[TP1]Getting started
Sign in, describe your first indicator, and follow the build from your request to the chart.
The Indicator Lab turns a plain-English description into a TradingView indicator. It writes the Pine Script, checks it on real market data, and draws it on a chart you can keep changing.
Signing in
Open the Lab and use Continue with Discord. The Lab is part of the membership. If it won't let you in, the message says why; see Signing in and access.
The launch deck
- Your indicators (the Yours tab): every saved build, with its number of versions, whether it's Own code or Protected, and when you last edited it. Click a name to open it.
- The composer: type your idea and press Build it. That starts a fresh conversation with your idea waiting in the Assistant's message box. Press Enter to send it.
- Continue conversation shows up when you have a conversation going, and takes you back to it.
- Show me around replays a three-step tour. It runs by itself on your first visit.
- Start from Pine brings in a script you already have. See Starting from your own Pine.
Describing an indicator
Say what it should mark, on which market, and when. “Mark the first fair value gap that forms after the 9:30 New York open each day, and shade the RTH session lightly” gets a better build than “FVG indicator”. A new conversation offers four starters: First FVG after the open, Prior day high/low sweeps, NQ / ES SMT divergence and Killzone highs & lows. Click one to send it as-is.
If the Lab needs something from you, it asks one question with a few options. One is marked Recommended. Pick an option, or use Other… to type your own answer. It asks at most two questions in a conversation, then goes with its best assumptions.
Watching the build
A build card tracks four stages: Queued, Draft, Verify and Finish.
| The card says | What it means |
|---|---|
| Waiting for a builder | Your build is queued and starts shortly. |
| Writing the Pine indicator | The code is being drafted. |
| Verifying on market data | The script is compiled and checked on real bars. |
| Repairing the indicator | A check found a problem and the Lab is fixing it. |
| Saving the checked indicator | Last checks, then it's saved to your library. |
| Build complete | It's on the chart and saved to Your indicators. |
| Build stopped | You pressed Stop. Your last saved version is untouched. |
| Build needs attention | It hit a problem you can retry. Press Try again. |
| Build could not finish | It can't be retried as-is. Reword the request and send it again. |
As the card says, it's “Safe to leave this tab after the build is accepted.” Close the tab or press Continue in background and the build keeps running on our side; the Lab reconnects when you come back. If the build's record is gone by then, the card says “That build record expired or could not be found.” Press Try again.
When it's done
The result lands in the chat and on the chart. The reply can include a How I read your request card: what the indicator will show, the question it answers, whether it draws On the price chart or in a Separate pane, and an Assumptions list. If an assumption is wrong, say so in your next message. Below it, the result card shows the version, the verification result (for example “Sample checks passed”, or “Partially checked” when only some checks ran) and a Revise button.
The workspace
- Assistant: the conversation. Ask for changes or ask questions here.
- How it works: the indicator's rules in plain language. See Changing a build.
- Chart: your indicator on market data. A new session opens on NQ.
On a phone, the tabs switch between them. Above the chart: the indicator's name (click it for settings), an eye to hide it, the market picker and timeframes, with more behind ⋯. The bar at the top adds Replay, Backtest and Finish (Code, Publish) once you have an indicator. The header holds Community, Saved, New indicator, notifications and your usage meter.