Judgment
What it decides, and what it refuses.
A case is put against every candidate before capital moves, and what it decides is ranked against everything else it could be doing instead.
- macOS
- Windows
- iOS
- Google Play
AI Agentic Trading
A trading desk that
thinks for itself.
AI automated agentic trading desk for macOS, iOS, Android and Windows. 17+ trading strategies simultaneously scan and compete for one pool of capital, under pre-trade risk checks and protective orders you set. The program scans, decides, enters, exits, judges and adjusts, completely automated and intelligently.
Autonomous trading involves substantial risk of loss, including loss of all capital connected and, in margin accounts, more. TradeAgentic is software you license and run; RLG, LLC does not manage accounts or give personalized advice. Read the risk disclosures before licensing.
Judgment
A prosecutor stands between
the green light and your capital.
Clearing the entry rules is not permission to trade. It is permission to be argued with. Every surviving candidate has a case built against it from evidence the desk already holds, and only those that survive it are funded.
Every refusal is recorded and scored later against what the market actually did, so the cost of saying no is a measured quantity rather than an article of faith.
Its accountability is structural, not aspirational.
Every refusal is carried through the same lifecycle as a funded trade, so what it saved or cost becomes a measured quantity rather than a claim. That record is reported, and the desk answers to it: refusing is held to the same standard of evidence as acting.
Nothing here waits on a human.
There is no approval queue, no discretionary override and no setting that lets an operator wave a trade past the chain. You set the account and the risk limits. It runs the desk, whether you are watching or asleep.
Decision loopIdle
Observe·
Judge·
Act·
Grade·
StandbyEvery candidate is argued against before capital moves. The ones that survive reach your account with a stop already resting at the broker. Both outcomes are recorded and later graded against what the market actually did.
Illustrative and simplified. Not a live feed, not the decision path itself, and not a recommendation regarding any security shown.
Intelligence
It knows
how much it knows.
Almost no trading system can tell you whether it is getting better, because nothing in it is keeping score of itself. This one carries a standing measure of its own competence, and it keeps the history so that change is something you can see rather than something you argue about.
It is assembled from measures of the desk’s own record, computed from the same history it trades on, and that history is kept so that improvement is something you can see rather than something you argue about.
Every part of it is computed from the desk’s own recorded outcomes. Nothing in it is a matter of opinion, nothing responds to how the last few days felt, and nothing moves the measure except outcomes that have actually been recorded.
Shown in outline. How the measure is built is provided to licensees.
The agent loop
Every trade earns its way
to your account.
A desk that thinks for itself is a claim, so here is what it means. The software forms its own view on every candidate, tests that view against the evidence it already holds before any capital moves, and afterwards finds out what the market actually did about the decision. No discretionary override and no trade that bypasses the checks. Every decision is recorded as it happens, whichever way it went.
Grading its own decisions is the part most systems never attempt, which is why most of them cannot tell you whether they are getting better. Almost every self improving system learns only from what it took. This one also learns from what it declined, which is the great majority of what a disciplined desk does.
Execution
The edge has to survive
the fill.
Expected edge is a claim. What the account is left with after costs is the result. The desk answers to the result.
Stated as a principle. How it is enforced is provided to licensees.
How it works
Three things it does.
All of them measured.
They divide into what the desk decides, what it remembers, and what it changes about itself. Each is scored by the machinery it supervises, and each earns its authority from its own record.
Memory
What it keeps, so that it can learn at all.
Every yes and every no it has ever made, kept and later scored against what the market actually did. Exits are scored as carefully as entries.
Learning
What it changes about itself, and on what evidence.
Capital moves toward what its own record says is working. It re-examines its own record continuously, and reports what it finds.
Bedrock
What no automated part of
this system may widen.
Under every autonomy mode, including full automatic. Only you can change these, and every change is recorded. The desk may tune some operating parameters within fixed ranges documented for licensees; it may not loosen anything below.
- Protective stops
- Placed at your broker, not held in this software. A stop does not guarantee a price in a gap or halt.
- The kill switch
- Always one action away, and nothing may gate it.
- The daily loss stop
- Your limit for the day. No AI layer may widen it.
- The profit target rule
- The target ends the day. It never trades bigger to win it back.
- Pre flight refusal
- Trading is blocked until it passes. In live mode it places and closes one small real order, about $10.50, at your cost.
- Account level caps
- How much of the account each asset class may use is set by you.
- The implausible order guard
- An order that does not make sense never reaches the broker.
The charter
A change that adds expected profit while weakening a loss bound fails this charter.
Governing document. Every audit the machine performs on itself is tested against it.
Every judgment the system makes, whether a control change, a veto, an audit finding or a line of code it proposes, has to justify itself: what it is for, what graded evidence supports it, how it will be measured afterwards, and what bounds the damage if it turns out to be wrong. Anything that cannot answer for itself on those terms is an observation, not an action.
The honesty requirements are enforced structurally rather than aspirationally. Claims are graded against outcomes rather than accepted on their reasoning. Changes carry receipts: what changed, why, the value that was persisted, and the metric that will later prove or disprove it. Every part of the system is held to the standard it applies to others.
Perception
Most of what a chart knows
is already in the price.
So the desk draws on sources that are not, and on shapes a chart cannot show you. Public disclosure, the state of the order book, and the changing geometry of the market as a whole. Every feed is best effort by design: a dead feed gates nothing, and the engine keeps flying.
- Public disclosure
- Ownership changes
- Corporate actions
- Funding
- Earnings calendar
- Volatility
- Market microstructure
- Cross asset shape
Abbreviated. The full perception stack, and how each source is weighted and validated, are documented for licensees.
The application
A native desktop app.
Not a dashboard in a browser tab.
A native Mac app, resident in the menu bar, watching an engine that runs as its own supervised process. Both run on your machine and talk to each other locally. The engine reaches out to your broker and its data sources. By default nothing accepts inbound connections; the optional signal webhook, if you turn it on, opens an authenticated inbound endpoint.
Set two things, then leave it
Broker keys and risk limits. The keys go into the macOS Keychain, are never written to a file by the software and are never sent to RLG. After that the engine runs its pre flight and starts. On the next launch it resumes in whatever mode it last ran, and after a reboot the login item brings it back without anyone pressing anything.
It shows its reasoning, not just its results
Every row in the ledger is inspectable: what the desk saw, what it decided, and, days later, what the market did about that decision. The dashboard carries live market context derived from the same data the engine trades on, so the screen can never disagree with the trading logic.
Watch its mind change
The learning feed records every adjustment the desk makes to itself, with the evidence that caused it. Every one of them evidenced, timestamped and readable afterwards. It is the difference between a system that claims to adapt and one that shows you the receipt.
Bring your own signals
Alerts from charting platforms, or anything that can post JSON, enter as ideas and never as orders. They clear every check anything built in has to clear, and they are treated as one more strategy with no head start, set aside if they do not pay. An alert carrying a quantity is refused outright, because sizing is never the sender’s job.
The suite that guards all of this includes a check which drives a live regulatory filing all the way to a protected order at the broker, with no human step anywhere in the path.
Licensing
Three ways in.
Sold to operators and firms who intend to run it, not to redistribute it. Every tier includes the charter, the owner’s guide and the reproducible research, because a system you cannot audit is one you should not run.
Before you ask for terms: the desk places trades in your account without asking you first. It will be wrong regularly. Losses are bounded by limits you set and by protective orders at your broker, not eliminated, and stop orders do not guarantee a price. Connect only capital you can afford to lose, and start in paper mode.
Consumer Grade
One operator, one machine, one account
- The complete application, no reduced build at any tier
- Starts in paper mode. It goes live when you decide, not before
- You set the risk envelope, the asset classes and which strategies are in scope
- Full pre flight, dashboards and the whole decision history
- Run the backtests and the year simulation yourself
- Owner’s guide and charter
Enterprise
Scoped to your seats, accounts and mandates
- Everything in Consumer Grade, across a team rather than one machine
- Strategy scope, risk envelope and asset classes set per seat
- Firm level limits the software operates inside and cannot widen
- Per seat decision record and attribution, reconciled to the broker of record
- Consolidated reporting across seats, entities and accounts
- Change control with a record of what changed, when and on what evidence
- Signed application, with updates for the license term
- Private support channel with the engineering team
- Onboarding covering how the risk settings work and how to test them in paper
Institutional
Perpetual, with build rights and governance
- Everything in Enterprise, without seat limits
- Full engine and application source, with build rights
- The research stack and study code, so findings can be validated independently
- Broker adapter development, with one port included
- Additional asset classes and venues scoped on request
- White label naming and icon
- Governance pack for diligence: charter, decision record format and reconciliation reports
- Continuity arrangement, in escrow or under license
- Architecture handover and a named engineering contact
Inquiries
Tell us what you
would run it on.
Account size, the asset classes you care about, and whether you intend to operate it or build on it. Terms follow from that.
The detail behind all of this is withheld for the same reason the desk refuses most of its own signals: the value is in the calibration, and calibration given away is calibration destroyed. Licensees receive it in full.
Answered by an engineer, not a queue
Questions
What agentic trading
actually means here.
- What is AI agentic trading?
- An agentic system does not wait to be told what to do at each step. It perceives the market, forms a judgment, acts on it, measures what happened, and adjusts its own behavior from the result. TradeAgentic.ai runs that whole loop continuously across crypto, equities and options, with no human approving individual trades.
- Does it really trade without approval?
- Yes. There is no approval queue and no discretionary override. What you control are the boundaries: which account it uses, how much it may risk, and when it must stop for the day. Inside those boundaries it decides on its own, and every decision it makes, including every refusal, is recorded and later graded against what the market did. It will take positions you would not have taken and it will be wrong regularly; its limits bound losses, they do not prevent them.
- How is this different from a trading bot?
- A bot executes rules you give it and cannot tell whether those rules are still working. This desk argues against its own candidates before committing, scores the value of its own refusals, withdraws capital from strategies its evidence says have stopped paying, and revises its own assumptions on that evidence. The rules are not the product. The judgment about the rules is.
- Which broker does it connect to?
- It connects to your own brokerage account over that broker's API, using your own credentials, which are held in the macOS Keychain, never written to a file by the software and never sent to RLG. Supported brokers and the market data feeds available through them are confirmed during licensing. Additional broker adapters are in migration, and a source license includes one broker adapter port.
- What happens if the software stops while a position is open?
- Protective stops are placed at the broker, so a stop already resting there stays working if the software is not running. A stop does not guarantee a price: in a gap, a halt or a thin market it can fill well beyond the stop, and a crypto stop-limit may not fill at all. On the next start it compares its records against the broker, adopts anything it does not recognize, and re-arms any missing stop before it is allowed to trade again.
- Can I see why it did something?
- Every decision it has made is inspectable: what the desk saw, what it decided, and, days later, what the market did about that decision. The learning feed separately records every adjustment the desk has made to itself, with the evidence that caused it.
- Do I need to write code or pick strategies?
- No code, and no picking individual trades. You set the envelope: your keys, the risk limits, which asset classes are in scope and which strategies are available to the desk. Inside that, they compete for one pool of capital and it decides between them on measured results rather than asking you which to run today.
- What does it run on?
- One Mac running macOS 13 or later on Apple silicon or Intel. It starts automatically at login and resumes trading after a reboot without anyone pressing anything. No market or account data is sent to RLG. The optional AI review, if you turn it on, sends trade candidates and summary account figures to a third-party AI provider under your own API key.