KeyAPI
Model Context Protocol

Social data for AI agents,
via MCP

KeyAPI exposes its full social media data suite as native MCP tools using Streamable HTTP. Plug it into any MCP-compatible AI client and let your agents fetch creator profiles, videos, comments, reviews, and more — across 11 platforms in real time.

TikTokInstagramYouTubeTwitter / XThreadsRedditLinkedInFacebookGooglePinterestAmazon
Get API Key

What is MCP?

A universal language for AI tools

The Model Context Protocol (MCP) is an open standard that lets AI models discover and call external tools through a consistent interface. KeyAPI implements MCP over Streamable HTTP — no SSE required, just plain HTTP POST requests.

Each platform has its own endpoint. Select the ones you need, paste the generated config, and your agent can fetch live social data without leaving the conversation loop.

Select platforms:

claude_desktop_config.json

{
  "mcpServers": {
    "keyapi-tiktok": {
      "url": "https://mcp.keyapi.ai/tiktok/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

API Reference

How to call the MCP server

All endpoints use Streamable HTTP (JSON-RPC 2.0 over POST).

01

Initialize session

Start a Streamable HTTP session by sending an initialize request.

curl -X POST https://mcp.keyapi.ai/tiktok/mcp \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "initialize",
    "params": {
      "protocolVersion": "2024-11-05",
      "capabilities": {},
      "clientInfo": {
        "name": "my-client",
        "version": "1.0.0"
      }
    }
  }'
02

List available tools

Discover all tools exposed by the server for this platform.

curl -X POST https://mcp.keyapi.ai/tiktok/mcp \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 2,
    "method": "tools/list",
    "params": {}
  }'
03

Call a tool

Invoke any tool by name with your API key in the Authorization header.

curl -X POST https://mcp.keyapi.ai/tiktok/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $KEYAPI_TOKEN" \
  -d '{
    "jsonrpc": "2.0",
    "id": 3,
    "method": "tools/call",
    "params": {
      "name": "search_shop_creator",
      "arguments": {
        "unique_id": "maccagames",
        "region": "US"
      }
    }
  }'

Replace tiktok in the URL with any supported platform slug.

Available Tools

Everything your agent needs

229+ tools across 11 platforms.

TikTok
75 tools
influencer_list_analytics

Search and filter TikTok influencers with multi-dimensional metrics including follower trends, engagement rates, and sales performance.

get_shop_creator_detail

Retrieve basic profile and performance data for a TikTok Shop creator, including follower count, sales metrics, and GMV.

Instagram
31 tools
get_user_info

Get detailed information about an Instagram user. Supports querying by username or user ID.

get_user_posts

Get a list of all posts by an Instagram user. Supports paginated retrieval.

YouTube
17 tools
get_video_information

Get detailed information about a YouTube video, including full metadata, player response, and initial data.

get_channel_description

Get detailed information about a YouTube channel including subscriber count, view count, join date, and social links.

Twitter / X
11 tools
get_single_tweet_data

Get single tweet data by tweet ID.

search

Search across Twitter/X content. Supports Top, Latest, Media, People, and Lists search types.

Reddit
24 tools
fetch_single_reddit_post_details

Get single post details by post ID, optionally including the context of a specific comment.

fetch_reddit_app_popular_feed

Get Reddit's most popular posts across the site with sort and time range filters.

LinkedIn
10 tools
get_user_profile

Get LinkedIn user profile information.

get_user_posts

Get posts published by a LinkedIn user.

Threads
24 tools
get_user_posts

Get the list of posts by a Threads user.

get_user_reposts

Get the list of reposts by a Threads user.

Amazon
19 tools
product_search

Search Amazon products by keyword or ASIN. Filter by category, price range, brand, and more.

products_by_category

Retrieve a paginated list of products within a specific Amazon category.

Facebook
6 tools
profile_details_by_url

Get profile details via Facebook profile URL.

profile_posts

Get public Facebook profile posts.

Pinterest
10 tools
search_users

Search Pinterest users by keyword.

get_user_information

Get user information for a Pinterest profile.

Google
2 tools
web_search

Perform a Google web search and retrieve ranked results with titles, snippets, and URLs.

image_search

Search Google Images with country and language targeting.

More platforms

More platforms coming soon

Why KeyAPI MCP

Built for how agents actually work

One-line integration

Point any MCP-compatible client at our server endpoint. No SDKs to install, no auth flows to wire up — just a URL and your API key.

Works with any AI agent

Claude, GPT-4, open-source LLMs, and custom agents all speak MCP. KeyAPI's server exposes the same tools regardless of the model.

11 platforms, one connection

TikTok, Instagram, YouTube, Twitter, Threads, Reddit, LinkedIn, Facebook, Google, Pinterest, and Amazon — all accessible through a single MCP server.

Structured JSON responses

Every tool returns clean, typed JSON that's immediately usable downstream — no regex scraping, no HTML parsing.

Compatibility

Works with your existing stack

Any client that supports MCP Streamable HTTP can connect to KeyAPI.

Claude DesktopCursorWindsurfClineContinueCustom agentsLangChainAutoGen

FAQ

Frequently asked questions

What is MCP and why should I use it?

MCP (Model Context Protocol) is an open standard that lets AI models discover and call external tools through a consistent interface. Using KeyAPI via MCP means your AI agent can fetch live social data — creator profiles, trending content, product reviews — without any custom integration code.

Which AI clients support KeyAPI MCP?

Any client that supports MCP Streamable HTTP works out of the box: Claude Desktop, Cursor, Windsurf, Cline, Continue, LangChain, AutoGen, and custom agents. If your client supports JSON-RPC 2.0 over HTTP POST, it's compatible.

Do I need to install any SDK or library?

No. KeyAPI MCP uses Streamable HTTP — plain HTTP POST requests with JSON-RPC 2.0 payloads. There's nothing to install. Just point your MCP client at the server URL and add your API key as a Bearer token.

How do I authenticate with the MCP server?

Pass your KeyAPI API key as a Bearer token in the Authorization header: Authorization: Bearer YOUR_KEYAPI_TOKEN. You can generate your key from the Dashboard.

Can I connect to multiple platforms at once?

Yes. Each platform has its own endpoint (e.g. mcp.keyapi.ai/tiktok/mcp, mcp.keyapi.ai/instagram/mcp). Configure multiple servers in your MCP client config and your agent will have access to all of them simultaneously.

What's the difference between MCP and the REST API?

The REST API is for direct programmatic access from your own code. MCP is designed for AI agents — it exposes the same data as typed tool functions that any MCP-compatible AI client can discover and call autonomously, without you writing any glue code.