About social-data-mcp
social-data-mcp exposes public profile, content, and platform-caption data from X, YouTube, TikTok, and LinkedIn to AI agents for social research. It runs through Apify Actors, charges for results returned, and is available as a hosted remote or local stdio MCP server.
Key features
- 01Fetch X profiles, tweets, and individual tweets
- 02Retrieve YouTube video and channel data
- 03Get TikTok profiles and video statistics
- 04Extract YouTube and TikTok caption transcripts
- 05Look up LinkedIn companies and public profiles
- 06Connect through remote or local stdio MCP
Use cases
- 01Compile public social profiles across platforms
- 02Research video data with platform captions
- 03Look up LinkedIn companies and public career profiles
MCP support
Connect MCP clients through Apify-hosted remote endpoints or the local stdio server. It exposes tools from six Apify Actors for social profiles, content, statistics, and transcripts.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y social-data-mcpAPIFY_TOKENSecretYour Apify API token (console.apify.com/account/integrations). Free account works; you pay per result.Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEsocial-data-mcp
Pay-per-result data tools — X/Twitter profiles & tweets, YouTube videos & channels, TikTok profiles & video stats, YouTube / TikTok transcripts, and LinkedIn companies & public profiles — exposed to AI agents as MCP servers and as Agent Skills. They run on Apify. No platform API key is needed and nothing logs in on its own; the one exception is full X timelines, which need cookies from a session of your own (X shows logged-out visitors only an account's latest few posts). You are charged only for results that actually came back, at the same price on every Apify plan (free included).
| Tool | Apify Actor | Price |
|---|---|---|
X (Twitter) profiles, tweets, single-tweet lookup, engagement filters, incremental runs (full timelines with your own cookies) | nick.cheng/x-twitter-profile-tweets-scraper | $0.005/account · $0.0003/tweet · $0.005/tweet lookup |
| YouTube video metadata & channel data, filters by views/duration/date | nick.cheng/youtube-video-channel-scraper | $0.015/video · $0.005/channel |
| TikTok profiles & per-video stats | nick.cheng/tiktok-profile-stats-scraper | $0.02/profile · $0.003/video |
| YouTube transcripts — spoken-language captions as text + timed segments, from links or channels | nick.cheng/youtube-transcript-scraper | $0.005/transcript; no captions, no charge |
| TikTok transcripts — TikTok's captions as text + timed segments | nick.cheng/tiktok-transcript-scraper | $0.003/transcript; no captions, no charge |
| LinkedIn companies (industry, size, HQ, website, followers, employees, latest job openings) and public profiles (name, headline, titles, companies, location, education) — no login | nick.cheng/linkedin-company-profile-scraper | $0.004/company · $0.004/profile |
The YouTube and TikTok tools can also return the transcript alongside the stats (includeSubtitles: true,
no extra charge).
Narrower single-purpose variants of the same code, for when you want exactly one thing:
x-profile-data ·
single-tweet-scraper ·
youtube-video-stats ·
tiktok-follower-count ·
tiktok-video-stats.
Use as an MCP server (hosted, no install)
Apify hosts an MCP endpoint for every Actor. Add one URL to any client that supports remote MCP servers (Claude, Cursor, VS Code, …) and authenticate with your Apify account when prompted:
https://mcp.apify.com/?tools=nick.cheng/x-twitter-profile-tweets-scraper,nick.cheng/youtube-video-channel-scraper,nick.cheng/tiktok-profile-stats-scraper,nick.cheng/youtube-transcript-scraper,nick.cheng/tiktok-transcript-scraper,nick.cheng/linkedin-company-profile-scraper
Or pick one: https://mcp.apify.com/?tools=nick.cheng/youtube-transcript-scraper.
The servers are also listed in the official MCP Registry as
io.github.mrocker/x-twitter-profile-tweets, io.github.mrocker/youtube-video-channel,
io.github.mrocker/tiktok-profile-stats, io.github.mrocker/youtube-transcripts,
io.github.mrocker/tiktok-transcripts, io.github.mrocker/linkedin-company-profile (see servers/), and the first three on Smithery:
nick-cheng/x-twitter-profile-tweets ·
nick-cheng/youtube-video-channel ·
nick-cheng/tiktok-profile-stats
(the Smithery gateway forwards your Apify token as a token query parameter).
Use as a local stdio server
Published on npm as social-data-mcp and in the
MCP Registry as io.github.mrocker/social-data-mcp:
{
"mcpServers": {
"social-data": {
"command": "npx",
"args": ["-y", "social-data-mcp"],
"env": { "APIFY_TOKEN": "<your token>" }
}
}
}
This is a thin wrapper that starts Apify's own @apify/actors-mcp-server pre-scoped to the six
Actors above. Get a token at https://console.apify.com/account/integrations (free account is enough).
Use as Agent Skills
npx skills add mrocker/social-data-mcp
Installs six skills — x-twitter-profile-tweets, youtube-video-channel, tiktok-profile-stats,
youtube-transcripts, tiktok-transcripts, linkedin-company-profile — each a short SKILL.md telling the agent when to call the tool, how (one curl), what comes
back, and what the tool cannot do.
What these tools will not do
Keyword search on X, follower lists, reply threads, TikTok video lists per account, LinkedIn people/company search, employee lists and Sales Navigator (all login-only), and speech recognition of their own (transcripts are the platforms' own captions; a video without captions gets none). Full X timelines need cookies you supply; nothing here signs in on its own or uses anyone else's account. The READMEs on Apify say so up front.
Correction (October 2026): earlier versions of this page said YouTube transcripts were impossible and that nothing here needed a login. The first was true only of the YouTube web player's caption links, the only ones we had tested; the second stopped being true for X timelines when X rebuilt its web app.
License
MIT for the code in this repository. The Actors themselves are used under Apify's terms.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.