What data sources does Cryptoscholar use?
Data is primarily sourced from Binance, with CoinGecko as a fallback. Market context data (stablecoin supply, etc.) comes from DefiLlama.

@cryptographer11
基本情報
Crypto technical analysis inside Claude. Ask about any coin — get live TA from Binance (EMA, RSI, MACD, ADX, Bollinger Bands, TSS score), macro market context (BTC dominance, ARS, MRS), coin ranking, and a grounded bull/bear debate. No API key needed except for the debate tool.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"cryptoscholar": {
"command": "cryptoscholar",
"env": {
"ANTHROPIC_API_KEY": "your-key-here"
}
}
}
}Full TA snapshot: EMA-20/50/200, RSI-14, MACD, ADX-14, ATR, Bollinger Bands, Historical Volatility, and TSS composite score (0–100). Data from Binance (CoinGecko fallback).
Rank a watchlist by Trend Strength Score. Pass a list of symbols, get them back sorted with full TA data.
Macro signals: BTC dominance trend, ETH/BTC ratio, TOTAL3 altcoin market cap, stablecoin supply (DefiLlama), ARS (Altcoin Rotation Score), MRS (Market Readiness Score).
Grounded bull/bear debate from live indicator values. Returns bull case, bear case, and a one-sentence bottom line.
Cryptoscholar is an MCP server that brings crypto technical analysis directly into Claude. It provides a suite of tools for coin analysis, ranking, market context, and bull/bear debates, pulling live data from Binance (with CoinGecko fallback) and DefiLlama. It is designed for traders and analysts who want to avoid switching between charts or copying data.
Install the package with pip install cryptoscholar, then configure it as an MCP server in your Claude environment. Once connected, you can invoke any of the exposed tools—such as analyze_coin, rank_coins, market_context, or debate—directly through Claude’s interface.
Data is primarily sourced from Binance, with CoinGecko as a fallback. Market context data (stablecoin supply, etc.) comes from DefiLlama.
The tool returns EMA-20/50/200, RSI-14, MACD, ADX-14, ATR, Bollinger Bands, Historical Volatility, and the Trend Strength Score (TSS, 0–100).
Install it via pip: `pip install cryptoscholar`. No additional configuration is required beyond setting up the MCP server in Claude.
Yes, use the `rank_coins` tool. Pass a list of coin symbols, and it returns them sorted by Trend Strength Score with full technical analysis data.
Yes, the `market_context` tool delivers macro signals including BTC dominance trend, ETH/BTC ratio, TOTAL3 altcoin market cap, stablecoin supply, Altcoin Rotation Score (ARS), and Market Readiness Score (MRS).

Let your AI send invoices and take payment — card or ACH. Free. Holdings MCP server for Claude, ChatGPT & Cursor.
Ledger service for AI agents. Each bot gets its own SCDO x402 wallet (USDC on Base), pays 0.01 USDC per call, writes signed work rows to the public SCDO Agent Ledger, and gets an ERC-8004 identity. One line: uvx scdo-age

Memecoin rug pull forensics for AI assistants. Scan any token on Solana, Ethereum, Base, BSC, Robinhood, Arc or Hyperliquid for launch bundles, snipers, insider networks, funding origins, dev wallets and a 0-100 rug scor

Household finance planner for AI assistants: cash flow, net worth, bills, debts, savings goals and retirement projections from your own Beyond Payday data. OAuth sign-in; read-only or read & write.

Build, backtest and diagnose algo trading strategies in your own Tradetron account in plain English. Remote server with OAuth sign-in; it never places a live order.
コメント