ae_douyin_search_douplus_user
## Purpose
Find Douyin users by account ID or nickname in the DOU+ promotion context, with cursor-based pagination.
## Inputs
Provide a required `request_body` object containing:
- `keyword` (required string): Douyin account ID or nickname to search for.
- `count` (optional integer or null): Results per page; defaults to 20.
- `cursor` (optional integer or null): Pagination cursor; defaults to 0. Use 0 for the first page, then the cursor returned by the previous response.
## Results
Returns matching users with details such as nickname, Douyin account ID, `sec_uid`, and follower count, plus pagination information. The returned `cursor` is used to request the next page, and `has_more` indicates whether more results are available. The exact user-record structure is not fixed by the response schema.
This tool searches accounts only; it does not create or submit a promotion campaign.
ae_douyin_fetch_hot_search_result
Fetch Douyin hot search results to see trending topics. No input is required; call with an empty object. The response includes hot search data in the data field, which may be null, plus response status and metadata. Individual entry fields are not specified. Keyword searches and date-based filtering are not supported.
ae_douyin_index_fetch_item_base
Retrieve basic information for a specific video in Douyin Index video analysis.
Input: item_id is required and must be the video's Douyin aweme ID as a digits-only string, for example "7660090930283744575".
Result: The response data is documented to include the video's title, cover, duration, publication time, author information, and basic metrics. The exact data fields are not fixed by the response schema, and data may be null. Interpret only the fields actually returned. This tool looks up a known video ID rather than searching for videos.
ae_x_get_trending
Retrieve trending data on X (Twitter) for a selected country.
Input:
- country (optional string): Country identifier; defaults to UnitedStates when omitted. Use the documented spelling, for example UnitedStates, UnitedKingdom, Japan, India, China, or Germany.
- No date, pagination, or result-limit inputs are available.
Returns:
- A response containing the country's trending data in data, with status and response metadata. The detailed structure of the trending data is not specified, and data may be null; specific fields such as ranks or post counts are not guaranteed.
Example: {"country":"UnitedStates"}
ae_kuaishou_web_video_data_by_url
Retrieve data for one Kuaishou video using the required url. Supports a video URL or share short link. Video records may be returned in data.photos; fields depend on the actual response. Photo-only posts are not supported.
ae_instagram_search_hashtags
Search Instagram hashtags related to a word or phrase.
Input:
- `keyword` (required, string): The term to search for, such as `cat`.
Return:
- `data.items`: A list of matching hashtags, including hashtag names and post counts where available.
ae_x_twitter_getusersfollowers
Get one page of followers of a user by user ID. Page size can vary; continue using the returned pagination token.
ae_douyin_index_fetch_daren_great_user_fans_info
Retrieve fan analysis for a Douyin creator to understand the audience's demographics and activity patterns.
Input: user_id is required and must be the creator's numeric Douyin UID, passed as a string, for example "3100268042915212".
Returns: response data with audience-analysis information such as fan gender, age and geographic distributions, and active times. The detailed result structure is not specified; interpret the fields and units actually returned. This tool provides audience analysis rather than individual follower records.
ae_douyin_index_fetch_multi_keyword_interpretation
Retrieve index interpretations for multiple keywords on Douyin or Toutiao within a specified date range, including composite, search, and content indices.
Inputs:
- keyword_list (required): Keywords supplied as one comma-separated string.
- start_date and end_date (required): Start and end dates in YYYYMMDD format.
- app_name (optional): aweme for Douyin (default), or toutiao for Toutiao.
- region (optional): Comma-separated regions; leave empty or omit for nationwide results.
Returns: Keyword index interpretation data for the requested filters. The response's data payload is not documented with a fixed structure; available details depend on the returned data. Index interpretations should not be equated with raw search or content counts.
ae_douyin_web_encrypt_uid_to_sec_user_id
Convert a Douyin user uid to sec_user_id, its encrypted identifier representation.
Input: uid is required and must be a string. It is documented as the user's uid (short_id), for example "1673937488185292".
Returns: a JSON response with status and message metadata and a data payload described as user information. The detailed payload structure and the field path for sec_user_id are not specified; the response is not documented as a bare identifier string.
ae_wechat_get_article_stats
## Purpose
Retrieve read and interaction counts for a WeChat Official Account article from its URL.
## Inputs
- `request_body` (required): object containing the article URL and optional output preference.
- `request_body.url` (required): WeChat MP article link on `mp.weixin.qq.com`, such as `https://mp.weixin.qq.com/s/TSNQKkRpN1qbKsT7BvzqIw`. Short links and long article links with `__biz` are supported when they match `^https?://mp\.weixin\.qq\.com/s([/?].+)?$`. Maximum length: 2048 characters.
- `request_body.raw` (optional boolean; default `true`): use `false` for simplified counts or `true` for the full raw statistics object.
## Results
Statistics are returned in `data`. With `raw=false`, the seven counters are:
- `data.read_num`: reads.
- `data.like_count`: likes.
- `data.old_like_count`: Wow (在看) count.
- `data.share_count`: shares.
- `data.collect_count`: favorites.
- `data.comment_count`: comments.
- `data.star_num`: stars / likes.
With `raw=true`, `data` includes the same counters plus visibility flags (`show_like`, `show_old_like`, `show_share`, `show_collect`, `show_read`, `show_star` and corresponding `*_gray` flags) and status fields (`comment_enabled`, `get_data_succ`, `is_subscribed`, `verify_status`, `can_use_star`). These fields describe visibility and status rather than additional engagement counts.
The result is returned directly; no task polling is needed. Responses can be slow, so a 30-second client request timeout is recommended. This tool returns statistics, not article text or individual comments.
ae_wechat_get_video_comments
Read comments and replies on a WeChat Channels video.
Inputs are supplied inside request_body:
- object_id (required): the video's numeric objectId as a digits-only string, up to 32 characters. Use the video ID from a video list or video detail result, not a video URL.
- comment_id (optional): the parent comment's numeric ID as a string, up to 32 characters. Omit it or leave it empty to read first-level comments; provide it to read that comment's second-level replies.
- last_buffer (optional): base64 pagination cursor. Omit it or leave it empty for the first page. If the response indicates another page, pass its returned cursor with the same video and parent-comment context.
- raw (optional, default true): true returns the original structure; false returns a simplified structure.
Results:
- With raw=false, data.comments contains comment IDs, commenter nickname/username/avatar, content, like and reply counts, creation time and IP region. data.count is the number of comments on this page; data.down_continue indicates whether another page is available; data.last_buffer is the next-page cursor. data.object_id echoes the video ID; data.comment_id echoes the parent comment ID when fetching replies, otherwise it is null.
- With raw=true, comments are in data.commentInfo, the response status is in data.baseResponse.ret, and pagination uses data.downContinueFlag and data.lastBuffer (also available as data.last_buffer). Comment fields include commentId, nickname, username, content, likeCount, createtime, headUrl and ipRegionInfo.regionText. Video-wide counts are available in data.monotonicData.commentCount.commentCount and data.countInfo.
Important: video and comment IDs may exceed JavaScript's safe integer range. Keep them as strings and parse numeric IDs losslessly; do not pass them through JavaScript Number. Responses may be slow; set the client request timeout to 30 seconds.
ae_tiktok_get_user_country
# Purpose
Look up the country or region associated with a TikTok account.
# Input
- `username` (required): The account handle, such as `tiktok` from `https://www.tiktok.com/@tiktok`. Enter the handle without `@`.
# Result
Returns the account's country or region and may include available account identifiers and metadata. This indicates the account's associated country, not necessarily the user's current location or nationality.
ae_youtube_get_general_search
Search YouTube by keyword and optionally narrow results by upload time, video duration, content type, or feature, with configurable sorting and regional settings.
Inputs:
- search_query (required): search keywords, up to 200 characters.
- language_code: language code such as zh-CN or en-US; default zh-CN.
- country_code: country code affecting regional relevance, such as US or CN; default US.
- time_zone: time zone such as America/Los_Angeles or Asia/Shanghai; default America/Los_Angeles.
- upload_time: hour (past hour), today, week (this week), month (this month), or year (this year).
- duration: short (under 4 minutes), medium (4–20 minutes), or long (over 20 minutes).
- content_type: video, channel, playlist, or movie.
- feature: live, 4k, hd, subtitles, creative_commons, 360, vr180, 3d, hdr, location, or purchased.
- sort_by: relevance (documented default), upload_date, view_count, or rating.
- continuation_token: optional pagination token for another page; use a token returned by an earlier search when available.
All inputs except search_query are optional. Choose at most one value per filter; filter fields can be combined.
Returns the search response directly, with raw YouTube search data in data. The result structure requires parsing and is not a standardized list of video, channel, or playlist records; specific item fields are not guaranteed. This tool searches for content rather than downloading it.
Example: search_query="Python tutorial", upload_time="week", duration="short", content_type="video" searches for short Python tutorial videos uploaded this week.
ae_xiaohongshu_note_comment_reply_list
获取小红书(RedNote)指定笔记评论下的二级回复列表,支持游标分页,用于查看话题互动、舆情分析或评论管理。首次查询省略分页状态;后续页使用本工具上次返回的游标、搜索或会话状态,不能复用旧供应商的游标。
ae_xiaohongshu_pgy_creator_core_metrics
获取小红书蒲公英创作者的核心数据指标,支持按业务类型(日常笔记/合作笔记)、笔记形式(图文/视频)、统计周期(近30天/近90天)及广告投放筛选,便于进行达人表现分析与对比。参数请放在必填的 request_body 对象中。