01 · BEGINNERJust getting started
What you can do even without knowing how to analyze the market
Maybe you already have a trading account and buy something now and then. Maybe you act on tips, news or gut feeling. And maybe you're still figuring out how to start.
Reactor is a trading operating system that brings together AI, market data and your trading accounts. You can use it to understand information, test ideas, research methods, and give AI Agents tasks of analysis, execution and management.
You don't need to build everything from scratch. You can start from ready-made examples and adapt them to your needs. The instructions you give decide what the Agent checks, how much judgment it uses, and which actions it's allowed to take.
The examples below show a few ways to use the system.
01
EXAMPLE 1
"I don't know how to analyze. Can the system check it for me?"
Say there's an asset you're interested in. Instead of going through charts and data yourself, you can give an Agent a task: check the market, assess the opportunity and explain its decision.
The Agent can gather information, weigh several factors and decide whether the opportunity fits, or whether it's better to wait. The instructions define its process and how much judgment it uses.
Task
→
Gathering information
→
Reasoned decision
WHAT YOU'LL SEE
The task given to the Agent, the information it checked and its reasoned decision. You'll also see how to ask for a simple explanation of anything that isn't clear.
Questions that may come up
Do I need to know how to write the instructions?
You can start from examples we provide. We'll show what each part of an instruction does and how to adjust it.
Can the Agent also buy?
Yes, when it's given the right tools and permissions. You can also keep it to analysis only, or require approval before any action.
✓ HOW WILL I KNOW I GOT IT?
You can explain what you asked the Agent, what it decided, and whether it was allowed to act.
02
EXAMPLE 2
"I want help with trading, but I don't have time for it every day."
You can run an Agent at set times. On every run it gathers the relevant information and carries out the task it was given.
You can also split the work between several Agents: one finds opportunities, another analyzes and decides, and a third manages open trades. The output of one step can move on to the next.
Opportunity finder
→
Analyzes and decides
→
Trade manager
WHAT YOU'LL SEE
A short process that starts with spotting an opportunity and reaches a decision, with your approval before execution. We'll show where you can give the process more independence.
Questions that may come up
Do I have to approve everything?
It depends on the way of working you choose. You can get suggestions only, approve actions, or allow automatic execution within defined limits.
What do I need to define in advance?
The task, the account it works on, the working hours, the limits on action, and the situations that require approval.
Can I start from a ready-made process?
You can start from ready-made examples. You still need to check that the instructions and permissions fit your account and your goal.
✓ HOW WILL I KNOW I GOT IT?
You can identify which steps the system performs on its own and at which step it comes back to you.
03
EXAMPLE 3
"If I let the system trade, what is it working from?"
A strategy is a way of working that defines which opportunities to look for and how to handle them. Reactor lets you explore that path with data and test each part of it separately.
In a real example, we ran an Agent that scans the market every four hours and spots a specific price pattern. Then we used our research tools to check how price behaved after those signals.
The findings helped us choose a 3% stop-loss and a 4.5% profit target. We connected a buying Agent that buys the signals and collected results. As of preparing this example, by the measurement taken, 800 trades closed with a 65% win rate.
In this case the buying Agent buys every signal. That's a way to test the detection and the trade management before adding judgment to filter opportunities.
4h
Scan and detect the pattern every four hours
3%
Stop-loss target
4.5%
Profit target
800
Closed trades
65%
Win rate
As of preparing this example, by the measurement provided.
WHAT YOU'LL SEE
How a finding in the data led to choosing the targets, and how detection, buying and outcome connect.
Questions that may come up
What exactly is being tested?
What happened after the signals: how far price moved in the trade's favor or against it, how long it took, and how the trades ended.
How do you know what to improve?
Separate opportunity detection, the buying decision and trade management. That way you can see where a change might help.
If the method worked so far, will it keep working?
Results describe the period that was tested. We keep measuring while it runs, to see how the method behaves in other market conditions.
✓ HOW WILL I KNOW I GOT IT?
You can explain which part of the method was chosen based on the data, and which part can still be tested and improved.
04
EXAMPLE 4
"I saw the strategy was profitable. How much money do I need for it?"
To connect the results to your money, you need to understand how they were produced too.
A results report shows how many trades closed, how many were profitable, the size of wins versus losses, and what was left after costs. Then you look at how much money was allocated per trade, whether leverage was used, and how many trades were open at once.
The same price movement can produce a different money result when the trade size changes. Increasing the amount or the leverage also affects the possible loss.
Method performance
Closed trades · profitability · win vs. loss · costs
Your money
Amount per trade · leverage · trades open at once
WHAT YOU'LL SEE
A results report with an explanation of the key figures, then a comparison of the same example at different trade sizes.
Questions that may come up
Does a 65% win rate mean the method is profitable?
You also need to check how much the winning trades make, how much the losing ones lose, and what the costs are.
Does the number of trades tell me how much money I need in the account?
You need to know how many trades were open at the same time. The same money may be reused for more trades after a trade closes.
What's the difference between the cost of Reactor and the money I trade with?
The payment for the system covers using it. The trading capital sits in your own account and is used for market activity.
✓ HOW WILL I KNOW I GOT IT?
You can tell the difference between the method's performance and the effect of amount and leverage on the money result.
05
EXAMPLE 5
"After I buy, who keeps track, and what happens when things change?"
Managing a trade can be a task of its own. An Agent can check an existing position, follow market conditions and decide on an action according to its instructions and authority.
For example, you can give an Agent the role of risk management, protecting profit or partial take-profit. You can define precise rules, and also allow judgment in situations defined for it.
Risk management
Protecting profit
Partial take-profit
WHAT YOU'LL SEE
An open trade, a change in its situation and the management Agent's response — including the data and the reasoning that led to the action.
Questions that may come up
Does the Agent that buys also have to manage the trade?
You can separate the roles, giving each Agent its own task and permissions.
Can it decide to exit?
Yes, if it was given authority and according to the instructions. You can also have it suggest an exit and wait for approval.
Does risk management prevent losses?
It helps manage the risk. Managed trades can still lose, and actual execution depends on market conditions.
✓ HOW WILL I KNOW I GOT IT?
You can explain what the management Agent is allowed to change, and when it has to ask for your approval.
06
EXAMPLE 6
"If it works for me, how do I stay in the picture?"
Reactor lets you look through Agent runs and see the information they used, tool results, decisions and actions. You can also review failures and usage costs.
That way you can go back to a specific case and understand the path to the outcome: what was found, what information was available, what the Agent decided and what was executed.
What was found
→
What information was available
→
What was decided
→
What was executed
WHAT YOU'LL SEE
One run, from the task to the outcome, alongside the action in the account and the cost of the run.
Questions that may come up
Do I have to go through every run?
You choose your level of involvement. The record lets you check specific cases and assess the process over time.
What do I do if the outcome isn't good?
Check the data, the instructions, the decision and the execution, to understand what needs to change.
✓ HOW WILL I KNOW I GOT IT?
You can find what the Agent did in the case shown, what information it relied on, and what the outcome was.
SEE IT LIVE
Want to go through these examples with us?
In a live demo we'll open the system and walk through the examples that fit you, at your pace.
ANOTHER TYPE?
Not quite you?
Go back and pick the description closest to where you are today.
