Max Agent
Articles

Build an alert you can explain: charts, intervals and automation checks

A practical MaxData learning workflow for organizing charts, documenting alert conditions, investigating missing notifications and reviewing automation evidence.

MaxData EditorialPublished Updated

Choose a small learning task: configure a chart, describe one condition, and record what happened when you checked it. This guide proposes an editorial workflow for that task, then extends it to interval comparisons and automation supervision. Treat the example intervals and record formats as exercise conventions, rather than trading parameters.

The product descriptions below come from the supplied MaxData help materials. They are presented as documented behavior, without asserting access under a particular subscription or promising that an interface will match every description. The checklists are our proposals and should be recorded separately from product documentation.

Define the observation before opening settings

Our proposed starting sentence is: “Notify me when this named series meets this written condition.” Keep the exercise focused on configuration and observation. For this workflow, define completion as an understandable record of the rule, the evidence checked and any unresolved questions.

Create an observation sheet with the instrument, trading pair, metric name, metric definition, data provider, unit, selected data field and chart interval. Add the measurement period used by the condition and the horizon you intend to inspect after an event. Include fields for typical publication delay, data limitations and an alternative interpretation. Write “not established” wherever you lack a supported answer.

The official Help Hub introduces MaxData Info Bot as an assistant for questions about MaxData and the workspace. MaxData Help Hub As our proposed support workflow, turn the unresolved entries on your sheet into specific questions. Ask which source supplies the series and where its timing is documented. Retain the answer alongside the configuration rather than silently filling the gap yourself.

Give each chart a defined job

The Max Terminal guide describes saving chart analyses with the top-right ⌄ control, adding indicators through “Indicator” and loading predefined settings through “Load charts.” It also describes multiple time intervals, regular and logarithmic scales, and editing a metric’s appearance or data source by double-clicking its line. Max Terminal guide

For our exercise, assign a chart one job: inspect the series used by the condition. Start with the price series and one additional metric if your question requires it. Name the saved analysis with the instrument, purpose and a version number, such as “BTC condition review v01.” Treat this as your own naming convention, not an automatic product feature.

Before adding another element, write the question it should answer. Record presentation choices separately from data choices. For example, put line color in a presentation section and the selected source in a data section. Our convention is to create a new configuration version whenever a data choice changes. Keep a screenshot or written record with the version so the exercise has a definite reference point.

Compare intervals without rewriting the question

Our proposed comparison uses two views of the same instrument, perhaps a 15-minute view and an hourly view. Choose one observation window with a fixed start and end. Copy the instrument, source, selected field and scale into a short worksheet for each view. Verify the alert settings separately rather than assuming that a chart interval change has updated them.

Use the same question for both views: “What did the selected series look like during the hour before the recorded event?” In the shorter view, write a description of the sections you want to inspect. In the longer view, describe the chosen horizon as a whole. Avoid ranking the views by which appears more persuasive; that ranking is outside this exercise.

Our proposed acceptance rule is that both descriptions must identify their settings and time window. If you cannot establish the same instrument or data source, mark the comparison incomplete. List possible explanations for a mismatch as questions to investigate, such as whether the fields or time boundaries differ. Do not promote an untested explanation into a conclusion.

Proposed exercise

Two views, one observation window

15-minute view
Assigned task
Describe selected sections
Record together
Instrument, source and time window
Acceptance check
Settings identified
Hourly view
Assigned task
Describe the chosen horizon
Record together
Instrument, source and time window
Acceptance check
Settings identified
Example worksheet roles; neither interval is a recommended trading setting. A proposed comparison of a 15-minute view and an hourly view with the same question, source and observation window.

Translate a sentence into an alert

The Pro Alerts guide instructs users to display the relevant metric on the chart, right-click the Pro Charts background and edit a saved alert in the right sidebar. It says at least two alerts must be created before building a Pro Alert. Pro Alerts guide

That guide distinguishes a single-condition Alert from a Pro Alert combining several conditions. It describes a time window for those conditions, sleep time after activation, activation when a condition starts, remains or stops being met, one-time or repeated triggering, expiration and data-field selection. Alert settings

Our proposed build sequence starts with a written rule and a single component. After saving, copy the displayed parameters back into the observation sheet. Check the instrument, field, threshold and measurement period against your sentence. If the available controls do not let you express the intended sentence, stop the exercise and prepare a support question.

Only then prepare a second component. Before combining them, write a separate timing specification: the allowed window, the relevant condition state and the planned end of the exercise. Do not substitute the word “together” for those details in your own notes. For this workflow, require a complete written timing specification before considering the combined configuration ready for observation.

Investigate a missing notification in three passes

Our diagnostic convention separates three questions: did the observed data match the written rule, is there evidence of a trigger, and is there evidence of message receipt? The Max Terminal guide describes chart markers showing when an Alert or Pro Alert triggered. Trigger markers

In the first pass, compare the series and saved settings. Mark each entry as matched, different or unknown. Include the selected data field, threshold, measurement period and any combined-condition timing. Keep the observation window fixed during this pass.

In the second pass, look for available trigger evidence. Record its timestamp and the configuration version you are reviewing. If evidence is missing, write “trigger not confirmed.” Do not write a delivery diagnosis solely because no message arrived.

In the third pass, document receipt separately: selected destination, saved notification settings, event time and the time you checked for the message. Our proposed testing rule permits one configuration change per attempt. Give the attempt a number and write the expected result before recording the observed result. Use a separate row for unresolved behavior rather than repeatedly editing the same note.

Proposed diagnostic workflow

Check rule, trigger and receipt

  1. 1

    Rule

    Compare the observed series with the saved parameters.

  2. 2

    Trigger

    Record available evidence and the configuration version.

  3. 3

    Receipt

    Document the selected destination and what you found there.

Keep each result separate and use unknown when evidence is unavailable. Three proposed diagnostic passes covering the saved condition, evidence of a trigger and evidence of message receipt.

Read a backtest as a historical report

The English Backtests guide describes checking how often an Alert or Pro Alert would have triggered in the past. It describes average and median price-change lines around the event, testing one or several conditions, and selecting a backtest period. Backtests guide

Our proposed reading order begins with the sample period and trigger count, then the observation horizon and reported lines. Record those details before writing an interpretation. Keep the meaning of the exercise narrow: describe what the report shows for the selected historical configuration.

Add a second review sheet for assumptions: fees, slippage, liquidity, instrument or market match, sample size and the possibility of fitting choices to the historical period. These are questions in our proposed evaluation checklist; this guide does not establish how the backtest models each item. Mark an assumption unresolved if you cannot find a documented answer.

Before changing parameters, write a review rule for another period. State which observations would lead you to reject the working interpretation. In our editorial convention, a historical description and a future expectation must occupy separate fields. Leave the future-expectation field blank for this configuration exercise; do not turn the historical report into a promised return.

Review automation as a chain of evidence

The Max Terminal guide describes trading automations based on Pro Alerts. Automation overview The following steps are our proposed supervision protocol, rather than instructions to place an order.

First, document the alert version, intended action, permitted instruments, named reviewer and stopping procedure. Require a supported answer for each item before proceeding beyond analysis. Ask whether a simulation or non-executing test option exists; do not assume that it does.

For an observed event, prepare four entries: data observation, trigger evidence, action confirmation and execution result. Record available timestamps and identifiers. Where you lack access to an entry, mark it unknown. For this workflow, confirmation at one entry does not count as your review of the next entry.

Finish with a reconciliation note: what you intended, what you observed, what remains unresolved and which configuration version you inspected. Preserve that note with the chart and alert records. Define successful completion of this learning task as an evidence-based review that another reader can follow, without requiring a profitable trade or a particular market direction.

Proposed review sequence

Follow the event through your record

  1. Observation

    Record the data and timing available for the event.

  2. Trigger

    Attach available evidence to the saved alert version.

  3. Action

    Record any confirmation of the intended action.

  4. Result

    Reconcile available execution evidence with the written intention.

This is a documentation protocol, not an execution instruction or a claim about available logs. Four proposed stages of an automation review: observation, trigger evidence, action confirmation and execution result.

Sources

  1. MaxData Help Hub – Pomoc i przewodniki Mój workspace — agent.maxdata.app; accessed October 27, 2026
  2. Max Terminal — agent.maxdata.app; accessed October 27, 2026
  3. Pro Alerts — agent.maxdata.app; accessed October 27, 2026
  4. Backtests — agent.maxdata.app; accessed October 27, 2026

Start working with your own agent

The agent knows this article context and still follows the assigned department checklist.

Create a MaxData account
Max Agent
MaxData charts, alerts and automation review workflow