# tradingview-mcp [Health: Active]

**Category:** 💰 Finance & Fintech  
**Repository:** https://github.com/atilaahmettaner/tradingview-mcp  
**GitHub Stars:** 4614  
**npm Downloads (last month):** 1632  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/tradingview-mcp

## Description
Real-time market data, screeners, technical analysis & backtesting for stocks, crypto and forex.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

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

## Documentation

## What the tradingview-mcp MCP server does

The tradingview-mcp MCP server exposes market research and technical-analysis functions to MCP-compatible AI clients. Its scope includes stocks, cryptocurrencies, foreign exchange, and futures across global exchanges. The README describes 37 tools covering live market data, technical indicators, screeners, financial news, sentiment, multi-timeframe analysis, and backtesting.

Example workflows include finding the top crypto gainers on an exchange, analyzing a stock such as NVDA, reviewing gold across several timeframes, or testing an RSI-based strategy on BTC. The server is intended for informational and educational analysis. It does not place trades, manage money, or provide investment advice.

## How it works

Requests are handled through server-side access to public market-data endpoints. The implementation does not sign in to TradingView, scrape a TradingView session, control a browser, or use a TradingView account. That makes the tradingview-mcp MCP server different from MCP projects that automate the TradingView Desktop application.

A hosted deployment is available through a connector URL and can keep responding without a local process running. The hosted version also supports MCP Apps charts: chart requests can render interactive candlestick views in the conversation, with selectable 1-day, 1-hour, or 15-minute intervals and an optional Bollinger Bands overlay. Self-hosting is available as an open-source option.

## Setup and configuration

The project supports Python 3.10 through 3.13 and is published on PyPI under the package name `tradingview-mcp-server`. The README presents two deployment paths: connect an MCP client to the hosted service, or install and run the server locally. The hosted option is aimed at users who do not want to manage Python dependencies; self-hosting does not require a paid TradingView account or TradingView API key.

The hosted service is a paid option with a 3-day free trial and plans described as Pro and Pro+. The repository states that self-hosting remains free. Follow the project’s current installation and client-configuration instructions for the exact local launch command and connector URL settings.

## Tools and capabilities

The tradingview-mcp MCP server includes capabilities for:

- Retrieving prices and extended-hours information
- Ranking market movers and scanning for volume breakouts
- Running technical-indicator and combined analyses
- Comparing multiple timeframes
- Reading financial news and market sentiment
- Backtesting and comparing nine documented strategy types
- Producing hosted interactive candlestick charts with optional Bollinger Bands

Several frequently used operations were converted to asynchronous execution, and the project describes retry handling and short-lived caching for some TradingView screener requests. These implementation details may improve concurrent request handling but do not change the data-source limitations.

## Limitations and notes

Market data comes from third-party public sources and may be delayed, incomplete, inaccurate, or unavailable. Users are responsible for complying with the terms that apply to those data sources. Results such as indicators, scores, signals, entries, stop losses, and targets are computed outputs rather than recommendations. The software does not guarantee trading results, and users remain responsible for their own financial and regulatory decisions.

The project is independent and is not affiliated with or endorsed by TradingView Inc. Its name refers to the data and analysis use case, not to an official TradingView product.

_Full upstream README: https://allmcps.com/mcp/tradingview-mcp/readme_