Build with SOTW Data

🤖 MCP Server

The Model Context Protocol (MCP) lets AI agents like Claude directly query SOTW data. Two options: hosted remote server or local npm package.

Option 1: Remote MCP (Recommended)

Point your MCP client at our hosted endpoint. No installation needed.

// Claude Desktop config (~/.claude/claude_desktop_config.json)
{
  "mcpServers": {
    "statistics-of-the-world": {
      "type": "url",
      "url": "https://statisticsoftheworld.com/api/mcp"
    }
  }
}

Option 2: Local MCP Server

// Install globally
npm install -g @sotw/mcp-server

// Claude Desktop config
{
  "mcpServers": {
    "statistics-of-the-world": {
      "command": "npx",
      "args": ["@sotw/mcp-server"]
    }
  }
}

Available Tools (7)

⚡ REST API v2 (Agent-Optimized)

Enhanced endpoints that return enriched data with units, sources, related indicators, and suggested follow-up queries. Supports format=markdown for direct LLM consumption.

curl https://statisticsoftheworld.com/api/v2/country/USA?format=markdown

curl "https://statisticsoftheworld.com/api/v2/indicator/IMF.NGDPD?limit=10&order=desc"

curl "https://statisticsoftheworld.com/api/v2/history?indicator=IMF.NGDPD&country=USA"

curl "https://statisticsoftheworld.com/api/v1/search?q=gdp"

curl "https://statisticsoftheworld.com/api/v1/compare?countries=USA,CHN,JPN"

curl -X POST https://statisticsoftheworld.com/api/chat -H "Content-Type: application/json" -d '{"message":"top 5 by GDP"}'

OpenAPI & GPT Actions

Use our OpenAPI 3.1 spec with any AI framework — LangChain, CrewAI, AutoGen, GPT Actions, and more.

# Python — LangChain
from langchain_community.tools import OpenAPIToolkit

toolkit = OpenAPIToolkit.from_openapi_spec(
    "https://statisticsoftheworld.com/api/openapi.json"
)

# JavaScript — fetch
const spec = await fetch("https://statisticsoftheworld.com/api/openapi.json")
  .then(r => r.json());

AI Discovery Files

Machine-readable files for AI crawlers and answer engines.

Key Indicator IDs

Most commonly used indicator IDs for quick reference.