TradingCalc MCP: Options, Forex, Risk Stats, Prediction Markets, On-Chain & Crypto Futures MCP server

TradingCalc MCP Server

Ask Claude or Cursor trade questions and get exact numbers back, not AI guesses.

"What's my PnL if I buy 0.5 BTC at $80k and sell at $95k with 5x leverage?"

"Size my position: $10k account, 1% risk, long BTC at $83k, stop at $81k."

"Is this carry trade worth it? 0.01% funding long, 0.05% short, $50k, 30 days."

40 deterministic tools across trade planning, risk & margin, funding/carry, market-structure (Market Profile) analysis, chain-agnostic on-chain tools — Solana (token safety, swap price impact, bonding curve) plus Solana + 5 EVM chains for market cap comparison and wallet flag check — prediction-market odds and fair value across Kalshi, Polymarket and ADI Predictstreet's crypto categories, and Deribit BTC/ETH options math (payoff/breakeven, Black-Scholes price/Greeks, straddle/strangle, covered call/protective put, implied volatility). Formulas verified against 35 canonical test vectors: same inputs always produce the same outputs. Every response is also signed with ECDSA P-256, so you can verify offline that it actually came from us. Free, no signup.

Access via MCP (Claude Desktop / Cursor / VS Code) or a plain HTTP POST to the MCP endpoint. Free, no signup.

Endpoints

MCP transport: Streamable HTTP (MCP spec 2024-11-05)

Quick Start

Claude Desktop

Add to claudedesktopconfig.json:

{
  "mcpServers": {
    "tradingcalc": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://tradingcalc.io/api/mcp"]
    }
  }
}

Cursor / VS Code

{
  "tradingcalc": {
    "url": "https://tradingcalc.io/api/mcp"
  }
}

Agent Skill

For coding agents that support the skills.sh ecosystem (Claude Code, Cursor, GitHub Copilot, and others) — installs a SKILL.md that teaches the agent when to reach for these tools instead of estimating trade math itself:

npx skills add SKalinin909/tradingcalc-mcp

Direct HTTP

curl -X POST https://tradingcalc.io/api/mcp \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
      "name": "workflow.run_liquidation_safety",
      "arguments": {
        "side": "long",
        "entryPrice": 95000,
        "leverage": 10
      }
    }
  }'

Example prompts

After connecting, just ask naturally: the AI picks the right tool automatically:

Trade P&L

"I bought 0.5 BTC at $80,000 and want to sell at $95,000 with 5x leverage. What's my net profit after fees?"

Position sizing

"I have a $10,000 account and want to risk 1% going long BTC at $83,000 with a stop at $81,000. How many coins should I buy?"

Liquidation check

"Long ETH at $3,200 with 10x leverage, where do I get liquidated?"

Full pre-trade check

"Analyze this setup: long BTC at $83,000, stop $81,000, target $90,000, $10k account, 1% risk, 5x leverage. Is it worth taking?"

Funding cost

"I'm holding 0.5 BTC long on Bybit at $83,000 with 0.01% funding rate. How much will funding cost me over 3 days?"

Carry trade

"Is this carry trade worth it? Long on Bybit at 0.01% funding, short on Binance at 0.05%, $50k notional, 30 days."

DCA average entry

"I bought BTC at $78k (0.2 BTC), $80k (0.3 BTC), and $82k (0.1 BTC). What's my average entry and breakeven?"

Scale-out plan

"I'm long 1 BTC from $80k. I want to close 30% at $88k, 40% at $92k, 30% at $96k. What's my total P&L?"

Shortened. The full README is on GitHub.