hive-crypto

Hive-crypto is an MCP server that exposes a broad suite of analytics for cryptocurrencies, Web3, and traditional financial markets. It provides access to 351 specialized tools spanning 14 analytics categories, drawing from data providers like CoinGecko, LunarCrush, DefiLlama, GeckoTerminal, Codex, DeBank, GoPlus, GoldRush, CCXT, Finnhub, and FRED. This enables AI assistants to query real-time and historical market data, on-chain metrics, DeFi analytics, NFT insights, security and risk signals, social sentiment, and economic indicators in a unified MCP endpoint. To use it, run the server via the recommended MCP package (mcp-hive) and connect your MCP client to the hive-crypto endpoint. The client will orchestrate tool calls and aggregate results across the 14 categories, facilitating complex analytics workflows without requiring you to manage individual data sources directly.

The server supports the full spectrum of analytics categories: Market Data & Price for prices and OHLCV; On-Chain DEX & Pool for liquidity and trader stats; Portfolio & Wallet for wallet balances and DeFi positions; Token & Contract for token metadata and forensics; DeFi Protocol for TVL and yields; NFT Analytics for collection and holder data; Security & Risk for honeypot and rugpull signals; Network & Infrastructure for gas, blocks, and chain status; Search & Discovery for token discovery; Social & Sentiment for influencer and news analytics; Stocks & Equities for quotes and company profiles; Forex & Commodities for FX and commodities; Economic Indicators for GDP, CPI, and time series; and Alternative Data for insider activity and earnings calendars. Tools are accessed through the MCP API, which routes requests to the appropriate providers (CoinGecko, LunarCrush, DefiLlama, GeckoTerminal, Codex, DeBank, GoPlus, GoldRush, CCXT, Finnhub, FRED, and others), and returns consolidated results suitable for downstream decision making.

Example workflows include: retrieving real-time price data across multiple assets, analyzing on-chain liquidity and pool activity, tracking DeFi yields and treasury balances, scanning for security signals like honeypots or phishing indicators, and monitoring macroeconomic indicators to contextualize market moves. Integration with your existing MCP client is straightforward; configure the hive-crypto server in your MCP client configuration and begin issuing queries to the consolidated endpoint.

Prerequisites:

• Node.js (LTS) installed on your system
• npm (comes with Node.js) or npm/yarn as your preferred package manager
• Access to the internet to install the mcp-hive package

Installation steps:

1. Install dependencies and fetch the MCP server package

   npm install
   

2. Build the server (if a build step exists in the package.json)

   npm run build
   

3. Start the server

   npm start
   

4. Validate the server is running by checking logs for a listening message and/or by performing a basic MCP query against the hive-crypto endpoint provided by your MCP client configuration.

Notes:

• If you are deploying in a container or orchestration system, ensure the working directory is the project root where package.json resides.
• In production, consider using a process manager (e.g., pm2 or systemd) to keep the server running and to manage restarts.
• When upgrading, review the release notes of mcp-hive for breaking changes or updated data provider integrations.

Tips and considerations:

• Environment variables: If the MCP server requires API keys or provider-specific configuration, set them in your environment and expose them to the node process (e.g., process.env and a corresponding env config).
• Rate limiting: Some data providers enforce rate limits. Implement backoff handling and respecting provider quotas through the MCP client layer.
• Data freshness: Depending on your needs, configure cache or refresh intervals to balance latency against data freshness.
• Security: Monitor for security-related signals (GoPlus/GoPlus-like checks, honeypot detection, phishing signals) and ensure secure handling of tokens or keys.
• Troubleshooting: If the server fails to respond, check the MCP client configuration for correct server name (hive-crypto) and the package name (mcp-hive). Ensure network access to data providers is not restricted by firewall rules.
• Debugging: Use the MCP inspector tooling (if available in your environment) to trace requests and inspect aggregated results from multiple tools.