aminer-paper-search
Search for papers by passing the corresponding parameters.
Web search plus first-class access to the Chinese platforms — Weibo, WeChat articles, Xiaohongshu, Bilibili.
LABSCO SUMMARY
A great deal of Chinese technical and industry writing lives inside WeChat's walled garden, where crawlers do not reach and search engines index poorly. Same for Xiaohongshu. If your questions touch the Chinese market, that coverage is the difference between an answer and a shrug — and it is not something you can substitute with a better prompt. YouTube subtitles are the other quietly useful one: the transcript without watching.
A search server whose distinguishing feature is coverage of platforms a general web search reaches badly or not at all.
A 302AI API key, and npx. MIT licensed; the vendor's own package.
One command plus a key — npx -y @302ai/web-search-mcp, then supply credentials
USE CASES
WeChat articles and Xiaohongshu, which general search misses.
Pull the subtitles.
Web, Wikipedia and arXiv from the same connection.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
aminer-paper-searchSearch for papers by passing the corresponding parameters.
arxivSearchSearches ArXiv papers and translates their titles.
bilibiliVideoInfoRetrieves detailed information about a specific Bilibili video by its BV ID.
metasoChatGet answers to questions based on query, with real-time search results
metasoReaderFetch full text content from a webpage based on the input URL
metasoSearchSearch web pages and other data sources for real-time results based on questions or keywords
RealTimeSearchReal-time search service from unifuncs
tiktokVideoSearchSearches for videos on TikTok platform using keywords.
twitterSearchSearches for content on Twitter/X platform using keywords.
UnifiedSearchInterfaceUnified search service supporting multiple search APIs including tavily, search1_search, exa and more
webCrawlIf you need to crawl the web for information, you can use this tool. It will crawl the web for the information you need.
webSearchIf you need to search the web for information, you can use this tool. It will search the web for the information you need.
wechatMpArticleRetrieves detailed content of a specific WeChat official account article by its URL.
wechatMpArticleListRetrieves a list of articles from a specific WeChat official account (MP) by its ID.
weiboUserPostsRetrieves posts from a specific user on Weibo (微博) platform.
wikipediaSearchSearches Wikipedia and returns summaries of the top results.
X_PostThis tool allows users to retrieve posts made by a user on platform X by providing the user ID, with options to specify the number of posts and paginate through results.
X_SearchThis tool allows users to search content on platform X using keywords, with options to filter by type and paginate results.
X_UserInfoThis tool allows users to retrieve detailed information about a user on platform X by providing their username (screen_name).
xiaohongshuNoteRetrieves detailed information about a specific note (post) on Xiaohongshu (小红书) platform by its ID.
xiaohongshuSearchSearches for notes (posts) on Xiaohongshu (小红书) platform using keywords.
youtubeVideoInfoRetrieves detailed information about a specific YouTube video by its ID.
youtubeVideoSearchSearches for videos on YouTube platform using keywords.
youtubeVideoSubtitlesRetrieves subtitles/captions for a specific YouTube video by its ID.
Zhihu_AI_SearchZhihu AI Search is an asynchronous API that requires obtaining a message_id first and then retrieving results via ZhihuGetTask. It leverages advanced AI models combined with Zhihu's trusted community content to provide real-time, accurate search and Q&A functionalities.
ZhihuHotListGet real-time Zhihu hot topics information
ZhihuTaskThis interface is used to query results by using the message_id obtained from Zhihu_AI_Search (Zhihu AI Search).
BEFORE YOU INSTALL
27 tools when we connected.
The documented client runs the server on your machine.
CONTEXT COST
Rated by the size of the 27 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
npx -y @302ai/web-search-mcp{
"mcpServers": {
"302ai-web-search-mcp": {
"command": "npx",
"args": [
"-y",
"@302ai/web-search-mcp"
],
"env": {
"302AI_API_KEY": "<YOUR_AI_API_KEY>"
}
}
}
}