TradingView AI Copilot: What It Can and Can't Tell You About Your Strategy

TradingView AI Copilot is an AI assistant built into the side panel of TradingView's charts that can set up and mark up charts, answer questions from TradingView data, write and apply Pine Script, and read a strategy's backtest report to explain numbers such as profit factor, max drawdown and win rate. It launched on 28 September 2026 as a paid monthly add-on for Essential, Plus, Premium and Ultimate plans, on web and desktop but not yet in the mobile apps.

For strategy traders, the useful question is narrower than "is it good?". Copilot can get a strategy onto your chart and give you a plain-language read of its report quickly. What it reports is the strategy's headline results. The questions that decide whether you should trade a strategy, such as where it loses, whether the edge survives costs and a period it never saw, and how live fills compare, still need work outside the chat. This guide covers what Copilot does, how to use it on a strategy, where its answers stop, and how it compares with connecting Claude through TradingView's MCP server.

What TradingView AI Copilot is

TradingView announced AI Copilot on its blog on 28 September 2026. It sits in the right-hand panel of Supercharts and in the desktop app, and according to TradingView's getting started guide it "sees what's on your chart" and works from TradingView data: quotes, indicators, financials, news, earnings, economic calendars and screener data.

The facts that matter before you pay, as of 2026-10:

Two details from the usage page are worth knowing if you plan to use it for Pine Script. Pine Script requests run on higher-cost models and use the most tokens, along with "complex analytics". And when you hit the limit, new requests pause until the allowance resets; you can't buy more yet. Some accounts have been given a 30-day promotional trial, which TradingView assigns and which expires on its own.

Not the Chrome extension. In April 2026 TradingView released a free AI Chart Copilot browser extension in public beta. That post now says the extension "is no longer actively supported and will eventually be discontinued" and points users to the built-in Copilot. Reviews describing a free extension are describing that older beta, not the built-in add-on.

What it can do

TradingView's help centre lists these jobs:

• Chart setup: symbols, timeframes, indicators, layouts, visible range and theme.
• Chart markup: support and resistance, chart patterns and market structure.
• Research: technical analysis, fundamentals, news, sentiment and macro context, and plain-language screening.
• Trading plans: structured plans, including position sizing.
• Pine Script: writes indicators and strategies and applies them to the chart.
• Backtest review: reads a strategy's report and explains the results.
• Watchlists and alerts: creates and manages them.

The help centre is also clear about limits: it isn't financial advice, it "can be wrong or use outdated information", so verify important figures (help centre), it relies on the context it has (your messages and the active chart), and it handles one request at a time in a chat.

How to use Copilot on a strategy

TradingView's page on creating indicators and strategies with AI Copilot describes a three-part loop: write, optimise, review.

1. Open the chart you'll trade. Set the symbol and timeframe first. Copilot works from the active chart, and the help centre recommends including the ticker and timeframe in your prompt.
2. Describe the strategy precisely. TradingView's example is a long entry when the 20 EMA crosses above the 50 EMA, exiting on the opposite cross. Be specific about entries, exits, stops, timeframe and any inputs. Copilot writes the Pine Script, checks it for errors and applies it to the chart. If it fails to compile, ask Copilot to fix it; it can read the error itself.
3. Set realistic properties before reading anything. Ask Copilot to add commission and slippage that match your broker, or set them yourself in the strategy's properties. Left at zero, they flatter any strategy that trades often. Our guide to why strategies work in backtest but not live covers the settings that matter.
4. Ask for a backtest review. According to TradingView, Copilot reads the backtest report, explains profit factor, max drawdown, win rate and average trade, and then judges whether the strategy looks viable based on your prompt.
5. Be careful with optimisation. The same help page says Copilot can test different parameter combinations and report which performed best. That's convenient, and it's also the fastest route to an overfit strategy: the best of many combinations on one stretch of history is partly the best fit to that history's noise. Keep a period the optimisation never saw, and see our overfitting checklist.

Useful prompts at the review stage are the ones that make it check rather than praise:

Review this strategy's backtest. How many trades is it based on?
What happens to profit factor if I double commission and slippage?
Which three trades contribute the most profit, and what is the
profit factor without them?

Whether Copilot can answer the second and third questions depends on what it can reach from the report on your chart. TradingView's documentation names the summary metrics and doesn't say how far it goes into the individual trades, so check its answers against the Strategy Tester's List of Trades.

What it can't tell you about your strategy

None of the following is a criticism of Copilot as a chart assistant. They are questions a chat over one backtest report isn't set up to answer, and TradingView's documentation doesn't describe Copilot doing them.

Breakdown by market condition

The single profit factor in a report is an average over every condition the strategy traded in. A strategy with a respectable overall number can be losing steadily in the Asian session and carried by London and New York, or making all its money on high-volatility days. You can't see that from the summary metrics TradingView says Copilot explains. You need each trade tagged with the conditions it happened in and summarised per bucket, with the trade count beside each result. How to analyse a TradingView backtest by session, weekday and hour shows how to do this in a spreadsheet from the List of Trades export.

Backtest vs live

Copilot reviews the backtest report on your chart. A strategy's real test is how its live or paper fills compare with that backtest over the same weeks: entries that slipped, exits that filled later, trades the backtest took that never triggered live. That comparison needs your broker's trade history next to the backtest, trade by trade.

Testing rule changes on the same trades

Copilot can edit a script and rerun it, which tells you how the modified strategy does on the same history. That's useful, but every rerun is another look at the same data. A cleaner first step for a stop, target or filter change is to apply it to the trades the strategy already took and see which ones it would have changed, then confirm the change on a period you didn't use to find it.

Where a dedicated analysis fits

This is the part Tessera is built for. You export the Strategy Tester's List of Trades, import it into Tessera, and the reports break the results down by session, weekday, time of day, volatility and symbol, with the trade count behind each slice. The simulator replays stop, target and filter changes on real candles against the trades the strategy took, and rules that hold up can be exported into a new copy of your Pine Script. Live trades can come in from broker sync (Liquid Brokers, Interactive Brokers, OANDA, cTrader and Binance) or file import, so you can compare them with the backtest.

Tessera's journal doesn't edit a Copilot strategy or generate new backtest fills for it; it analyses trades the strategy has already taken. (Tessera's Strategy Lab is a separate no-code backtester for strategies you build there.) In practice that makes it a second step after Copilot rather than a replacement: let Copilot get the strategy onto the chart and summarise it, then check where it makes and loses money.

Copilot vs connecting Claude through TradingView's MCP server

TradingView also runs an official MCP server that lets assistants such as Claude and ChatGPT reach your account from outside TradingView. Its documentation lists 35 tools across watchlists, market data, symbol search, screener, news, fundamentals, documents, calendars and alerts, included in Essential plans and above (not trials). It has no Pine Script or Strategy Tester tools.

For strategy work on the chart, Copilot is the more direct tool. The MCP route suits research and monitoring workflows in an assistant you already use, and some traders write Pine Script in Claude and paste it into TradingView. Our guides cover connecting Claude to TradingView and the differences between the official and community TradingView MCP servers, including desktop bridges that can reach the Strategy Tester with unofficial methods and real caveats.

Is TradingView AI Copilot worth it?

It depends on what you'd use it for, and TradingView doesn't publish the price on its help pages, so check it at checkout. As of 2026-10, it's likely worth trying if you:

• spend a lot of time setting up charts, marking levels and pulling news or fundamentals into one view;
• want to turn a strategy idea into compiling Pine Script without learning the language;
• want a quick plain-language read of a Strategy Tester report.

It's less useful if you already write Pine Script comfortably, mostly trade from mobile, or expect it to tell you whether a strategy will make money. TradingView's own page says backtest results "describe the past and don't guarantee future performance". A Copilot-written strategy should go through the same checks as any AI-written strategy; how to test a strategy ChatGPT or Claude wrote has the full sequence.

Trading carries risk, and no AI assistant or analysis tool makes a strategy profitable.

If you've got a strategy on the chart already, export its List of Trades and upload it for a free Entry Analysis report to see its results by session, weekday, time and volatility. No card needed.

For how Copilot compares with other tools that turn plain English into a strategy, see the best AI trading strategy builders.