tradingview-mcp

Runs price queries, technical analysis, backtests, and sentiment scans across stocks, crypto, and global exchanges.

1|Updated May 6, 2026
One-click install
npx skills add https://github.com/surfingalien/FinSurfing --skill tradingview-mcp-surfingalien
Or copy as Structured Prompt for Agentâ–¼
Please help me install this Agent Skill.
Skill: tradingview-mcp
Source: https://github.com/surfingalien/FinSurfing/tree/main/openclaw
Command: npx skills add https://github.com/surfingalien/FinSurfing --skill tradingview-mcp-surfingalien

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires tradingview-mcp-server, and includes scripts (resource) components.

What problem does it solve? Answering trading questions requires juggling multiple data sources for prices, indicators, backtests, and sentiment. This Skill consolidates live quotes, 30+ technical indicators, six backtestable strategies, walk-forward overfitting checks, Reddit sentiment, and news into a single bash-callable CLI. ## Core Features & Use Cases - Live Prices & Market Snapshots: Fetch quotes for US stocks, crypto, ETFs, indices, Turkish BIST, and Egyptian EGX symbols via Yahoo Finance. - Technical Analysis & Backtesting: Compute RSI, MACD, Supertrend, Bollinger Bands and backtest six strategies with trade logs, equity curves, Sharpe, drawdown, and profit factor metrics. - Overfitting Detection & Sentiment: Run walk-forward validation with robustness scores and combine Reddit sentiment with news for confluence analysis. - Use Case: Ask "Should I buy AAPL?" and get price data, an RSI backtest with walk-forward validation, and current Reddit sentiment in one combined answer. ## Quick Start Ask the assistant to backtest the RSI strategy on AAPL over the past year and include the full trade log.

Frequently Asked Questions about tradingview-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I backtest a trading strategy on a stock?â–¼

Run the backtest command with a symbol, strategy key, and period, for example backtest AAPL rsi 1y. Available strategies are rsi, bollinger, macd, ema_cross, supertrend, and donchian, with optional interval and trade log output.

How do I check if a backtested strategy is overfitted?â–¼

Use the walkforward command, which splits historical data into folds and compares in-sample training returns against out-of-sample test returns. It outputs a robustness score where higher values indicate the strategy generalizes beyond its training window.

Which markets and symbols does the price tool support?â–¼

It supports US stocks like AAPL and NVDA, crypto pairs like BTC-USD, ETFs like SPY, indices like ^GSPC, Turkish BIST symbols like THYAO.IS, Egyptian EGX symbols like COMI.CA, and forex pairs like EURUSD=X.

Why does the trading script fail with an import error?â–¼

The script requires the tradingview-mcp-server package installed via uv tool install. If site-packages are not auto-discovered, the error message instructs you to run uv tool install tradingview-mcp-server to resolve it.

Can I combine technical analysis with Reddit sentiment?â–¼

Yes, the sentiment command analyzes Reddit discussion for a symbol, and the confluence analysis combines technical indicators with sentiment signals. This is recommended for buy/sell questions alongside price and backtest data.