> ## Documentation Index
> Fetch the complete documentation index at: https://takeprofit.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Frequently Asked Questions

> Short answers to the questions we get most often — where the data comes from, how drawings sync, how fast alerts are, and where to start with Indie and MCP.

Answers to the questions that come up most often in [Discord](https://discord.gg/WVk8TjwU7p) and support. Each one links to the full page if you want the detail.

## Charts and drawings

<AccordionGroup>
  <Accordion title="How does drawing sync work?">
    Every drawing has a **sync level** that decides which charts it appears on. You set it per drawing, in the drawing's settings or from the [Object Tree](/docs/guide/platform/chart-widget/Object-tree):

    * **Local** — the drawing stays on the chart you drew it on. This is the default.
    * **Channel** — it appears on every chart in the same [linked channel](/docs/guide/getting-started/Widget-Hub#widget-linking).
    * **Workspace** — it appears on every chart in the current workspace.
    * **Global** — it appears on every chart, in every workspace.

    This is what trips people up most often: a chart you forked and dropped into a workspace as an unlinked widget is its own chart, so a drawing made on it at the default **Local** level lives only there. To use the same level elsewhere, raise its sync level — to **Workspace**, for example — and it shows up on the other charts too.

    Sync works across timeframes as well as across charts, so a level drawn on the 1H stays where it belongs on the daily. See [Sync tool](/docs/guide/platform/chart-widget/Utility-tools#sync-tool).
  </Accordion>

  <Accordion title="Do you use your own charts, or TradingView?">
    Our own. We use open-source components where it makes sense, but the rendering engine (WebGL), the drawings, the indicators, the settings, and the UX are all ours. That is why the charts feel more responsive than an embedded widget — and it is also why indicators are written in [Indie](/docs/indie/What-is-Indie) rather than in someone else's language.
  </Accordion>
</AccordionGroup>

## Market data

<AccordionGroup>
  <Accordion title="Why do volumes on US stocks look different from other platforms?">
    US exchanges charge licensing fees to display real-time quotes, and each licence covers a different slice of the market. For intraday US equity quotes we currently use **CBOE** data — the same category of licence other retail charting platforms use. It is a consolidated view rather than the primary exchange feed, so the volume printed on a bar can differ from a platform quoting a primary feed. Prices track closely; volume is where you see it.

    We are moving coverage onto direct exchange feeds as our brokerage integrations go live. See [Markets We Cover](/docs/guide/market-data/Market-data-overview) for what is available today.
  </Accordion>

  <Accordion title="Where does crypto data come from, and why do I sometimes see an abnormal candle?">
    It depends on the venue. For the exchanges we are integrated with directly — **Binance** and **Bybit** among them — the data comes straight from the exchange. The rest of our [coverage](/docs/guide/market-data/Market-data-overview) currently arrives through a third-party market-data aggregator.

    Abnormal candles almost always come from the aggregated feed rather than from our side: the same print is usually present in the provider's own API. We cannot guarantee the same quality there as on a direct connection, which is why we keep moving venues over to direct exchange feeds one at a time. Trading is offered only on venues we connect to directly.
  </Accordion>
</AccordionGroup>

## Alerts

<AccordionGroup>
  <Accordion title="How fast are alerts? Is there a delay?">
    We measure it end to end — from receiving the tick from the exchange, through triggering the alert on our server, to delivering it to the server that receives your webhook. In our measurements it is **typically around 300 ms, and up to about 1 second in the worst case.**

    That is a measurement of the alert pipeline, not a guaranteed delivery time. Delivery to a messenger or an email inbox adds whatever that channel adds, and per-channel rate limits apply — see [Alert limitations and delivery](/docs/guide/alerts/Alert-limitations-delivery-history).
  </Accordion>

  <Accordion title="Can I put an alert on my own indicator?">
    Yes — an alert can trigger on an indicator value, not only on price. See [Alert trigger criteria](/docs/guide/alerts/Alert-trigger-criteria) for the conditions available.

    Through the AI assistant and the [MCP server](/docs/guide/platform/ai-assistant/Mcp-server-guide#your-private-scripts) this also works on a private indicator of your own: once it is saved to **My scripts**, the alert can be created on it without publishing it to the Marketplace first. The assistant can write the indicator, save it, and set the alert in the same conversation.
  </Accordion>
</AccordionGroup>

## Indicators and AI

<AccordionGroup>
  <Accordion title="How do I port my Pine Script indicators?">
    Give the script to the AI assistant and ask it to convert it to [Indie](/docs/indie/What-is-Indie). It knows the language, compiles what it writes, and fixes the errors it finds. Work in the [AI Chat](/docs/guide/platform/ai-assistant/AI-chat-page), where a script can be opened in a real IDE tab and added to a chart, or in the [IDE assistant](/docs/guide/platform/ai-assistant/AI-assistant-IDE) if you are already in the editor.

    Convert one indicator at a time and say what it is supposed to do — a conversion that keeps the intent beats one that keeps the syntax. Indie is a Python subset, so most Pine logic has a direct equivalent; see [Language differences with Python](/docs/indie/Language-differences-with-Python) for what does not.
  </Accordion>

  <Accordion title="How do I set up the MCP server?">
    The [MCP Server Setup Guide](/docs/guide/platform/ai-assistant/Mcp-server-guide) has step-by-step instructions for Claude, Claude Code, Codex, Cursor, VS Code, and any application that can run `mcp-remote`. In short: add `https://mcp.takeprofit.com/mcp` as a custom connector and sign in when prompted.

    Once connected, your AI application can write and validate Indie code, search the Marketplace, work with your private scripts, manage your alerts, and read and edit your watchlists.
  </Accordion>

  <Accordion title="What is the difference between the AI Chat and the MCP server?">
    The [AI Chat](/docs/guide/platform/ai-assistant/AI-chat-page) is ours: it runs on the platform, draws on the [hosted budget](/docs/guide/platform/ai-assistant/AI-assistant-overview#usage-and-limits) that comes with All-In (or on your own Anthropic key), and hands back charts, screens, and IDE tabs you can keep working with. The [MCP server](/docs/guide/platform/ai-assistant/Mcp-server-guide) puts the same platform capabilities inside an AI application you already use — Claude, Cursor, VS Code — and runs on that application's own subscription.
  </Accordion>
</AccordionGroup>

<Note>
  Question not answered here? Ask in [Discord](https://discord.gg/WVk8TjwU7p) or email [support@takeprofit.com](mailto:support@takeprofit.com).
</Note>