Best MCP Servers for Social Media Data in 2026

The Model Context Protocol (MCP) is how Anthropic's Claude — and increasingly Cursor, Codex, Zed, and Windsurf — call external tools. If you're an AI agent that needs Instagram or TikTok data, you don't write integration code; you point at an MCP server and call its tools. Here's what's worth installing.

What is MCP?

The Model Context Protocol is an open standard introduced by Anthropic in late 2024 that lets AI assistants discover and call external tools through a uniform interface. An MCP server exposes tools (functions the model can call), resources (data the model can read), and prompts. The client (Claude Code, Cursor, Codex, Zed, Windsurf) handles the wiring.

For social-media data, an MCP server is essentially a thin proxy that turns Instagram/TikTok API endpoints into tools your AI agent can invoke without you writing any glue code.

1. hikerapi-mcp — Instagram (105+ tools)

hikerapi-mcp exposes every HikerAPI Instagram endpoint as an MCP tool. Profiles, posts, stories, reels, followers, comments, hashtags, locations, search, threads — 105+ tools, no glue code.

HIKERAPI_KEY=... npx -y hikerapi-mcp

Why it's the most-used Instagram MCP: widest endpoint coverage, stable backend (HikerAPI's own infrastructure), thin npm package (no Python runtime required).

Free tier: 100 requests on HikerAPI, no credit card.

2. lamatok-mcp — TikTok

lamatok-mcp is the TikTok counterpart. Videos, profiles, comments, hashtags, music — exposed as MCP tools.

LAMATOK_KEY=... npx -y lamatok-mcp

Same install pattern as hikerapi-mcp, same MCP-client compatibility. Lamatok includes 100 free requests.

3. datalikers-mcp — datasets & cache

datalikers-mcp exposes 29 dataset and cache-API tools across both Instagram and TikTok. Cheaper per call than hitting the live APIs; slightly stale data.

DATALIKERS_KEY=... npx -y datalikers-mcp

Choose if: your AI workflow can tolerate hours-old data and you'd rather pay less. Common case: ML pipelines, retrospective analytics, "find creators who posted about X last month".

Install across MCP clients

Claude Desktop / Claude Code

// claude_desktop_config.json
{
  "mcpServers": {
    "hikerapi": {
      "command": "npx",
      "args": ["-y", "hikerapi-mcp"],
      "env": {"HIKERAPI_KEY": "your_key"}
    },
    "lamatok": {
      "command": "npx",
      "args": ["-y", "lamatok-mcp"],
      "env": {"LAMATOK_KEY": "your_key"}
    }
  }
}

Cursor

Cursor reads MCP config from .cursor/mcp.json in your repo or globally from ~/.cursor/mcp.json. Same JSON shape as Claude.

Zed

Zed supports MCP servers via its extensions system. Configure under Settings → Extensions → Model Context Protocol.

Windsurf / Codex

Both follow Anthropic's reference config. Use the same JSON config as Claude Desktop.

Why use MCP instead of direct API calls?

• Tool discovery. The model sees the schema of every endpoint and picks the right one without you specifying.
• Context-friendly. Tool returns are structured JSON the model can iterate on naturally.
• Composable. Combine with other MCP servers (filesystem, GitHub, Postgres) in the same conversation.
• Zero glue code. Install once, all 105+ Instagram endpoints become callable.

What's missing in the social-media MCP space

As of April 2026, MCP servers for these are not shipping (or are toy-grade):

• YouTube — only the official Data API has wrappers; no MCP-native server with depth.
• Twitter/X — official API restrictions block most viable MCP work.
• LinkedIn — no good MCP server exists; commercial APIs are closed.
• Reddit — multiple lightweight MCPs exist but none with full coverage of public endpoints.

If you need any of those, you're still writing glue code (or using EnsembleData/Apify directly).

Real-world MCP recipes

Find followers of a creator and summarize their bios

With hikerapi-mcp installed, ask Claude:

"Get the first 200 followers of @nasa, then summarize their bios — group them by what they say they do."

Claude calls user_followers_v2, then iterates user_info_by_id in batches, then synthesizes. No code from you.

Build a hashtag report

"For #datavisualization, get the top 50 recent posts and bucket them by post type, engagement rate, and account size."

OSINT dossier

"Pull profile, last 20 posts, last 10 stories, and follower-of-follower graph for @username. Save as JSON."

For non-MCP workflows, see Best Instagram API in 2026 and Best TikTok API in 2026.