ae_xiaohongshu_creator_inspiration_feed
获取小红书创作者中心热门灵感流,支持通过游标分页拉取更多热门选题和创作思路。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。
AgentEarth Tool Catalog
Query social media profiles, posts, notes, comments and interaction data for account research, topic tracking and content performance analysis across platforms such as X (Twitter), RedNote (Xiaohongshu), Reddit and TikTok.
Tool List
ae_xiaohongshu_creator_inspiration_feed
获取小红书创作者中心热门灵感流,支持通过游标分页拉取更多热门选题和创作思路。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_douyin_fetch_douplus_video_ranking
## Purpose Retrieve the Douyin DOU+ video ranking board to discover popular videos and compare ranking dimensions within a content category and time-range code. ## Inputs Provide a required `request_body` object. All fields inside it are optional and nullable; `{}` uses the documented defaults. - `time_range`: Time-range code, documented as 1, 2, or 3; default 2. The corresponding calendar durations are not specified. - `tag_id`: Content category ID; default 634 (Entertainment). Categories: 601 Drama; 602 Celebrities; 606 Music; 607 Anime/comics/games; 608 Gaming; 609 Current affairs; 612 Dance; 614 Agriculture and rural life; 615 Science and technology; 616 Finance; 617 Parenting; 619 Lifestyle; 621 Health; 623 Relationships; 624 Culture; 625 Workplace; 626 Education; 627 Photography; 628 Food; 629 Travel; 631 Fashion; 633 Sports; 634 Entertainment; 635 Automotive. - `dim_type`: Ranking dimension; default 5. Values: 1 = breakout videos from low-follower accounts; 2 = high completion rate; 3 = high like rate; 4 = high follower-growth rate; 5 = popularity index. - `adv_id`: Optional advertiser ID string; defaults to an empty string and may be left empty. ## Result Returns ranking data with video, author, and metric information when available. The response's `data` payload has no fixed documented field structure and may be null. The tool retrieves rankings; it does not launch promotions or modify campaigns.
ae_douyin_fetch_multi_video_statistics
Retrieve statistics for up to 50 Douyin videos in one request. Input: aweme_ids is required and must be a string. For one video, provide its video ID; for multiple videos, separate the IDs with commas. Do not include more than 50 IDs. Returns: Video statistics including likes (digg_count), downloads (download_count), plays (play_count), and shares (share_count). Example: aweme_ids = "7448118827402972455,7126745726494821640".
ae_douyin_xingtu_get_item_report_comments
Retrieve comment analysis data for a single video. Input: Provide the required request_body object with item_id, the video item ID, as a string. Result: The response's data field contains the video's comment analysis data and may be null. The detailed analysis fields and metrics are not specified, so availability of particular metrics or individual comment records is not guaranteed. Example input: {"request_body":{"item_id":"7451907165419359488"}}
ae_weibo_comprehensive_search
## Purpose Search Weibo by keyword, either across general results or within a selected category. ## Inputs - `query` (required string): Search keyword. - `page` (optional integer): Page number, starting at 1; defaults to 1. Use this parameter to retrieve additional pages. - `search_type` (optional integer): Search category; defaults to 1. Documented values: 1 = general, 61 = realtime, 3 = users, 64 = videos, 63 = images, 62 = following, 60 = hot, 21 = whole network, 38 = topics, 98 = super topics, 92 = places, 97 = products. ## Results Returns matching search data, which may include post content, author information, images, or videos, depending on the category and available results. A uniform result-item structure is not specified. Results are documented as sorted by relevance. Pagination uses only `page`. ## Example `query="NVIDIA", page=1, search_type=1` searches for general results on the first page.
ae_kuaishou_search_creator_ids
按主播昵称关键词搜索快手磁力聚星账号,直接返回 star_id、昵称、普通快手用户 ID、快手号及账号摘要,包括粉丝数、预期播放、互动指标和人民币报价等信息。star_id 是磁力聚星达人标识,不等同于普通快手用户 ID。结果可能含同名或推荐账号,必须核对后选择目标,不能默认取第一条。查直播主播时传 star_type=4;默认搜索短视频作者。支持分类筛选、排序和分页。选中的 star_id 可用于查询主播整体观众画像和账号近30天带货统计。本工具只按昵称搜索,不接受快手主页 URL 直接转换,也不能保证所有快手账号都已被磁力聚星收录。
ae_xiaohongshu_pgy_note_performance_metrics
通过小红书蒲公英平台获取创作者的笔记聚合表现指标,支持多种业务与内容维度筛选。参数请放在必填的 request_body 对象中。
ae_douyin_search_fetch_image_search
## Purpose Search Douyin for image posts and photo collections using a text keyword. This tool retrieves search results and image links; it does not search by an uploaded image or generate images. ## Inputs Provide a required `request_body` object. The fields inside it are optional in the schema: - `keyword`: Search text; default `"猫咪"`. - `cursor`: Integer pagination cursor; default `0` for the first request. - `search_id`: Search session identifier; default `""` for the first request. - `backtrace`: Pagination backtrace value; default `""`. - `content_type`: Explicitly use the string `"2"` for image posts. The schema defaults to `"0"` and lists `"0"`=all, `"1"`=video, `"2"`=picture, `"3"`=article, but this operation is documented for image search; support for other content types is not established. - `sort_type`: `"0"`=comprehensive, `"1"`=most likes, `"2"`=latest; default `"0"`. - `publish_time`: `"0"`=unlimited, `"1"`=last day, `"7"`=last week, `"180"`=last half year; default `"0"`. - `filter_duration`: Video-duration filter: `"0"`=unlimited, `"0-1"`=within one minute, `"1-5"`=one to five minutes, `"5-10000"`=over five minutes. Default to `"0"` for image searches. For later pages, reuse the previous response's `cursor`, `search_id` and `backtrace` when available; do not invent missing pagination values. `has_more` indicates whether more results are available (`1`=yes, `0`=no). ## Example ```json {"request_body":{"keyword":"cat","cursor":0,"search_id":"","backtrace":"","content_type":"2","sort_type":"0","publish_time":"0","filter_duration":"0"}} ``` ## Results The response is returned directly and includes a `data` payload plus response metadata such as a status code and message. Documented search-result fields include pagination information and post entries containing `aweme_info`: content ID (`aweme_id`), description (`desc`), creation time, author information, `image_post_info.images` with image URL lists and pixel dimensions, engagement statistics, visibility status and a share link. These are common fields, not a guaranteed fixed structure: the payload schema is flexible and nullable, so use the actual response to determine available fields.
ae_chanmama_video_hot_rank
View Chanmama's most popular videos for a past hour, day, week, or month, optionally by category. This ranking cannot be searched by title, topic, event, or keyword.
ae_youtube_get_video_captions
## Purpose List the caption languages available for a YouTube video or retrieve existing captions in a selected language. ## Inputs - Provide at least one of `video_id` (recommended) or `video_url` (the full video URL). - Omit `language_code` or leave it empty to list available caption languages. To retrieve caption content, supply a code from that list, such as `en`, `zh-Hans`, or `a.en` when available. - `format` defaults to `srt`. Supported values are `srt` (numbered subtitles with timestamps), `xml` (raw XML), `json3` (YouTube JSON caption structure), and `txt` (plain text without timestamps). ## Results and usage First request the language list if you do not know which languages are available, then request content using a returned language code. The language-list result includes `video_id` and a `captions` list with `language_code` and `language_name` entries. A content request returns captions in the selected format. Larger videos may require asynchronous processing. A `processing` status with a `job_id` means the result is pending, not that caption text is ready; wait for processing to complete before using the final content. If the video has no captions, the result contains an empty `captions` list and a message indicating that no transcript is available. Only existing captions are retrieved: this tool does not generate captions from audio or translate them into unavailable languages. ## Pending caption jobs This tool returns the provider response from one request without automatically polling or creating an AE background task. If the response contains status=processing and a job_id, call ae_youtube_get_video_captions_result with that provider job_id and the same format. The provider job_id is not an AE task_id and must not be passed to ae_get_result. If the query remains queued or active, query again later as needed; an empty captions list with a no-transcript message is a valid no-captions result.
ae_x_twitter_getusersfollowing
Get one page of accounts followed by a user. Page size can vary; continue using the returned pagination token.
ae_xiaohongshu_pgy_fan_summary
通过达人用户 ID 获取小红书蒲公英达人的粉丝数据概览,用于查看账号的整体粉丝规模与受众状态。参数请放在必填的 request_body 对象中。
ae_douyin_billboard_fetch_hot_total_high_fan_list
## Purpose Retrieve the Douyin high follower-growth-rate leaderboard for a recent time window, optionally filtered by keyword and content category. ## Inputs Place query options inside `request_body`: - `page`: Integer page number; default `1`. - `page_size`: Integer number of items per page; default `10`. - `date_window`: Time window in hours: `1`, `24`, `72`, or `168`, meaning the last hour, day, three days, or seven days. Default: `24`. - `keyword`: Search keyword to filter the leaderboard; default `""` means no keyword filter. - `tags`: Array of content-category filters; an empty array means all categories. Each object uses `value` for a top-level category ID and optional `children` containing objects with subcategory IDs in `value`. Omit `children` to include all subcategories under that top-level category. Obtain IDs from the content-tag lookup operation. ## Result Returns the high follower-growth-rate leaderboard for the requested window, page, and filters in the response's `data`, alongside status and response metadata. The detailed structure of leaderboard entries and the growth-rate calculation are not specified. ## Example ```json {"request_body":{"page":1,"page_size":10,"date_window":24,"keyword":"","tags":[]}} ```
ae_tiktok_get_hashtag_videos
Retrieve videos associated with a hashtag. The required ch_id is the hashtag ID. Optional parameters are cursor (page cursor; defaults to 0), count (number of results; defaults to 10), and region (defaults to US; supported values: US, GB, FR, JP, VN, SG). Use the cursor returned in a previous response to request the next page when available. The response contains video result data and response metadata.
ae_weibo_web_post_comments
获取指定微博的一级评论列表。 参数: - id:微博ID,必填。 - count:评论数量,可选,默认值为 10。 - max_id:翻页游标,可选;首次请求留空,后续请求使用上一次返回的 max_id。 返回:评论列表数据及用于继续翻页的 max_id;没有更多评论时,max_id 为空。
ae_dumplingai_list_youtube_channel_videos
List long-form uploaded videos from a YouTube channel with sorting and pagination support.