One MCP server, every SocialKit tool
Connect Claude, Cursor, or any MCP client to social data across YouTube, TikTok, Instagram, Facebook, X, and LinkedIn - transcripts, stats, comments, and more. No install, just a URL.
- Works with Claude, Cursor & any MCP client - no install, just a URL
- Same key and pricing as the REST API - 20 free credits to start
Loved by 20,000+ data lovers
# Add to your MCP client config:
{
"mcpServers": {
"socialkit": {
"url": "https://mcp.socialkit.dev/mcp",
"headers": {
"Authorization":
"Bearer YOUR_ACCESS_KEY"
}
}
}
}
# Then in Claude / Cursor:
> Summarize this Instagram Reel
and tell me the engagement:
https://www.instagram.com/...
✓ Calls instagram_stats
✓ Calls instagram_comments
✓ Returns one-paragraph brief{
"mcpServers": {
"socialkit": {
"url": "https://mcp.socialkit.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_ACCESS_KEY"
}
}
}
}Restart Claude. Ask it to summarize a YouTube video. Done.
Use the Agent Skill instead
If you work in Claude Code, Cursor, or Codex, skip the MCP config. The SocialKit Agent Skill teaches your agent every endpoint, parameter, and response shape in one command — and it calls the REST API directly, so no MCP server is required.
- One command, no config file to maintain
- Seven workflow skills: transcripts, summaries, research, discovery, engagement, downloads
- Same access key, same credits as MCP and REST
# 1. Install the skills
$ npx skills add SocialKit-API/skills --all
✓ added 7 SocialKit skills
# 2. Add your key to the environment
$ export SOCIALKIT_API_KEY="sk_live_..."
# 3. Ask for a real outcome
> Summarize this Reel and tell me how
the top comments feel about it.
• Using SocialKit skill · instagram/summarize
• Using SocialKit skill · instagram/comments
→ one-paragraph brief + sentimentThings you can ask your agent
Once the MCP is connected, your AI has every SocialKit endpoint available as a tool. Talk to it like you'd talk to a human researcher.
Every SocialKit endpoint, as an MCP tool
Same inputs, same response shapes as the REST API. Your agent gets the full SocialKit surface as native tools.
YouTube
- youtube_transcript
- youtube_summarize
- youtube_stats
- youtube_comments
- youtube_channel_stats
- youtube_videos
- youtube_search
- youtube_download
TikTok
- tiktok_transcript
- tiktok_summarize
- tiktok_stats
- tiktok_comments
- tiktok_channel_stats
- tiktok_channel_video_metrics
- tiktok_search
- tiktok_hashtag_search
- tiktok_download
- instagram_transcript
- instagram_summarize
- instagram_stats
- instagram_comments
- instagram_channel_stats
- instagram_channel_posts
- instagram_channel_reels
- instagram_reels_search
- instagram_download
- facebook_transcript
- facebook_summarize
- facebook_stats
- facebook_comments
- facebook_channel_stats
Twitter / X
- twitter_profile
- twitter_tweet
- twitter_tweets
- twitter_transcript
- linkedin_profile
- linkedin_company
- linkedin_company_posts
- linkedin_post
- linkedin_transcript
Direct video
- video_transcript
- video_summarize
Your agents already know how to use this
A first-party Claude Code skill, a hosted MCP server, and a plain REST API - all official, all maintained.
Agent skill
A drop-in skill for Claude Code, Cursor, and Codex. It teaches them every endpoint, param, and response shape.
$ npx skills add \
SocialKit-API/skills --all
✓ added 7 SocialKit skills
> summarize the last 5 reels from
@nasa and rank them by viewsMCP server
One config block. Claude, Cursor, or any MCP client gets every SocialKit endpoint as a first-class tool.
{
"mcpServers": {
"socialkit": {
"url": "https://mcp.socialkit.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_KEY"
}
}
}
}REST API
Pipe-friendly and framework-agnostic. Wrap the endpoints as tools in OpenAI, LangChain, CrewAI, or anything with function-calling.
$ curl https://api.socialkit.dev/\
tiktok/stats?url=... \
-H "x-access-key: YOUR_KEY"
→ { likes, views, shares, ... }
1 credit usedZero install
Hosted at mcp.socialkit.dev over HTTP-streamable transport. No binary to download, no `npx -y` to update, no local process to babysit.
Same key, same credits
Uses the SocialKit access key you already have. Tool calls cost the same as the REST endpoint they wrap — 1 credit per success, failures free.
Stateless + safe
Server stores nothing. Your access key is forwarded to api.socialkit.dev, never logged. Open source if you want to self-host.
Hook your agent up in 30 seconds
Grab a free API key, paste 8 lines of JSON into your MCP client, and watch your AI start understanding social media.
Frequently Asked Questions
Ready to Extract Social Media Videos Data?
Extract video summaries, transcripts, comments, and engagement metrics from social media content with our powerful Social Media Data Extraction API.
Video discusses upcoming social media trends including AI-generated content...