One person. Coding agents. Two servers. A live trading system with my own money on-chain.

I built SKYNET to bring trading intelligence and execution into one platform.

It tracks Telegram signals and on-chain wallet activity, with rankings that make their historical performance easier to compare.

Wallet profiles reveal trading results and behavior, while market intelligence adds context about changing conditions.

The strategy engine lets you replay historical signals with different exit rules and position sizes.

The trading terminal brings together Phantom wallet swaps, automated copy trading through my custom Rust engine, and live position monitoring.

Detailed transaction timelines show the journey from the source signal through execution to on-chain confirmation.

SKYNET connects the research, the trade, and the evidence behind it.

Skynet brings research, strategy design, position sizing, and trade execution into one connected workspace. Let’s walk through how it works.

It starts with signals from Telegram and tracked wallets. The research layer puts those signals in context. Strategies define the rules, sizing defines the exposure, and the terminal brings execution and monitoring together.

Start with the source rankings. Open a profile, and you can look beyond the headline numbers: how active the source is, the market caps it enters, and how different models perform. Time, chain, and market filters help narrow the comparison.

Here’s Trader A. We’re using a hundred-dollar paper entry across the full history. The leading models are below zero, but the regime labels reveal a pattern: positive results in bull markets, and losses in sideways and bear markets. Let’s follow Model Six through the filters.

Select Bull, and the ranking changes. Model Six moves to the top, with three thousand two hundred and fifty dollars in historical replay profit. Models Four and Two are positive as well. We haven’t changed the entry size. We’re looking at the calls that fell within bull-market conditions.

Now switch to Bear. Model Six shows a replay loss of one thousand five hundred and fifty dollars, and the leading models are below zero. That’s why market context matters. The overall result combines periods where the same strategy behaved very differently.

Before changing the sizing, return to all regimes. Model Six shows a three thousand five hundred dollar loss across seven hundred and seventy-two cases. Its exit rules are a fifty-percent stop loss and a five-times take profit. We’ll keep those rules fixed for the comparison.

Open Custom Sizing and select Test Four. This preset switches off entries below three hundred thousand dollars in market cap. From there, it allocates one thousand dollars to the next tier, two thousand to the one-to-ten-million tier, and three thousand above ten million. The two-percent limit is set to trim mode.

Apply the preset, and the strategy cards update. The date range, chain, and exit rules stay the same. In this recorded snapshot, Model Six moves from a three thousand five hundred dollar loss to fifty-five thousand dollars in historical replay profit. Models Four and Two are also positive. The monthly curves, regime breakdown, and ranking all reflect the new settings. Model Six now includes three hundred and fifteen cases.

Switch back to Paper, and Model Six returns to its three thousand five hundred dollar loss. Apply Test Four again, and the positive replay returns. Two things have changed: which entries are included, and how much is allocated to them. That changes both the results and the exposure. Maximum drawdown in the custom view is twenty-eight thousand dollars. These are historical simulations with trading costs excluded. The comparison helps you understand the effect of your rules; it doesn’t establish how they’ll perform in future trading.

Strategy Center brings these trader-and-model combinations into one view. Compare exit rules, replay results, market regimes, and consistency across the cards. When something deserves a closer look, open its statistics or continue into simulation.

Market Intelligence adds the wider context: daily regimes, intraday alerts, and reversal indicators. Coverage and freshness are shown alongside the signals, so you can see how much information is available and how recently it was updated.

Now let’s move into the trading terminal. The chart, swap controls, strategies, and trade history share one workspace. You can move from researching a setup to managing its execution, then come back to inspect the result.

The wallet balances are loaded here. You can see the available Solana balance and the tokens held in the wallet. Open the selector to inspect quantities and displayed values. Choose a holding, and its available balance appears in the swap panel, ready for the amount you want to enter.

For a manual Phantom swap, the system prepares an unsigned transaction. You review and approve it in your wallet. Phantom signs it, and the selected engine handles submission.

The Ultra option uses Jupiter’s order and execution flow. With Phantom connected, the prepared transaction returns to the wallet for signing. The selected tokens, amount, and active engine remain visible together in the swap panel.

The custom route is labelled Legacy in the terminal. Jupiter supplies the quote and swap instructions. The Rust engine builds the transaction, Phantom signs it, and the custom submission path sends it through Helius Sender.

Automation uses a different signing flow. The bot checks incoming signals against the configured strategy, and its execution wallet handles signing. The Rust service connects transaction preparation, submission, confirmation, and the monitoring of the resulting position.

Inside the custom engine, the instructions become a transaction with the configured compute budget, priority fees, and submission settings. Once the trade is confirmed, that execution connects back to its position record.

The published measurements report a median engine time of sixty-six milliseconds, and around one point three seconds from signal to confirmation. Those figures describe different stages. Next, let’s inspect the timings of an individual recorded trade.

Open a recent copy trade from Positions. This entry was triggered by the automation. The record brings its outcome, costs, and entry context together. From here, we can open the timing breakdown and follow what happened after the signal arrived.

The displayed end-to-end time for this entry is two point four two three seconds. Within that, quoting, swap instructions, transaction building, and submission took eighty-one milliseconds combined. Network confirmation is shown separately at four hundred and seventeen milliseconds. You can see the engine’s speed in context, alongside the rest of the recorded workflow.

This position was later closed manually. The exit record shows seventy-nine milliseconds of engine work, with confirmation measured separately at four hundred and sixteen milliseconds. You can also inspect the route, slippage, fees, and transaction links. Open the log trace to follow the recorded events in more detail. The timing numbers connect to evidence you can check.

That’s what makes this view useful. You can inspect the stages behind an execution, see where time was spent, and follow the transaction evidence. Manual swaps and copy trades each keep their own breakdown.

Back in the terminal, the strategy rail lists each configured source and model, its status, and recent performance. Search or sort to find a strategy, then open its overview, positions, history, or configuration. Those views stay connected within the same workspace.

The overview brings net profit and loss, win rate, trade count, and exit outcomes together. The realized performance curve shows how results developed over time. The fee breakdown helps explain how gross trading results translate into net performance.

Open Configuration to see the rules behind those results: base investment, daily trade limits, take profit, stop loss, the selected wallet, and buy and sell slippage. The model and status stay visible, keeping the strategy’s setup close to the activity it produced.

Positions connects each open lot to its source and model. Entry details, valuation fields, exit rules, and transaction links sit together. You can move directly from a strategy to the positions it created and inspect their current state.

Trade History gives you the individual records behind the summary: entries, exits, execution status, outcomes, and transaction references. Together with the fee breakdown, it helps explain the result over time. These execution records are kept distinct from the historical replays used in research.

The wider system also includes vault and wallet views. Portfolio monitoring, manual wallet approval, and automated execution each have their own role, with different signing responsibilities.

Monitoring continues throughout the workflow. Token tracking keeps entry context, current observations, historical highs, and data freshness visible. Wallet research adds another view of activity, while the strategy and execution records show how the system responded to the information it received.

That’s Skynet: from the first signal to the strategy, the trade, and the evidence behind the result.