Model Context Protocol

128+ data APIs, inside your AI assistant

One remote MCP server exposes the whole catalog to Claude, Cursor, ChatGPT, VS Code and any other MCP client. Your assistant can search for the right API, read its endpoint reference and fetch live data with your own key.

Server URLhttps://dataandapis.com/mcpStreamable HTTP · protocol 2025-06-18 · no session required

Authentication is your normal API key, sent as the X-Api-Key header (or Authorization: Bearer). Searching and reading references work without a key; live calls need one. Create a free account for 500 tokens a month.

Tools

search_apis: Find APIs by the data they hold. “barcode nutrition”, “IP geolocation”, “company by domain”, “OFAC”.
get_api: Every endpoint of one API with parameters, example values, request body and an example response.
call_api: Make a live request with your key: slug, endpoint path, query parameters and an optional JSON body.

Setup

Claude.ai and Claude Desktop

In Claude, open Settings → Connectors → Add custom connector and paste the server URL with your key in it: https://dataandapis.com/mcp?key=your-api-key. Claude cannot set custom headers, so the key travels in the URL (over HTTPS only; treat the URL as a secret). For Claude Desktop versions that read a config file, use the mcp-remote bridge:

{
  "mcpServers": {
    "dataandapis": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://dataandapis.com/mcp", "--header", "X-Api-Key:${DATAANDAPIS_KEY}"],
      "env": { "DATAANDAPIS_KEY": "your-api-key" }
    }
  }
}

Cursor, Windsurf, VS Code

Add to .cursor/mcp.json, mcp_config.json or your editor's MCP settings:

{
  "mcpServers": {
    "dataandapis": {
      "url": "https://dataandapis.com/mcp",
      "headers": { "X-Api-Key": "your-api-key" }
    }
  }
}

Plain HTTP

The server is ordinary JSON-RPC over POST, so you can script it directly:

curl https://dataandapis.com/mcp \
  -H "Content-Type: application/json" \
  -H "X-Api-Key: your-api-key" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_apis","arguments":{"query":"sanctions screening"}}}'

Try asking your assistant

“Is Viktor Bout on any sanctions list?”
“What product is barcode 3017620422003 and what is its Nutri-Score?”
“Which public holidays does Mexico have in the next 60 days?”
“Enrich stripe.com: industry, size, headquarters, tickers.”

Prefer files? Every API also publishes an OpenAPI 3.1 document at /apis/<slug>/openapi.json, and /llms.txt lists the whole catalog for language models.