douyin

name	douyin
description	Work with Douyin (抖音) data via TikHub — fetch videos, user profiles and post lists, video comments, and run video/user/general search via Douyin's dedicated search series. Use when the task targets Douyin. Covers App-V3 and the Douyin Search series.

Douyin / 抖音 (via TikHub)

Deep coverage of Douyin. Exhaustive endpoints: tikhub-find-endpoint "<goal>" --platform douyin.

Setup gate

[ -z "${TIKHUB_API_KEY:-}" ] && echo "Set TIKHUB_API_KEY first (see tikhub-onboarding)."

Coverage: App-V3 (video/user/comments) and the dedicated Douyin Search series. (Douyin-Web, Douyin-Creator, Douyin-Index, Xingtu, and Billboard are de-scoped — reach them via discovery.)

Search note: Douyin has its own specialized search series (/douyin/search/..., POST with a JSON body). Prefer it over the App-V3 search endpoints.

Key endpoints

Goal	Method + path	Key params
One video	GET /api/v1/douyin/app/v3/fetch_one_video_v2	aweme_id
Video by share URL	GET /api/v1/douyin/app/v3/fetch_one_video_by_share_url	share_url
User profile	GET /api/v1/douyin/app/v3/handler_user_profile	sec_user_id
User's videos	GET /api/v1/douyin/app/v3/fetch_user_post_videos	sec_user_id, max_cursor, count
Video comments	GET /api/v1/douyin/app/v3/fetch_video_comments	aweme_id, cursor, count
Video search	POST /api/v1/douyin/search/fetch_video_search_v2	body: keyword, cursor, sort_type, publish_time
User search	POST /api/v1/douyin/search/fetch_user_search_v2	body: keyword, cursor
General search	POST /api/v1/douyin/search/fetch_general_search_v2	body: keyword, cursor, sort_type

Example

# Video search — POST with a JSON body
curl -s -X POST "https://api.tikhub.io/api/v1/douyin/search/fetch_video_search_v2" \
  -H "Authorization: Bearer $TIKHUB_API_KEY" -H "Content-Type: application/json" \
  -d '{"keyword": "美食"}'

# One video by id
curl -s "https://api.tikhub.io/api/v1/douyin/app/v3/fetch_one_video_v2?aweme_id=7637462264047710705" \
  -H "Authorization: Bearer $TIKHUB_API_KEY"

Pagination

App-V3 user/video lists use max_cursor + count; comments use cursor + count; the Search series pages via a cursor field in the JSON body. Loop on the returned cursor/has_more. Cap pages — each is billed.

Hand off to task skills

• Download videos → social-media-downloader
• Analyze an account → creator-analytics
• Competitor benchmarking → competitor-analysis
• Hashtag/keyword work → hashtag-research
• Comment mining → comments-analysis
• Large pulls → bulk-data-export

Red flags

• The Search series is POST with a JSON body (keyword, cursor) — not GET query params.
• User endpoints need sec_user_id; resolve it first if you only have a share URL/nickname.