Sort

AgentEarth Tool Catalog

Social Media

852Tools
12tags
14Providers

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.

  1. Find information using accounts, content URLs or keywords, and retrieve followers, likes, comments and other data through the relevant tools.
  2. Support brand monitoring, competitor account research, topic tracking and content performance analysis with data for agent research tasks.

Tool List

852 tools

ae_wechat_official_accounts_search_articles_with_time_and_sort

Search WeChat Official Account articles by keyword for topic research and content monitoring. A keyword is required. Choose comprehensive, newest-first, or hottest-first sorting (default: comprehensive), and filter by all time, the past day, the past seven days, or the past six months (default: all time). Start with currentPage=1, offset=0, and an empty cookies_buffer; to continue, use the page number and the offset and opaque pagination state returned by the preceding search. Returns a JSON response with search data (data), a response code (code), a message (message), and recordTime. The search-data structure is not specified, so full article text and particular article fields are not guaranteed.

ae_douyin_index_fetch_hot_words

Retrieve the current hot search keywords on Douyin or Toutiao. Input: app_name is optional. Use aweme for Douyin or toutiao for Toutiao; the default is aweme. Results: Hot keyword data, which may include keyword names, search indices, and growth rates to help assess search popularity and changes. Individual fields depend on the returned data; a fixed set of metrics is not guaranteed. This tool is for current hot-word discovery, not historical trend queries.

ae_wechat_generate_video_share_url

Generate a shareable WeChat Channels video short link (weixin.qq.com/sph/…) from a known video object ID. Inputs: - request_body is required. - request_body.object_id is required: a string containing 1–32 digits, such as "14941130915890399732". Use the video id obtained from a video-list or video-detail lookup, not an account ID or share URL. - request_body.raw is optional and defaults to true. Set it to false for a simplified result or leave it true for the raw response. Result: - With raw=false, data.object_id echoes the input and data.share_url contains the generated share link. - With raw=true, the share link is in data.feedH5Url or in entries of data.urlList, where feedH5Url is the link and objectId identifies the video. The raw status code is in data.baseResponse.ret. - The link can be used for sharing or a video-detail lookup that accepts a share URL; it is not a video download URL. Input accepts one object ID per call. Responses may be slow. Set the calling client's request timeout to 30 seconds; timeout is not an input parameter of this tool.

ae_xiaohongshu_pgy_note_get_detail

通过笔记 ID 获取小红书蒲公英笔记详情,用于达人投放研究或单篇笔记复盘。参数请放在必填的 request_body 对象中。

ae_douyin_fetch_hashtag_video_list

Retrieve videos for a specified Douyin hashtag, with pagination and a choice of sorting order. Inputs: - ch_id (required string): The hashtag ID, not its name or text. - cursor (optional integer, default 0): Use 0 for the first page. For the next page, use the cursor returned by the previous response. - sort_type (optional integer, default 0): 0 = comprehensive ranking, 1 = most likes, 2 = latest published. - count (optional integer, default 10): Keep the default value of 10; changing it is documented to cause errors. Returns: Video-list data for the specified hashtag in the response data payload. The documented response schema does not specify individual video fields. Example: ch_id = "1575791821492238", cursor = 0, sort_type = 0, count = 10.

ae_douyin_index_fetch_current_hot_topic

Retrieve Douyin's current hot-topic and rising-topic leaderboards. Inputs: None required; call without parameters. Results: Topic information such as names, popularity index values, and ranking changes. Popularity index values indicate topic popularity, while ranking changes describe movement in the leaderboard. The exact fields and data structure depend on the returned response; not every field is guaranteed. This tool does not accept topic filters or historical date ranges.

ae_douyin_index_fetch_daren_great_item_mile_info

Retrieve a Douyin creator's core metrics. Input: user_id is required and must be the creator's numeric Douyin UID supplied as a string, for example "3100268042915212". Returns: creator metrics such as follower count, like count, published video count, and engagement rate. The exact metric fields and engagement-rate definition are not specified; use the data actually returned. Suitable for reviewing a known creator's core performance metrics, not for searching for creators.

ae_douyin_index_fetch_item_user_profile

## Purpose Get the audience profile for a specific Douyin video over a chosen date range. ## Required inputs - `item_id`: The video's Douyin aweme ID, provided as a numeric string. - `start_date`: Start date as a string in `YYYYMMDD` format. - `end_date`: End date as a string in `YYYYMMDD` format. ## Result Returns audience distributions such as gender, age, region, and interests. These describe the video's audience composition, not individual viewers' personal profiles. The detailed data structure is not specified; use the dimensions and values actually returned. ## Example inputs ```json {"item_id":"7660090930283744575","start_date":"20260704","end_date":"20260710"} ```

ae_dumplingai_search_tiktok_users

Search for TikTok user profiles and creators by keyword using DumplingAI.

ae_xiaohongshu_get_keyword_daily

## Purpose View a keyword's daily search trend on Xiaohongshu to assess rising or falling interest, daily peaks, and possible seasonal patterns. ## Input Provide `request_body.search_word` as a string of 1–100 characters. Example: `{"request_body":{"search_word":"面膜"}}`. ## Result The business payload is at `resp["data"]["data"]`. Its `dailyData` list contains daily entries: - `dateKey`: date. - `searchScore`: search index, not a raw search count. - `noteNum`: daily note count. - `userNum`: number of users searching. ## Limits The platform determines the time window. Approximately the last three months has been observed, but the exact coverage is not guaranteed; no custom date range can be supplied. This tool provides daily trends rather than aggregate totals or period-over-period changes. A null payload means no result was returned, not zero activity; do not automatically repeat the same request solely because the result is empty.

ae_dumplingai_get_youtube_transcript

Get the transcript for a YouTube video, with optional timestamps, using DumplingAI.

ae_youtube_activities_list

List recent activities for a YouTube channel, such as new uploads, likes, or playlist updates.

ae_douyin_index_fetch_daren_great_user_top_video

### Purpose Retrieve a Douyin creator's popular videos within a specified date range. ### Required inputs - `user_id`: The creator's numeric Douyin UID, supplied as a string, for example `"3100268042915212"`. - `start_date`: Start date as a string in `YYYYMMDD` format. - `end_date`: End date as a string in `YYYYMMDD` format. The date range must not exceed 30 days; longer ranges cause an error. ### Result Returns the creator's popular-video list and performance information such as play and like counts. Detailed video fields depend on the returned data; a fixed video-item schema is not documented. ### Example `user_id="3100268042915212", start_date="20260316", end_date="20260414"`

ae_wechat_get_article_detail_h5

Retrieve the text and metadata of a WeChat Official Account article from its link. Inputs: - request_body is required. Its url must be a WeChat article URL, such as https://mp.weixin.qq.com/s/... or a long article link containing __biz, matching the declared URL format and no longer than 2048 characters. - request_body.raw is optional and defaults to true. Set it to false for core article fields; true adds metadata such as account avatars, copyright status and the fetched page's HTML character count. The top-level result structure is unchanged. Results: The response returns article details directly under data: url is the article URL, bizUin identifies the account, itemIdx is the article position, and content contains the parsed article. Typical details include the title, author, account name, summary, cover image, publication time, plain-text body and HTML body. Albums, tags and images may also be available, depending on the article. Publication timestamps are in seconds. Limitations and handling: - content._content_type distinguishes article, immersive, video and unknown page types. content.content_noencode contains HTML only for article pages; it is empty for other types. For unknown types, the body is the summary. - Readership metrics and comment lists are not included. The comment_id field is an identifier, not comment content, and may be empty on immersive or image pages. - Page retrieval and parsing can be slower than a lightweight detail lookup; a 30-second client timeout is recommended. - Preserve large IDs such as mid and comment_id as strings with lossless JSON parsing. Do not convert them through JavaScript Number.

ae_dumplingai_get_tiktok_transcript

Extract the text transcript from a TikTok video URL using DumplingAI.