Claude Code

Add one line to your terminal and Claude Code can pull current public data from 44 platforms, subject to each endpoint's documented cache window, with no HTTP plumbing required:

claude mcp add --scope user socialcrawl -e SOCIALCRAWL_API_KEY=sc_your_key_here -- npx -y socialcrawl-mcp

That single command registers the server and sets your key. The flags come before the -- separator; everything after it is the command Claude Code runs.

The socialcrawl-mcp server connects any MCP-compatible client to the SocialCrawl API. It is published on npm and the MCP Registry. The current package bundles 44 platforms and 357 endpoints; the REST API serves 48 platforms and 381 endpoints, so the newest additions land in the package a release behind. Call the REST API directly when you need something the package does not list yet.

How do I give Claude Code access to social media data?

Run the claude mcp add command above and restart Claude Code. From then on you can ask things like "Get the TikTok profile for @charlidamelio" or "Search YouTube for machine learning tutorials," and the agent calls the right endpoint for you. Discovery and documentation tools work even before you set a key, so the agent can explore what is available first.

Get a key at socialcrawl.dev (Dashboard → API Keys). Every account starts with 100 free credits, no card required.

How do I connect Claude Desktop, Cursor, VS Code, or Windsurf?

Add the server to your client's MCP config file, then restart the client.

Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json on macOS, %APPDATA%\Claude\claude_desktop_config.json on Windows), Cursor (.cursor/mcp.json), and Windsurf use the same shape:

{
  "mcpServers": {
    "socialcrawl": {
      "command": "npx",
      "args": ["-y", "socialcrawl-mcp"],
      "env": {
        "SOCIALCRAWL_API_KEY": "sc_your_key_here"
      }
    }
  }
}

VS Code (.vscode/mcp.json) uses the servers key with an explicit stdio transport:

{
  "servers": {
    "socialcrawl": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "socialcrawl-mcp"],
      "env": {
        "SOCIALCRAWL_API_KEY": "sc_your_key_here"
      }
    }
  }
}

Any MCP client with stdio transport works with the same pattern: command npx, args ["-y", "socialcrawl-mcp"], and SOCIALCRAWL_API_KEY in the environment.

What tools does the agent get?

Seven tools: socialcrawl_list_platforms, socialcrawl_list_endpoints and socialcrawl_get_docs for discovery, socialcrawl_request for any API call, socialcrawl_check_balance for your credit balance, socialcrawl_monitors for scheduled Monitors, and socialcrawl_web for scraping, crawling and browser sessions. The three discovery tools query bundled local data and need no API key, so the agent can list platforms, inspect an endpoint's parameters, and read docs for free. Only billed data requests spend credits; cache hits cost 0. Before every call, the server validates locally that the platform, endpoint, and required parameters exist, and tells the agent how to fix a bad call without burning a credit. For the full tool table and how validation works, see Skills & MCP and the MCP server page.

What does it cost?

Credits are billed exactly as the REST API bills them — most endpoints are 1 credit, heavier ones 5 or 10, and composite or bundle endpoints carry their own price (for example search/everywhere is a flat 20). Cache hits cost 0 credits. See Endpoint pricing for the exact figure per endpoint. The discovery and documentation tools are local, so they cost 0 credits too. See Credits for how the ledger works, or Pricing for credit packs.

Where to go next

• Skills & MCP reference — the full tool list, the Skills package for 40+ agents, and Skills vs MCP.
• AI Agent Integration for the machine-readable llms.txt, OpenAPI spec, and per-platform reference files.
• Vercel AI SDK and LangChain if you are building your own agent instead of using an MCP client.
• Browse the platform directory or start with the Quickstart.