MCP Server Fear & Greed Index

A Model Context Protocol (MCP) server that provides access to the CNN Fear & Greed Index for the US stock market. This server fetches real-time market sentiment data and presents it in both structuredContent and text content.

Features

• Real-time Fear & Greed Index: Get the current market sentiment score (0-100)
• Historical Comparisons: View previous close, week, month, and year data
• Detailed Market Indicators: Access individual component scores including:
  • Market Momentum (S&P 500 & S&P 125)
  • Stock Price Strength & Breadth
  • Put/Call Options Ratio
  • Market Volatility (VIX)
  • Junk Bond Demand
  • Safe Haven Demand
• Flexible Output: Choose between structured markdown or raw JSON format

Requirements

• Node.js 18 or newer
• VS Code, Cursor, Windsurf, Claude Desktop or any other MCP client

Getting Started

Local (Stdio)

First, install the Fear & Greed MCP server with your client. A typical configuration looks like this:

{
  "mcpServers": {
    "mcp-server-fear-greed": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-server-fear-greed@latest"
      ]
    }
  }
}

You can also install the mcp-server-fear-greed MCP server using the VS Code CLI:

# For VS Code
code --add-mcp '{"name":"mcp-server-fear-greed","command":"npx","args":["mcp-server-fear-greed@latest"]}'

After installation, the Fear & Greed MCP server will be available for use with your GitHub Copilot agent in VS Code.

Go to Cursor Settings -> MCP -> Add new MCP Server. Name to your liking, npx mcp-server-fear-greed. You can also verify config or add command like arguments via clicking Edit.

{
  "mcpServers": {
    "mcp-server-fear-greed": {
      "command": "npx",
      "args": [
        "mcp-server-fear-greed@latest"
      ]
    }
  }
}

Follow Windsurf MCP documentation. Use following configuration:

{
  "mcpServers": {
    "mcp-server-fear-greed": {
      "command": "npx",
      "args": [
        "mcp-server-fear-greed@latest"
      ]
    }
  }
}

Follow the MCP install guide, use following configuration:

{
  "mcpServers": {
    "mcp-server-fear-greed": {
      "command": "npx",
      "args": [
        "mcp-server-fear-greed@latest"
      ]
    }
  }
}

Remote (SSE / Streamable HTTP)

At the same time, use --port $your_port arg to start the browser mcp can be converted into SSE and Streamable HTTP Server.

# normal run remote mcp server
npx mcp-server-fear-greed --port 8089

You can use one of the two MCP Server remote endpoint:

• Streamable HTTP(Recommended): http://127.0.0.1::8089/mcp
• SSE: http://127.0.0.1::8089/sse

And then in MCP client config, set the url to the SSE endpoint:

{
  "mcpServers": {
    "mcp-server-fear-greed": {
      "url": "http://127.0.0.1::8089/sse"
    }
  }
}

url to the Streamable HTTP:

{
  "mcpServers": {
    "mcp-server-fear-greed": {
      "type": "streamable-http", // If there is MCP Client support
      "url": "http://127.0.0.1::8089/mcp"
    }
  }
}

In-memory call

If your MCP Client is developed based on JavaScript / TypeScript, you can directly use in-process calls to avoid requiring your users to install the command-line interface to use Fear & Greed MCP.

import { Client } from '@modelcontextprotocol/sdk/client/index.js';
import { InMemoryTransport } from '@modelcontextprotocol/sdk/inMemory.js';

// type: module project usage
import { createServer } from 'mcp-server-fear-greed';
// commonjs project usage
// const { createServer } = await import('mcp-server-fear-greed')

const client = new Client(
  {
    name: 'test fear greed client',
    version: '1.0',
  },
  {
    capabilities: {},
  },
);

const server = createServer();
const [clientTransport, serverTransport] = InMemoryTransport.createLinkedPair();

await Promise.all([
  client.connect(clientTransport),
  server.connect(serverTransport),
]);

// list tools
const result = await client.listTools();
console.log(result);

// call tool
const toolResult = await client.callTool({
  name: 'get_fear_greed_index',
  arguments: {
    format: 'json'
  },
});
console.log(toolResult);

API Reference

Tool: get_fear_greed_index

Fetches the current Fear & Greed Index and related market indicators.

Parameters

• format (optional): Output format
  • "structured" (default): Returns formatted markdown with organized data
  • "json": Returns raw JSON data

Example Usage

// Get structured output
await client.callTool("get_fear_greed_index");

// Get JSON output
await client.callTool("get_fear_greed_index", { format: "json" });

Response Structure

The tool returns data in the following structure:

{
  "fear_and_greed": {
    "score": 75,
    "rating": "greed",
    "timestamp": "2025-07-18T23:59:57+00:00",
    "previous_close": 75.31,
    "previous_1_week": 75.26,
    "previous_1_month": 54.29,
    "previous_1_year": 45.94
  },
  "fear_and_greed_historical": {
    "timestamp": 1752883197000,
    "score": 75,
    "rating": "greed"
  },
  "market_momentum_sp500": {
    "timestamp": 1752871567000,
    "score": 61.2,
    "rating": "greed"
  },
  "market_momentum_sp125": {
    "timestamp": 1752871567000,
    "score": 61.2,
    "rating": "greed"
  },
  "stock_price_strength": {
    "timestamp": 1752883197000,
    "score": 80,
    "rating": "extreme greed"
  },
  "stock_price_breadth": {
    "timestamp": 1752883197000,
    "score": 84,
    "rating": "extreme greed"
  },
  "put_call_options": {
    "timestamp": 1752871897000,
    "score": 79.6,
    "rating": "extreme greed"
  },
  "market_volatility_vix": {
    "timestamp": 1752869701000,
    "score": 50,
    "rating": "neutral"
  },
  "market_volatility_vix_50": {
    "timestamp": 1752869701000,
    "score": 50,
    "rating": "neutral"
  },
  "junk_bond_demand": {
    "timestamp": 1752877800000,
    "score": 88.8,
    "rating": "extreme greed"
  },
  "safe_haven_demand": {
    "timestamp": 1752868799000,
    "score": 81.4,
    "rating": "extreme greed"
  }
}

Fear & Greed Index Ratings

The index uses the following rating scale:

• 0-25: Extreme Fear
• 26-45: Fear
• 46-55: Neutral
• 56-75: Greed
• 76-100: Extreme Greed

Development

Access http://127.0.0.1:6274/:

npm run dev

Error Handling

The server includes comprehensive error handling:

• Network request failures are caught and reported
• Invalid API responses are handled gracefully
• Missing data fields are filled with sensible defaults
• All errors include descriptive messages