Back to MCP Servers

LunarCrush MCP Server

Official LunarCrush MCP server providing real-time social intelligence, sentiment, and market metrics for cryptocurrencies and stocks.

Finance by LunarCrush Bearer Token active
Overview

The LunarCrush MCP server connects AI agents to LunarCrush's social intelligence platform, which aggregates and analyzes social media activity across X, Reddit, YouTube, TikTok, and news sources to produce structured metrics for crypto and stock assets. It exposes LunarCrush's proprietary signals such as Galaxy Score, AltRank, social sentiment percentages, interactions, contributor counts, and trending topics through MCP so that LLM-based clients can pull live social context into their workflows.

The server is officially maintained by LunarCrush and is available in two forms. The recommended option is a fully managed remote server reachable at https://lunarcrush.ai/mcp (Streamable HTTP) or https://lunarcrush.ai/sse (SSE), authenticated with a Bearer token. The open-source lunarcrush/mcp-server repo provides a Node.js stdio transport that acts as a thin proxy to that remote server, useful for clients that only support local stdio. The proxy forwards all MCP primitives (tools, prompts, resources) from the remote server, so any new capability LunarCrush adds becomes available without updating the local install.

Under the hood the tools surface data from LunarCrush's v4 REST API, covering Topics, Categories, Creators, Coins, Stocks, Posts, and Time Series endpoints. Access requires a LunarCrush API key, which is tied to a paid LunarCrush subscription plan.

Tools

Tool Description
Topics (list, summary, time series, posts, news, creators) Query trending social topics, get 24-hour aggregated activity with sentiment, historical time series, top posts, top news articles, and top creators discussing a topic.
Categories (list, detail, topics, time series, posts, news, creators) Retrieve trending categories, top topics inside a category, aggregated metrics, historical category data, and the leading creators in that category.
Creators (list, detail, time series, posts) Rank trending social creators by interactions, return creator details and topic influence, and pull historical follower/interaction data and top posts.
Coins (list, detail, time series, meta) Snapshot or historical view of tracked cryptocurrencies including price, volume, Galaxy Score, AltRank, average sentiment, and social dominance.
Stocks (list, detail, time series) Snapshot or historical view of tracked stocks with price data and social metrics.
Posts and Searches Look up individual social posts, their interaction trends over time, and run saved or custom searches across LunarCrush data.
Setup Guide

Prerequisites

Option 1: Remote HTTP (recommended)

Point your MCP client directly at the hosted server. No install required.

{
  "mcpServers": {
    "LunarCrush": {
      "type": "http",
      "url": "https://lunarcrush.ai/mcp",
      "headers": {
        "Authorization": "Bearer ${input:lunarcrush-api-key}"
      }
    }
  }
}

An SSE endpoint is also available at https://lunarcrush.ai/sse using the same Bearer auth.

Option 2: Local stdio proxy

Clone the repo and run the Node.js proxy locally. This is useful for clients that only support stdio transport.

git clone https://github.com/lunarcrush/mcp-server.git
cd mcp-server
npm install

Then configure your MCP client:

{
  "mcpServers": {
    "LunarCrush": {
      "command": "node",
      "args": ["<absolute_path_to_project_root>/index.js"],
      "env": {
        "LUNARCRUSH_API_KEY": "<your_lunarcrush_api_key>"
      }
    }
  }
}

The local server proxies all requests to the remote LunarCrush MCP endpoint.

Use Cases
  • Track real-time social sentiment and Galaxy Score for a watchlist of crypto assets and surface notable shifts
  • Identify trending topics, categories, and influential creators discussing a coin or sector before price moves
  • Pull historical time series of social interactions vs price for a coin or stock to feed into a backtest or report
  • Compare two assets (for example BTC vs ETH) across sentiment, AltRank, and social dominance from inside an LLM chat
  • Summarize what is happening today around a specific ticker by combining top posts, top news, and aggregated sentiment
Example Prompts
  • "What is Bitcoin's current Galaxy Score and sentiment, and how has it changed over the past 7 days?"
  • "Show me the top 10 coins by AltRank right now along with their social dominance."
  • "List the most influential creators posting about Solana this week and summarize their take."
  • "Compare ETH and SOL on price, sentiment, and interactions over the last 30 days."
  • "What are the trending crypto topics in the last 24 hours and which news articles are driving them?"
Pros
  • Officially built and maintained by LunarCrush, not a community wrapper
  • Offers both a managed remote server and a local stdio proxy, so it works with any MCP client
  • Exposes proprietary signals (Galaxy Score, AltRank, sentiment) that are not available from generic price APIs
  • Covers both crypto and equities plus the underlying social posts and creators
Limitations
  • Requires a paid LunarCrush subscription to get an API key; there is no free tier for production usage
  • The open-source repo is just a transport proxy; tool definitions live on the remote server and are not enumerated in the README
  • Local stdio proxy is a small JavaScript project with limited public history (low star count, few commits), so debugging relies on LunarCrush support
Alternatives
  • CoinGecko MCP server for price and market data without the social layer
  • Santiment APIs, which offer similar on-chain and social metrics for crypto (no official MCP, community wrappers exist)
  • Generic news/social MCP servers (Reddit, X) combined with a price API if proprietary scores like Galaxy Score are not required