agentsclimarketplace

Creator collection

Skill creativault/skills/creator-scraper-cv/collection/creator-collection

通用AI助手,为creativault打造的技能包

Install
npx -y skills add creativault/skills --skill creator-collection

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

CreatiVault official batch creator collection and export skill. MUST be used when the user provides creator profile links, usernames, keywords, seed lists, or asks to collect, enrich, export, download, poll, or manage creator data collection tasks for TikTok, YouTube, Instagram, or Twitter/X. Use CreatiVault OpenAPI collection scripts as the authoritative source; do not replace collection with web search. Use when: batch collection, collect creators, scrape creators, enrich creator links, username batch, keyword collection, export creators, export xlsx, export csv, task status, poll collection task, download file, 批量采集, 达人采集, 数据采集, 链接采集, 用户名采集, 关键词采集, 离线采集, 导出达人, 导出Excel, 导出CSV, 下载采集结果, 查询采集任务. 批量达人数据采集与导出能力,支持 TikTok、YouTube、Instagram、Twitter 四平台。支持链接批量、用户名批量、关键词采集三种模式。异步任务机制,含提交、轮询、取数、导出完整生命周期。 Use when: 批量采集, 数据导出, 离线采集, batch collection, data export, keyword collection

SKILL.md

8.5 KB, as published. Nobody here has run it

概述

批量达人数据采集与导出能力。支持通过链接、用户名或关键词提交异步采集任务,自动轮询任务状态,完成后导出为 xlsx/csv/html 文件并提供下载链接。

支持平台:TikTok、YouTube、Instagram、Twitter。

Twitter 平台限制:仅支持 LINK_BATCH(链接采集)和 FILE_UPLOAD(用户名采集),不支持视频采集(CREATOR_VIDEOPOST_VIDEO)。

脚本引用

#脚本路径状态
1submit_collection_task.mjs../../scripts/submit_collection_task.mjs
2submit_keyword_task.mjs../../scripts/submit_keyword_task.mjs
3poll_task_status.mjs../../scripts/poll_task_status.mjs
4get_task_status.mjs../../scripts/get_task_status.mjs
5get_task_data.mjs../../scripts/get_task_data.mjs
6export_task_data.mjs../../scripts/export_task_data.mjs
7export_to_csv.mjs../../scripts/export_to_csv.mjs
8get_download_url.mjs../../scripts/get_download_url.mjs

异步任务生命周期

采集任务为异步操作(耗时 5~30 分钟),遵循四阶段流程:

提交(submit) → 轮询(poll) → 取数(get data) → 导出(export)
阶段脚本说明
1. 提交submit_collection_task.mjs / submit_keyword_task.mjs返回 task_id
2. 轮询poll_task_status.mjs每 60s 自动轮询直到终态
3. 取数get_task_data.mjs分页获取原始 JSON 数据(仅在用户明确要求时使用)
4. 导出export_task_data.mjs生成文件并返回下载链接

任务状态与终态判断

状态含义是否终态
processing处理中(爬取中或数据入库中)❌ 继续轮询
completed已完成✅ 可取数/导出
failed失败✅ 报告错误
timeout超时✅ 报告超时

[禁止] 在 status: "processing" 时提前报告结果。 即使 progress: 100%,只要 status 仍为 processing,说明数据仍在入库处理中,不能判定为"0 条数据"或"无匹配结果"。 必须等到 status 变为 completed / failed / timeout 之一后,才能向用户报告最终结果。

典型场景progress: 100%, completed: 0, status: processing — 爬取已完成但数据尚未入库,继续轮询等待 completed 状态。

规则:采集完成后(status: completed),必须先调用 export_task_data.mjs 生成可下载文件并展示链接给用户。不要直接调用 get_task_data.mjs 输出原始 JSON。

积分提醒规则:仅当接口明确返回错误码 40201 时提示积分不足。meta.quota_remaining 是当天剩余 API 请求次数,不是积分余额;采集、轮询或导出成功后,禁止根据该字段生成“剩余积分不足”提醒。

辅助脚本:

  • get_task_status.mjs — 单次查询任务状态(不轮询)
  • export_to_csv.mjs — 管道式本地 CSV 导出(接收 stdin JSON)
  • get_download_url.mjs — 获取已生成文件的下载链接

采集类型

类型task_type 值触发场景提交脚本
链接批量LINK_BATCH用户提供达人主页链接列表submit_collection_task.mjs
用户名批量FILE_UPLOAD用户提供用户名列表submit_collection_task.mjs
关键词采集用户提供关键词,按关键词批量采集submit_keyword_task.mjs

使用示例

# 链接批量采集
node ../../scripts/submit_collection_task.mjs '{"task_type":"LINK_BATCH","platform":"tiktok","values":["https://www.tiktok.com/@creator1","https://www.tiktok.com/@creator2"],"task_name":"Q1 collection"}'

# 用户名批量采集
node ../../scripts/submit_collection_task.mjs '{"task_type":"FILE_UPLOAD","platform":"tiktok","values":["creator1","creator2"],"task_name":"username batch"}'

# 关键词采集
node ../../scripts/submit_keyword_task.mjs '{"platform":"tiktok","keywords":["beauty tips","skincare routine"]}'

# Twitter 链接批量采集
node ../../scripts/submit_collection_task.mjs '{"task_type":"LINK_BATCH","platform":"twitter","values":["https://x.com/creator1","https://x.com/creator2"],"task_name":"Twitter collection"}'

平台支持矩阵

平台LINK_BATCHFILE_UPLOADCREATOR_VIDEOPOST_VIDEO关键词采集
TikTok
YouTube
Instagram
Twitter

Twitter 限制:仅支持链接采集和用户名采集,不支持视频采集(CREATOR_VIDEOPOST_VIDEO)。

参数说明

submit_collection_task.mjs

参数类型说明
task_typestring必填LINK_BATCH(链接)/ FILE_UPLOAD(用户名)
platformstring必填tiktok / youtube / instagram / twitter
valuesstring[]必填。链接或用户名数组,最多 200 条
task_namestring任务名称
webhook_urlstring完成回调 URL(HTTPS)

submit_keyword_task.mjs

参数类型说明
platformstring必填tiktok / youtube / instagram / twitter
keywordsstring[]必填。关键词列表,最多 10 个
task_namestring任务名称
webhook_urlstring完成回调 URL(HTTPS)

poll_task_status.mjs

参数类型说明
task_idstring必填。任务 ID
intervalinteger轮询间隔秒数,默认 60
max_attemptsinteger最大轮询次数,默认 45(约 45 分钟)

get_task_data.mjs

参数类型说明
task_idstring必填。任务 ID
pageinteger页码,默认 1
sizeinteger每页条数,默认 20,最大 100

export_task_data.mjs

参数类型说明
task_idstring必填。任务 ID(必须已完成)
formatstring必填xlsx / csv / html

重复调用相同 task_id + format 会返回缓存文件,不会重复生成。

输出格式

导出格式

格式说明
xlsxExcel 格式(默认推荐)
csvCSV 格式
htmlHTML 表格格式

下载链接

export_task_data.mjs 返回 file_url 字段,为 OSS 签名下载链接。如需重新获取链接,使用 get_download_url.mjs

node ../../scripts/get_download_url.mjs '{"file_id":"xxx"}'
# 或
node ../../scripts/get_download_url.mjs '{"file_name":"task_xxx.xlsx"}'

export_to_csv.mjs 为本地管道导出,通过 stdin 接收 JSON 数据:

node ../../scripts/get_task_data.mjs '{"task_id":"xxx","size":100}' | node ../../scripts/export_to_csv.mjs '{"output":"creators.csv"}'

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.