TradingView MCP Server

Connect your tradingview-mcp AI Trading Intelligence Framework to communication channels like Telegram, WhatsApp, and Discord using OpenClaw. This allows you to interact with trading tools via plain language messages from any device.

Prerequisites

• OpenClaw installed and running (openclaw doctor must return healthy).
• A running gateway (e.g., Hetzner VPS or local machine).
• uv installed on the same machine as OpenClaw.

To use the TradingView MCP server with Claude Desktop, add the following configuration to your claude_desktop_config.json file.

Config File Locations:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

This configuration uses uv to run the server directly from the GitHub repository.

{
  "mcpServers": {
    "tradingview-mcp": {
      "command": "uv",
      "args": [
        "tool",
        "run",
        "--from",
        "git+https://github.com/atilaahmettaner/tradingview-mcp.git",
        "tradingview-mcp"
      ]
    }
  }
}

The server supports complex analysis patterns through natural language:

• Multi-Timeframe Analysis: Compare performance across different intervals (e.g., 15m, 1h, 1D, 1W) to check for trend alignment.
• Pattern Recognition: Scan for specific candlestick patterns, such as consecutive bullish or bearish candles.
• Swing Trading: Find assets in oversold territory using RSI and rating filters.
• Day Trading: Identify volatile assets with high volume using Bollinger Band scans.

💬 You: "Find coins showing consecutive bullish candle patterns"
[Uses: consecutive_candles_scan(pattern_type="bullish", timeframe="1h")]

If you want to modify the code or run the server locally, follow these steps:

1. Clone the repository.
2. Install dependencies using uv sync.
3. Configure your MCP client (Claude Desktop or Codex) to point to your local installation.

Claude Desktop Config (macOS/Linux): Use uv run python and specify the cwd (current working directory) to your local clone path.

Claude Desktop Config (Windows): It is recommended to use the direct path to the .venv Python executable for better reliability.

# Clone and setup
git clone https://github.com/atilaahmettaner/tradingview-mcp.git
cd tradingview-mcp
uv sync

# Test the server locally
uv run python src/tradingview_mcp/server.py

You can install the TradingView MCP server directly from PyPI using pip. This provides access to 37 tools including real-time market data, technical indicators, screeners, and backtesting capabilities.

pip install tradingview-mcp-server

Use the server for broader market context:

• Cross-Exchange Comparison: Compare the price of a single asset across multiple exchanges (e.g., KuCoin vs Binance vs Bybit) to identify arbitrage opportunities or best entry points.
• Market Sentiment Analysis: Get an overview of market mood by combining gainers, losers, RSI, and breakout data.
• Sector Rotation: Analyze which categories (e.g., DeFi, Layer 1s, Meme Coins) are currently leading the market.

To get the most accurate and actionable results from the AI assistant, follow these prompting best practices:

The repository includes metadata for the Codex plugin. Use the following configuration to enable the server via the PyPI package.

{
  "mcpServers": {
    "tradingview": {
      "command": "uvx",
      "args": [
        "--from",
        "tradingview-mcp-server",
        "tradingview-mcp"
      ]
    }
  }
}

UV is required for dependency management. Install it using one of the following methods based on your operating system:

• macOS (Homebrew): brew install uv
• macOS/Linux (Direct): Use the curl script.
• Windows (PowerShell): Use the PowerShell command.

Verify the installation with uv --version.

If you prefer to run the server from the local repository using uv, follow these steps:

git clone https://github.com/atilaahmettaner/tradingview-mcp
cd tradingview-mcp
uv run tradingview-mcp

Request a comprehensive technical analysis for a specific symbol. The server provides data on:

• Price Action: Current price, 24h change, highs, and lows.
• Bollinger Bands: Rating (e.g., BUY/SELL), Bandwidth (BBW), and position.
• Technical Indicators: RSI, MACD, and EMA (50/200).
• Trading Signal: A summarized recommendation (e.g., BUY, SELL, NEUTRAL).

💬 You: "Give me a complete technical analysis of Bitcoin"
[Uses: coin_analysis(symbol="BTCUSDT", exchange="KUCOIN", timeframe="1D")]

To avoid timeout errors during the initial Claude Desktop startup, run the following command in your terminal to pre-install the server using Python 3.13. This ensures the virtual environment and dependencies are ready before Claude attempts to connect.

uv tool install --python 3.13 tradingview-mcp-server