Skip to main content
Guide2 min read

TradingView MCP server: is there one, and how to connect (2026)

There's no official TradingView MCP yet — here's the honest landscape of community options for charts, screening and market data, and how to use them safely.

Searches for a "TradingView MCP server" are rising fast — so here's the honest answer: there's no official TradingView MCP yet. What exists is a set of community servers that surface TradingView-style data (screening, indicators, symbol info). This guide covers the realistic options and how to use them responsibly.

Not financial advice. These tools surface data; they don't tell you what to trade, and an AI's market "opinions" should never be acted on blindly.

The honest landscape

TradingView doesn't publish an official MCP server. Community projects fill the gap, typically wrapping:

  • Screening / scanners — filter symbols by criteria (price, volume, indicators).
  • Symbol & indicator data — quotes, technical readings, fundamentals.
  • Chart snapshots — some servers render or link chart images.

Quality and maintenance vary widely, so vetting matters more than usual here — see how to vet an MCP server.

Typical setup

Most community servers run via npx or uvx and read data without an account; some need a TradingView session or a third-party data key. A representative shape:

{
  "mcpServers": {
    "tradingview": {
      "command": "npx",
      "args": ["-y", "<community-tradingview-mcp-package>"]
    }
  }
}

Swap in the exact package from the repo you choose, and read its README for any required cookies/keys. Restart the client; see MCP config file location.

Verify

Ask: "Screen for large-cap stocks above their 200-day moving average and summarise." Real screening data back means it works — treat the output as data, not a recommendation.

Safer alternatives for market data

If you mainly want reliable market data into an agent, a general data server (e.g. Bright Data) plus a spreadsheet server is often more dependable than an unmaintained TradingView wrapper.

Use responsibly

Read-only data is fine; never wire an AI to place trades on its own. Review any community finance server's code and permissions before running it. See MCP security best practices.

Going further

Browse the finance category or curated loadouts. For brokerage research (not advice), see Robinhood MCP.

Loadout

Build your AI agent loadout

The directory of MCP servers and AI agents that actually work. Pick the right loadout for Slack, Postgres, GitHub, Figma and 20+ integrations — with install commands ready to paste into Claude Desktop, Cursor or your own stack.

© 2026 Loadout. Built on Angular 21 SSR.