What we are building
AI can write reports and track competitors, but it cannot read TikTok, Instagram, Xiaohongshu or Douyin: official APIs are hard to get and limited, and open-source scrapers break every few weeks. FeedMCP turns public data from TikTok, YouTube, Instagram, X, Threads, Telegram, Reddit, LinkedIn, Facebook, Pinterest, Meta Ad Library, Google Ads Transparency, Xiaohongshu (RedNote), Douyin, Kuaishou, WeChat Official Accounts, WeChat Channels, WeChat Search, Weibo, Zhihu and Bilibili into one interface that AI agents and developers can call per use.
What we do
- 193 endpoints on 21 platforms, one response shape
- Pay per call, failures free, prices on every page
- An MCP server so AI agents can ask in plain words
What we do not do
- Nothing behind a login
- No personal contact details, no people search
- No video file downloads
- No fake engagement
Where the data comes from
From what each platform shows publicly, fetched in real time through our data partners. Every answer says when it was fetched. FeedMCP is not affiliated with these platforms.
Contact
[email protected] — we answer within one business day.