crypto-feargreed-mcp

Purpose

crypto-feargreed-mcp is an MCP server designed to provide real-time and historical cryptocurrency Fear and Greed Index data to MCP-compatible clients (such as Claude Desktop). The index data comes from Alternative.me.

Main Capabilities

• Current Index: Get the latest Fear and Greed Index value and its classification.
• Historical Data: Get historical index values for a specified number of days.
• Trend Analysis: Analyze trends over a period of time through statistics such as averages and trend direction.
• Tool-Only Support: Includes a tool version of all resources to be compatible with MCP clients that only support tools.
• Prompt Generation: Provides prompt templates for interpreting index values.

Usage

The server exposes the following resources and tools:

• Resources:
  • fng://current: The current cryptocurrency Fear and Greed Index.
  • fng://history/{days}: Historical data for the cryptocurrency Fear and Greed Index.
• Tools:
  • get_current_fng_tool() -> str: The current index, the same as fng://current .
  • get_historical_fng_tool(days: int) -> str: Historical index data, the same as fng://history/{days} .
  • analyze_fng_trend(days: int) -> str: Index trend analysis.

Installation:

• Automatic installation to Claude Desktop via Smithery:
  1. Clone the repository.
  2. Install for Claude Desktop.
  3. Then enable it in the Claude Desktop configuration.
• For other clients, add a server entry in the configuration file.

After installation, you can ask:

• "What's the current Crypto Fear & Greed Index?"
• "Show me the Crypto Fear & Greed Index trend for the last 30 days." Claude will automatically call the corresponding tool and provide a response.

Applicable Scenarios

Suitable for scenarios that require integrating the cryptocurrency Fear and Greed Index into MCP-compatible clients, such as querying and analyzing the index in Claude Desktop.

Limitations

This project is built on the MCP Python SDK, and the data is provided by the Alternative.me Fear & Greed Index API. It uses the MIT License. Other specific limitations are not provided for now.