TradingView MCP
Live prices, 30+ technical indicators, six backtestable strategies and a walk-forward overfitting check, through one MCP server.
- What
- Live prices, 30+ technical indicators, six backtestable strategies and a walk-forward overfitting check, through one MCP server.
- Cost
- Free
- Needs
- Use "TradingView MCP" with your Muse.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
⚠️ **Trading warning / Avertissement trading** : informational only, not investment advice. Indicators describe the past, backtests are run on historical data, and a strategy that looks robust in a walk-forward window can still lose money live. Curated by Skill Harbor: the most-starred repository of the October 2026 trading scan (4,973 stars), an MCP server that gives an AI assistant a trading intelligence toolkit: live prices for stocks, crypto, ETFs, indices and FX (plus the Turkish BIST and Egyptian EGX markets), more than 30 technical indicators (RSI, MACD, Bollinger Bands, Supertrend and more), backtesting for six built-in strategies with trade logs and equity curves, a walk-forward mode that scores robustness and flags overfitting, licensed news sentiment from Marketaux, confluence analysis that combines technicals and sentiment, and multi-market screeners (bullish, oversold, exchange-specific). The repo's single SKILL.md lives under openclaw/ and is written for the OpenClaw agent harness, driving the tools through a bash wrapper script; the server itself is a standard MCP server, hosted or self-hosted, and works from Muse and other MCP clients. From the atilaahmettaner/tradingview-mcp repository (MIT). Honest caveats: prices and screener data flow from third-party endpoints that can change shape, lag or rate-limit without notice, so treat any single reading as a quote to confirm, not a fact; the sentiment layer depends on the Marketaux service; and the skill file's own behaviour rules push the agent to answer fast and combine signals, which is a style choice to keep on a leash when real money is anywhere near the output. Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Copy the install package below, then paste it into MuseThe install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Use "TradingView MCP" with your Muse. Prerequisites: the tradingview-mcp server from the repo, either the hosted option or a self-hosted install (see the repo README for the MCP client configuration), and Python 3 if you use the skill file's bash wrapper. The news sentiment tools rely on the Marketaux service; a self-hosted setup may ask for your own Marketaux API key, which belongs in an environment variable, never in the chat. 1. Open the skill: https://github.com/atilaahmettaner/tradingview-mcp/blob/main/openclaw/SKILL.md and copy the full SKILL.md text. 2. Paste it into a chat with Muse and add: "Give me the current price and a technical analysis for AAPL, then backtest the RSI strategy over one year and run the walk-forward check on it." 3. Read the backtest metrics the honest way: trades count, win rate, max drawdown and the walk-forward robustness score, not just the headline return. Tip: ask for the confluence view (technicals plus sentiment) instead of a single indicator; one signal alone is how bad entries happen. Safety: a skill is plain-text instructions and an MCP connection; the server only reads market data and places no orders. Informational only, not investment advice.
Saved to your recent installs. Find it anytime on /connect.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.