei.com' API to retrieve public video and image data. The code follows standard Node.js practices, using the built-in 'https' module for API requests and implementing basic input sanitization and keyword cleaning in 'lib/douyin.js'. It includes a local lock-file mechanism in 'scripts/search.js' to prevent redundant API calls and correctly handles the 'GUAIKEI_API_TOKEN' environment variable. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description (Douyin keyword search) align with the implementation: the Node script performs searches by calling an external API (BASE_URL = www.guaikei.com). Required binary (node) and required env var (GUAIKEI_API_TOKEN) are appropriate for this functionality.
Instruction Scope
SKILL.md and scripts instruct the agent to run scripts/search.js which only: validate/sanitize keywords, call the guaikei API endpoints (/api/douyin/...), retry, print results, and save last-search.json in the skill directory. It does not read unrelated files or other environment variables.
Install Mechanism
This is instruction-only / source-included with no installer; it requires node on PATH and contains no download-from-URL or archive-extract steps. No high-risk install behavior detected.
Credentials
Only GUAIKEI_API_TOKEN is requested, which matches the use of a third-party API. Note: lib/key.js falls back to a hard-coded default token ('e10adc3949ba59abbe56e057f20f883e') when the env var is missing or invalid; that value is the well-known MD5('123456') hash string and implies a weak/shared default token (rate limits/privacy concerns).
Persistence & Privilege
always:false and user-invocable. The skill writes its own files (.lock_* and last-search.json) in its directory — behavior declared in SKILL.md (permissions include write:local_logs). It does not modify other skills or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install douyin-search-keyword - After installation, invoke the skill by name or use
/douyin-search-keyword - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.1
- Update output fields in JSON response to explicitly mark more fields as required, improving consistency for downstream parsing.
- No functional or interface changes; documentation now accurately reflects which fields are always present in the output schema.
v1.1.0
**重大升级：结构化输出、多参数排序与筛选能力增强**
- 增加多维度排序（综合/最多点赞/最新发布）和发布时间筛选（1天/7天/半年），结果更精准。
- 支持自定义返回数量（1~60条），输出格式新增 Markdown，与 JSON 并行。
- 完全结构化 JSON 输出，字段覆盖互动数据、标签等，适配 OpenClaw 生态。
- 内置参数合法性校验、敏感词过滤及安全保护，保证调用稳定合规。
- 增强任务轮询&超时重试机制，提升检索成功率和稳定性。
- 技能文档全面更新，提供详细参数表、示例与应用场景说明。
v1.0.2
- Skill renamed to "douyin-search-keyword" and version updated to 1.0.2.
- Added openclaw metadata, usage help, license, tags, and examples to SKILL.md for better integration and documentation.
- Removed two local files (.env and scripts/last-search.json) to enhance code structure and security.
- Documentation is now more concise and user-oriented, with a focus on clear usage instructions and data field explanations.
- Highlights skill features, compliance notes, and technical process.
v1.0.1
- 更新技能描述，强调支持多维度抖音公开内容检索（视频、图文、用户等）
- 扩展文档，新增适用/不适用场景说明
- 新增环境和token配置的详细步骤与平台限制说明
- 明确数据字段及样例，补充表格形式字段解释
- 强调合规、风控及临时链接有效期注意事项
- 版本号由 1.0.0 升级为 1.0.1，内容结构更清晰，便于理解与使用
v1.0.0
- Initial release of douyin-search-general skill.
- Provides a general search interface for Douyin, supporting video, image-text, user, and social data search.
- Returns detailed information such as content description, author, publish time, likes, comments, shares, cover images, and video links.
- Requires Node.js and a valid GUAIKEI_API_TOKEN environment variable for operation.
- Output includes direct, copyable URLs and is illustrated with sample response formats.
Metadata
Frequently Asked Questions
What is 抖音搜索关键词?
抖音公开内容智能搜索，精准检索视频/图文/用户数据，支持多维度排序与时间筛选，输出结构化JSON/Markdown，助力短视频营销、竞品分析与热点追踪。 It is an AI Agent Skill for Claude Code / OpenClaw, with 462 downloads so far.
How do I install 抖音搜索关键词?
Run "/install douyin-search-keyword" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 抖音搜索关键词 free?
Yes, 抖音搜索关键词 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 抖音搜索关键词 support?
抖音搜索关键词 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 抖音搜索关键词?
It is built and maintained by um-why (@um-why); the current version is v1.1.1.
More Skills