ae_kuaishou_app_user_hot_posts
### 用途 获取指定用户的热门作品数据,并支持分页浏览。 ### 参数 - `user_id`:必填,用户的纯数字 ID。不支持用户 eid。 - `pcursor`:可选,作品分页游标。首次请求可省略;继续获取下一页时,使用上一页响应中返回的游标。 ### 返回 返回该用户热门作品对应的响应数据;如需继续浏览,可从响应中读取后续分页所需的游标。
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_kuaishou_app_user_hot_posts
### 用途 获取指定用户的热门作品数据,并支持分页浏览。 ### 参数 - `user_id`:必填,用户的纯数字 ID。不支持用户 eid。 - `pcursor`:可选,作品分页游标。首次请求可省略;继续获取下一页时,使用上一页响应中返回的游标。 ### 返回 返回该用户热门作品对应的响应数据;如需继续浏览,可从响应中读取后续分页所需的游标。
ae_kuaishou_app_video_comments
用于获取单个视频作品的评论数据。 参数: - photo_id:必填,作品 ID;支持纯数字 ID,也支持短字符串(eID)形式的 ID。 - pcursor:可选,评论分页游标。首次请求可省略或留空;翻页时填写上一页响应返回的 pcursor。 返回:包含该作品评论数据的响应。
ae_wechat_get_live_history
## Purpose List a WeChat Channels creator's live replays, one page at a time, with video metadata and media URLs. The tool returns information for downloading and decrypting replay videos; it does not download or decrypt them itself. ## Inputs Provide a `request_body` object containing: - `username` (required): the creator's finder identifier, in `v2_<hexadecimal characters>@finder` format, 10–256 characters. A display name is not a substitute. - `last_buffer` (optional): base64 pagination cursor; defaults to an empty string and also accepts null. Leave empty for the first page. For subsequent pages, use the cursor returned by the previous page when its continuation flag indicates more results. - `flag` (optional): fetch flag, an integer from 0 to 1000; default `12`. - `raw` (optional): defaults to `true` for the original response. Set to `false` for a simplified structure, recommended for media handling. ## Results For `raw=false`, `data.lives[]` contains replay video objects with IDs, creator information, titles, creation times, engagement counts, and a single `media` object. `data.count` is the number of items on this page; `data.total_count` is the total replay count, or `0` when none exist. `data.continue_flag` indicates whether another page is available (`1` means yes), and `data.last_buffer` is the next-page cursor. The simplified media fields are `url`, `url_token`, `full_url` (the URL with token already appended), and `decode_key`. For `raw=true`, replays are in `data.object[]`, total count in `data.totalCount`, continuation in `data.continueFlag`, and the cursor in `data.lastBuffer` (also documented as `data.last_buffer`). Raw media is at `data.object[N].objectDesc.media[0]`, using `url`, `urlToken`, and `decodeKey`; construct the full URL by concatenating `url` and `urlToken`. ## Handling notes Use the decryption key from the same response as the media URL; keys change between requests. Preserve large 64-bit IDs as strings with lossless JSON parsing, not JavaScript `Number`. Responses may be slow; configure a 30-second timeout in the calling client, not as a tool argument.
ae_youtube_get_channel_info
## Purpose Retrieve information about a specific YouTube channel using its channel ID. ## Input - `channel_id` (required, string): The YouTube channel ID to look up. Provide the ID rather than a channel name or search query. - Example: `channel_id = "UCXuqSBlHAE6Xw-yeJA0Tunw"`. ## Returns A response containing channel information in `data`. The documented response does not specify individual channel fields, and `data` may be null.
ae_reddit_app_fetch_comment_replies
Fetch replies to a Reddit comment. Required inputs: - post_id: The post ID, such as t3_1qmup73. It must start with t3_ followed by 1–17 lowercase letters or digits. - cursor: Copy the target comment node's more.cursor value from previously retrieved comment data. It must begin with commenttree: and be 13–4096 characters long; a comment ID is not a substitute. Optional inputs: - sort_type: CONFIDENCE (default), NEW, TOP, HOT, CONTROVERSIAL, OLD or RANDOM. - language: Preferred response language as an IETF language tag, default en-US. - need_format: Whether to clean and format the data, default false. First retrieve the post's comments, then find the desired comment with child replies and a more.cursor value. Submit that cursor unchanged with the same post ID. Returns JSON data for the retrieved replies, including sub-comment details such as text, authors and upvotes, along with pagination information. The data structure may vary; one response should not be assumed to contain every reply in the thread.
ae_reddit_app_fetch_subreddit_style
Retrieve the rules and style information for a Reddit subreddit. Inputs (all optional): - subreddit_name: Bare subreddit name, without "r/"; 2–21 letters (A–Z, a–z), digits, or underscores. Default: "pics". - language: Preferred response language as an IETF language tag, such as "en-US". Must be 2–16 characters and match ^[A-Za-z]{2,3}(?:-[A-Za-z0-9]{2,8})*$. Default: "en-US". - need_format: Whether to request data cleaning and formatting. Default: false. Returns: A JSON response whose data contains the subreddit's rules and style information. The detailed fields are not specified, and data may be null. Example: {"subreddit_name":"pics","language":"en-US","need_format":false}
ae_reddit_app_fetch_trending_searches
Retrieve current trending searches and topics from the Reddit app. Inputs (all optional): - language: Preferred response language as an IETF language tag, such as en-US. Must be 2–16 characters and match the accepted language-tag format; default: en-US. - need_format: Whether to request cleaned and formatted data; default: false. Returns: A JSON response with trending-search information in data. This may include trending keywords, topics, search-volume or popularity information, and related post previews. The data payload does not have a fixed item schema, so available details may vary. Example inputs: {"language":"en-US","need_format":false}
ae_xiaohongshu_get_blogger_filter_options
## Purpose Retrieve selectable filter values for Xiaohongshu creator discovery. This tool returns filter dictionaries, not creator profiles or search results. ## Input No parameters are required. Call with `{}`. ## Returned data The dictionary payload is documented at `resp["data"]["data"]`. Check that the data is not null before reading its fields. - `industryTags`: industry hierarchy, with `taxonomy1Tag` for the first level and `taxonomy2Tags` for the second. These are industries, not creator content categories. - `personalTagsV2`: creator personas, organized by family identity, occupation, or background, then groups and leaf tags. - `featureTags`: content expertise, with two levels. - `contentTheme`: content themes. - `audience20`: audience segments. - `automotiveIndustryTag`: industry-specific profiles. - `consumerBehavior`: estimated consumption behavior; may be null for some accounts. - `market`: marketing goals. - `activities`: campaigns; each item's `code` supplies an activity-code filter value for creator discovery. The documentation also mentions `contentTags` (creator content categories), `contentScenes` (note scenes), `areas` (creator locations down to district/county), and `fansAreas` (follower locations down to city). Use these dictionaries only when present in the actual response; do not assume category metadata is always included or substitute industry values for categories. ## Usage notes Filter choices are maintained by the platform and may change. Cache and reuse the dictionaries rather than hardcoding them or fetching them before every search.
ae_threads_get_post_comments
Retrieve the available comments for a Threads post. Inputs: - post_id (required string): The Threads post ID, for example "3390920896561588969". - end_cursor (optional string): A reserved cursor field with no effect. Omit it; pagination is not supported. Results: - comments: The returned comment list. - next_cursor: Always null; there is no usable next-page cursor. - has_more: Indicates whether more data exists, but cannot be used to fetch another page. One call returns the available batch, documented as approximately 25 comments. This is not a guarantee of the post's complete comment history, and this tool cannot retrieve additional pages.
ae_linkedin_get_company_profile
Retrieve a company's public LinkedIn profile using its company page URL. Input: url is required and must be a full company page URL, such as https://www.linkedin.com/company/microsoft/. The response is returned synchronously. Its data contains the available company profile information, which may include the company name, about text, size, industry, headquarters, follower count, and affiliated pages. Specific fields depend on the returned data and are not guaranteed. This tool looks up a company by URL; it does not search by company name.
ae_tencent_huxuan_list_official_account_articles
List a public-account creator's articles for content review and advertising planning. Provide the creator's appId and startDate/endDate in yyyyMMdd format. Optionally filter titles with keyword, choose articleType ALL (all articles, default) or ORDER (advertising collaboration order articles), and set page (starts at 1; default 1). An empty keyword applies no title filter. The response includes the query result in data, along with code, message, and recordTime. The available schema does not specify individual article fields or pagination metadata.
ae_wechat_channels_search_accounts_paginated
Find WeChat Channels creator accounts by searching the account category with a required keyword. Pagination is optional: start with currentPage=1, offset=0, and an empty cookies_buffer; for later pages, set the page number and reuse the offset and opaque cookies_buffer returned by the preceding search response. The response contains search data in data, a result code in code, a message, and recordTime. Individual account fields and pagination fields inside data are not specified, so use the information actually returned. This tool searches accounts, not videos or full account profiles.
ae_wechat_official_accounts_get_search_suggestions
Get WeChat search suggestions to help expand or refine a query before searching. Provide keyword as a string. Optionally choose businessType: ALL (default), ACCOUNT (official accounts), ARTICLE, CHANNEL (Channels), MOMENTS, LIVE, EMOJI, AUDIO, BOOK, WECHAT_INDEX, NEWS, MINI_PROGRAM, or ENCYCLOPEDIA. The response includes the suggestion payload in data, together with code, message, and recordTime. The payload's internal format is not specified; suggestions are not the underlying content or full search results.
ae_chanmama_brand_marketing_seeding_authors
Look up marketing seeding creators for a single brand in Chanmama over a specified date range. Provide a Chanmama brand ID or brand name and start and end dates in YYYY-MM-DD format. Supports pagination with page numbers starting at 1; the default is page 1.
ae_chanmama_brand_marketing_seeding_videos
Look up brand marketing seeding videos in Chanmama for a single brand and date range. Enter a Chanmama brand ID or brand name and start and end dates in YYYY-MM-DD format. The optional page number starts at 1 and defaults to 1.
ae_baidu_tieba_search_posts_in_forum
Search Baidu Tieba topic threads by keyword within a specific forum; comments are not searched. Supply request_body with both forum_name and keyword (both required, not either/or). Use the exact forum name, for example 方便面, without appending 吧; do not supply a forum URL or numeric ID, and no URL encoding is needed. keyword must be a term or phrase, not a post URL or pagination token. Optional sort_type accepts newest (default) or relevance; relevant is retained for legacy compatibility. Result order follows Tieba. For the first page, omit page_token or leave it empty. For subsequent pages, copy this interface's next_page_token unchanged and keep forum_name, keyword, and sort_type unchanged. An empty next_page_token marks the end. Only token pagination is exposed, not page numbers or page size; tokens from other search interfaces must not be reused. Results include items and next_page_token. Each item includes the thread ID, title, summary, author information, forum ID/name, post link, publication/update times, likes, shares, and available image/video data. Times are Unix seconds; update_time may be null. Summaries are not full post bodies. Media can be incomplete, empty arrays do not mean the post has no media, and media URLs may expire. Use site-wide post search when no forum restriction is needed.