The premarket skill

---
name: premarket
description: Prepare for a crypto trading day in ATAS through the ATAS MCP server: levels of the previous day or a chosen session, the current day's range, markup on a separate tab and a short summary. Use when the user asks to get them ready for the day or a session, do a premarket, mark today's levels, or build a day plan for an instrument ("get me ready for the day on BTCUSDT", "premarket for ETH for the US session"). Not for signals, entries or placing orders.
compatibility: Requires the ATAS MCP server and a connection to a crypto exchange. AI Connect gives the agent market data for crypto.
metadata:
  version: 0.2
  updated: 2026-11-05
---

# premarket

Get the trader ready for the trading day: collect the levels, draw them on the chart and give a one-screen summary. Answer in the user's language. You describe data. Do not say where price will go, and do not suggest entries, stops or position size. Never place orders.

## Hard rules

- Do not change the user's open charts. Work on a separate tab. If the tab already exists, ask whether to update it.
- Take every number from ATAS tools. No estimates from memory or from a screenshot.
- If there is no data, say why and stop. Do not substitute another instrument without the user's consent.
- Set alerts only if the user asked for them.

## 1. Clarify the request

You need the instrument and what counts as the previous day. Crypto trades around the clock, so by default a day is the UTC day, 00:00-24:00. If the user trades a specific session, for example the US, European or Asian session, take its hours and convert them to UTC, accounting for daylight saving time. If the exchange has both spot and perpetual futures, ask which one is needed. Ask once for whatever is missing. Unless the user says otherwise, use a 5-minute period.

## 2. Open the working tab

1. `workspace_get_overview`, then `instruments_search`. Pick the instrument from the user's crypto exchange connection. Open the chart with `modules_open` with `placement: "newTab"` and `tabName: "Premarket"`, depth via `timeFrame: {"minutes": 5, "daysToLoad": 7}`.
2. Read `dataSource` from `chart_get_bars` and name the exchange in the summary.
3. If ATAS does not return market data for the instrument, say that through AI Connect it is available for crypto, and suggest a crypto instrument from the user's connection.

If the chart has a simulator position or orders, do nothing with them and do not mention them in the summary.

## 3. Collect the levels

Find "Volume Profile & TPO" with `indicators_search`, add it to this chart with a period that matches the chosen day or session, and read `indicators_get_values`. Do not recalculate POC and value area from bars.

Collect:

- high, low and close of the previous day or session;
- POC, VAH and VAL of that day or session;
- high and low of the current day from its start up to the calculation time;
- average range of the last five completed days or sessions; weekends count as ordinary days unless the user says otherwise;
- where price is now relative to these levels.

State the day or session boundaries by time. Bar times are in UTC; the chart offset is in `chartTimeZoneOffset`.

## 4. Draw on the chart

In one `tools_batch`, add horizontal lines (`drawing_objects_add`, `HorizontalLine`) with labels like "Yesterday POC 78000". Show the markup with `chart_show_range`. Check the result with `module_screenshot`.

## 5. Summary

```
Instrument: <instrument, exchange, period>
Day for the levels: <date, start and end, UTC>

Level              Price   From current price
Yesterday high
Yesterday low
Yesterday close
POC
VAH
VAL
Today high
Today low

Average range of 5 days: <points and %>
Today's range vs average: <%>
Where price is now: <one sentence, no forecast>
Marked on the chart: <list>
```

End with one line: this is markup of the data; the trading decision is the trader's.

## 6. Cleanup

Say that everything can be undone with the phrase "undo the markup". If the user asks, use `changes_undo` and close the tab with `modules_close`.