Token导航 LogoToken导航

Agent Skill · 研究检索

virlo-short-form-video-training-data(virlo 简短视频训练数据)

用于辅助数据整理、表格处理、CSV/Excel 分析、指标计算和图表准备。它适合让 Agent 清洗字段、汇总数据、发现异常、生成统计口径或把分析结果转成可读说明。使用时需要确认数据来源、字段含义和时间范围,避免把样本数据当全量事实;涉及敏感数据、导出文件或批量写回时,应先确认权限和脱敏边界。

virlo-short-form-video-training-data 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 6,550。

总安装 21,790周安装 873GitHub Stars 2下载 7,054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
1.0.1

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:virlo-short-form-video-training-data(virlo 简短视频训练数据)
来源仓库:https://github.com/virlogit/virlo-short-form-video-training-data
安装命令:
openclaw skills install virlo-short-form-video-training-data
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 virlo-short-form-video-training-data

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。

提供 YouTube、TikTok 等平台的社交聆听与趋势摘要。支持病毒视频分析与主题标签热度排序查询。适合内容团队研究爆款规律与优化选题方向。使用时需指定时间段与关键词范围。virlo-short-form-video-training-data 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

数据为抽样统计,不代表全量真实分布。

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 virlo-short-form-video-training-data;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

virlo-short-form-video-training-data 怎么安装?

在终端执行 openclaw skills install virlo-short-form-video-training-data。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

virlo-short-form-video-training-data 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。

virlo-short-form-video-training-data 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。

virlo-short-form-video-training-data 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。

virlo-short-form-video-training-data 用的人多吗?

公开数据是累计安装 21,790 次,近一周安装 873 次,下载 7,054 次,GitHub 2 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。

virlo-short-form-video-training-data 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
virlo-short-form-video-training-data

virlo 简短视频训练数据 · 本页

21,7902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
virlo
description
Virlo social media intelligence — viral video analytics, hashtag rankings, trend digests, and social listening across YouTube, TikTok, and Instagram. Use for content strategy, trend discovery, competitive analysis, and niche monitoring.
license
MIT
metadata

Virlo

Social media intelligence for short-form video — Bloomberg for viral content.

Homepage: https://dev.virlo.ai Source: https://github.com/CalciferFriend/virlo-skill Full API docs: https://dev.virlo.ai/docs | Playground: https://dev.virlo.ai/docs/playground

Config

Set VIRLO_API_KEY environment variable. Your API key has the format virlo_tkn_<your_key> and can be obtained from the Virlo dashboard.

Context

The Virlo API provides cross-platform analytics across YouTube, TikTok, and Instagram. Key capabilities:

  • Hashtags — 500K+ hashtags ranked by usage count and total views
  • Trends — Daily curated trending topics updated at 1am UTC
  • Videos — 2M+ viral video performance data (views, likes, shares, comments)
  • Orbit — Keyword-based social listening with async analysis jobs
  • Comet — Automated niche monitoring with scheduled scraping

API Access

All endpoints use base URL https://api.virlo.ai/v1, snake_case naming, and return data in a { "data": ... } envelope.

Making Requests

Use curl directly with the VIRLO_API_KEY environment variable:

# GET request
curl -s -X GET "https://api.virlo.ai<endpoint>" \
  -H "Authorization: Bearer ${VIRLO_API_KEY}" \
  -H "Content-Type: application/json"

# POST request with JSON body
curl -s -X POST "https://api.virlo.ai<endpoint>" \
  -H "Authorization: Bearer ${VIRLO_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '<json-body>'

Examples

# List top hashtags
curl -s -X GET "https://api.virlo.ai/v1/hashtags" \
  -H "Authorization: Bearer ${VIRLO_API_KEY}" \
  -H "Content-Type: application/json"

# Top 10 viral videos
curl -s -X GET "https://api.virlo.ai/v1/videos?limit=10" \
  -H "Authorization: Bearer ${VIRLO_API_KEY}" \
  -H "Content-Type: application/json"

# Daily trend digest
curl -s -X GET "https://api.virlo.ai/v1/trends" \
  -H "Authorization: Bearer ${VIRLO_API_KEY}" \
  -H "Content-Type: application/json"

# Create an Orbit search
curl -s -X POST "https://api.virlo.ai/v1/orbit" \
  -H "Authorization: Bearer ${VIRLO_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"name":"AI research","keywords":["artificial intelligence","AI tools"]}'

API Reference

Authentication

All requests require a Bearer token:

Authorization: Bearer virlo_tkn_<your_key>

Never commit API keys to version control. Regenerate compromised keys from the dashboard.

Response Envelope

All responses use a { "data": ... } envelope. List endpoints include pagination metadata:

{
  "data": {
    "total": 500,
    "limit": 50,
    "offset": 0,
    "items": [ ... ]
  }
}

Pagination

List endpoints use offset-based pagination via limit and page query parameters:

ParameterTypeDefaultDescription
limitinteger50Items per page (1-100)
pageinteger1Page number (1-indexed)

Maximum of 1000 total results accessible via pagination.

Common Query Parameters

Many video-related endpoints support these filters:

ParameterTypeDescription
min_viewsintegerMinimum view count threshold
platformsstringComma-separated: youtube, tiktok, instagram
start_datestringISO 8601 date filter (earliest publish date)
end_datestringISO 8601 date filter (latest publish date)
order_bystringSort field (varies per endpoint)
sortstringSort direction: asc or desc (default: desc)

Platforms

  • youtube — YouTube Shorts and videos
  • tiktok — TikTok videos
  • instagram — Instagram Reels

Hashtags

Virlo tracks 500K+ hashtags ranked by usage count and total views.

List Hashtags
GET /v1/hashtags

Standard pagination (limit, page).

Response:

{
  "data": [
    {
      "hashtag": "#shorts",
      "count": 10926,
      "total_views": 869912593
    }
  ]
}
FieldTypeDescription
hashtagstringThe hashtag text (may or may not include #)
countintegerNumber of videos using this hashtag
total_viewsintegerSum of views across all videos with this tag

Trends

Daily curated trending topics updated at 1am UTC.

List Trend Groups
GET /v1/trends

Standard pagination (limit, page).

Response:

{
  "data": [
    {
      "id": "b88c0c23-8501-4975-a1e9-b7c1160c6342",
      "title": "Trends for Oct 15th",
      "trends": [
        {
          "id": "132ea402-804d-4515-b706-f3ff9c698c5e",
          "trend_id": "8ab75d1a-cb50-4885-b9b3-2e4ede2a3620",
          "trend_group_id": "b88c0c23-8501-4975-a1e9-b7c1160c6342",
          "ranking": 1,
          "trend": {
            "id": "8ab75d1a-cb50-4885-b9b3-2e4ede2a3620",
            "name": "NBA Season Opening Night",
            "description": "The 2025-26 NBA season officially tipped off...",
            "trend_type": "content"
          }
        }
      ]
    }
  ]
}
FieldTypeDescription
idstringUUID of the trend group
titlestringDisplay title (e.g., "Trends for Oct 15th")
trendsarrayArray of ranked trend entries

Trend entry fields: id, trend_id, trend_group_id, ranking, trend (object with id, name, description, trend_type).


Videos

2M+ viral videos across YouTube, TikTok, and Instagram.

List Top Videos (Cross-Platform)
GET /v1/videos
ParameterTypeDescription
limitintegerItems per page (1-100, default 50)
pageintegerPage number (1-indexed, default 1)
min_viewsintegerMinimum view count filter
platformsstringComma-separated: youtube, tiktok, instagram
start_datestringISO 8601 earliest publish date
end_datestringISO 8601 latest publish date
order_bystringpublish_date, views, created_at
sortstringasc or desc (default: desc)
Platform-Specific Endpoints
GET /v1/youtube-videos
GET /v1/tiktok-videos
GET /v1/instagram-videos

Same parameters and response shape as /v1/videos, filtered to a single platform.

Video Object Fields
FieldTypeDescription
idstringVirlo UUID
urlstringOriginal video URL
publish_datestringISO 8601 publish timestamp
viewsintegerView count
number_of_likesintegerLike count
number_of_commentsintegerComment count
descriptionstringVideo description
thumbnail_urlstringThumbnail image URL
hashtagsstring[]Extracted hashtags
typestringPlatform: youtube, tiktok, instagram
nichestringContent niche category
author_idstringUUID of the video author
bookmarksintegerBookmark/save count
external_idstringPlatform-specific video ID
regionstring/nullGeographic region code
durationintegerDuration in seconds
transcript_rawstring/nullRaw video transcript (when available)

Orbit — Social Listening

Queue keyword-based video discovery jobs that search across platforms. Supports async analysis, Meta ads collection, and creator outlier detection.

Create Orbit Search
POST /v1/orbit
FieldTypeRequiredDescription
namestringYesDescriptive name for the search job
keywordsstring[]YesKeywords to search (1-10)
platformsstring[]NoPlatforms: youtube, tiktok, instagram. Default: all
min_viewsintegerNoMinimum view count threshold
time_periodstringNotoday, this_week, this_month, this_year
run_analysisbooleanNoEnable AI social intelligence analysis (default: false)
enable_meta_adsbooleanNoEnable Meta ads collection (default: false)
exclude_keywordsstring[]NoKeywords to exclude from results
exclude_keywords_strictbooleanNoAlso check transcripts for exclusions (default: false)
List Orbit Searches
GET /v1/orbit

Paginated list of all search jobs. Standard limit/page parameters.

Get Orbit Results
GET /v1/orbit/:orbit_id

Poll until job completes. When run_analysis: true, includes AI analysis report.

Query params: order_by (views, likes, shares, comments, bookmarks, publish_date, author.followers), sort (asc/desc).

Response fields: name, keywords, analysis (markdown AI report when ready), results (contains total_videos, platform counts, trends, videos, ads, creators).

Get Orbit Videos (Paginated)
GET /v1/orbit/:orbit_id/videos

Standard pagination plus min_views, platforms, start_date, end_date, order_by, sort.

Get Orbit Ads
GET /v1/orbit/:orbit_id/ads

Meta ads collected (requires enable_meta_ads: true). Pagination plus order_by (created_at, page_like_count), sort.

Get Creator Outliers
GET /v1/orbit/:orbit_id/creators/outliers

Creators outperforming their follower count. High outlier_ratio = content reaching far beyond follower base. Standard pagination.


Comet — Automated Niche Monitoring

Create niche configs that automatically discover videos, ads, and creator outliers on a schedule.

Create Comet Config
POST /v1/comet
FieldTypeRequiredDescription
namestringYesDescriptive name (e.g., "Tech Reviews")
keywordsstring[]YesKeywords to search (1-20)
platformsstring[]YesPlatforms: youtube, tiktok, instagram
cadencestringYesdaily, weekly, monthly, or cron expression
min_viewsintegerYesMinimum view count threshold
time_rangestringYestoday, this_week, this_month, this_year
is_activebooleanNoDefault: true. Set false to create paused
meta_ads_enabledbooleanNoEnable Meta ads collection (default: false)
exclude_keywordsstring[]NoKeywords to exclude
exclude_keywords_strictbooleanNoAlso check transcripts for exclusions (default: false)
List Comet Configs
GET /v1/comet

Add ?include_inactive=true to include deactivated configs.

Get / Update / Delete Comet Config
GET /v1/comet/:id
PUT /v1/comet/:id       # Full replacement — all required fields must be provided
DELETE /v1/comet/:id     # Soft-delete, returns 204
Get Comet Videos
GET /v1/comet/:id/videos

Standard pagination plus min_views, platforms, start_date, end_date, order_by, sort.

Get Comet Ads
GET /v1/comet/:id/ads

Requires meta_ads_enabled: true. Standard pagination plus order_by (created_at, page_like_count), sort.

Get Creator Outliers
GET /v1/comet/:id/creators/outliers

Standard pagination.


Error Handling

CodeNameDescription
200OKRequest processed successfully
201CreatedResource created
202AcceptedAsync job queued (e.g., Orbit search)
204No ContentSuccessful deletion
400Bad RequestInvalid request parameters
401UnauthorizedMissing or invalid API key
403ForbiddenInsufficient permissions
404Not FoundResource not found
422Unprocessable EntityValid syntax but cannot process
429Too Many RequestsRate limit exceeded
500Internal Server ErrorServer error

Error response format:

{
  "error": {
    "type": "validation_error",
    "message": "keywords is required",
    "param": "keywords"
  }
}

Rate Limits

  • Max limit per request: 100 items
  • Max total accessible results: 1,000 items per query
  • On 429, back off and retry after the retry_after value (in seconds)

平台分布

OpenClaw

94.93%
按下载量换算6,696

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

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