APEX AI Trading is an event-driven signal engine. It continuously monitors macro markets and multi-source news, classifies what it sees into normalized events, evaluates them against a transparent declarative rulebook, and turns every actionable signal into a fully calculated, risk-managed trade proposal — all from free data sources, streamed live from the cloud to this dashboard with no page refresh. It also runs the Trade Book: the engine's own portfolio of S&P 500 stocks, bought on technical rules and weighed by the same news.
Daily OHLCV bars for ten macro instruments, plus roughly 130 headlines per cycle aggregated from three independent news feeds — all normalized into one shape and de-duplicated across sources.
Raw data becomes normalized events two ways: a macro instrument moving past its per-asset daily threshold, or a headline matching the keyword lexicon with word-boundary precision.
Active events are evaluated against a declarative 38-rule rulebook spanning eight market themes. Every rule is transparent and auditable — every signal traces back to the exact events that fired it.
Every actionable signal gets one year of daily candles analyzed for structure before any trade is proposed.
Signals are never a naked BUY/SELL — each ships as a concrete proposal anchored to the price structure found in Stage 4. The inverse logic applies to SELL setups.
Any S&P 500 stock or top US ETF can be analyzed from the search bar — even when no rule targets it directly.
The engine's own portfolio, open to every Pro member. It buys S&P 500 stocks that clear a set of technical rules, lets the news keep a stock out or move it up the queue, and sells on fixed exit rules. Long only, at most 15 stocks of $10,000 each. Every trade — and every trade it turned down — shows its reason.
At 4:30 pm ET every S&P 500 stock is scored on its monthly, weekly and daily charts. A stock becomes a candidate only if it passes all four core gates and at least three of four confirmations.
Each candidate gets a news score from the same live events that drive the signals above. Good news cannot buy a stock the scan didn't clear; bad news can stop one the scan did.
$150,000 of capital, split into 15 slots of $10,000. Free slots go to the best candidates, ranked by reward:risk plus a quarter of the news score.
A position closes at whichever rule it reaches first. The Trade Book shows the price each holding would close at, and which rule sets it.
When more stocks qualify than there are free slots, the weakest holding can give up its slot to a clearly better candidate.
While the market is open, every refresh of the news — about every 90 seconds — is checked against the holdings and the candidates.
Worked example — a stock kept out by the news (illustrative). The Trade Book lists every blocked stock like this under "Blocked by news", and names the rule behind every closed position — stop, loss cap, give-back, rotation or news.
| Step | Points | What happened |
|---|---|---|
| SCAN | — | A chip maker clears the technical scan: all four core gates and three confirmations, reward:risk 3.0. |
| SECTOR | −2.0 | Export restrictions — a headline about U.S. export controls. It doesn't name the company, but chip makers are sensitive to it. |
| MACRO | −1.5 | Treasury yields rising — the 10-year yield rose 3% on the day, a move large enough to count 1.5×. |
| RESULT | −3.5 | At or below −2, so the stock is kept out. The page shows each event with a link to its headline, and whether the block still holds. |
A background engine keeps this dashboard current on its own: it re-ingests every feed on a 90-second cycle, works out what is genuinely new, and pushes updates to every open browser over Server-Sent Events.
The server re-scans all market and news feeds every 90 seconds (configurable) and re-runs the full pipeline over the active event set — API reads stay near-instant because a live snapshot is always ready.
Every event has a stable identity — one per instrument per day for market moves, a headline fingerprint for news — so nothing is announced twice. Stale events expire automatically (market 8 h, news 6 h), so the board never goes stale.
Fresh snapshots stream over Server-Sent Events: the LIVE indicator shows connection state, newly arrived events and signals flash on arrival, and the stream reconnects automatically if the connection drops.
| Endpoint | What it does |
|---|---|
GET /api/stream | Live Server-Sent Events stream — a full snapshot on connect and after every ingest cycle. Powers this dashboard. |
GET /api/scan | Latest snapshot, returned instantly from the live engine. ?refresh=1 forces an immediate ingest cycle. |
GET /api/analyze?symbol=NVDA | On-demand analysis of any catalog symbol — bias, confidence, rationale, S/R levels, trade proposal. |
GET /api/tradebook | The Trade Book for Pro members — holdings, closed positions, candidates, blocked stocks and the news behind each decision. |
GET /api/catalog | The full 611-symbol search catalog (symbol, name, type, sector). |
GET /api/health | Liveness + real-time status: last cycle time, active events, per-feed headline counts, connected clients. |
The engine refreshes every 90 seconds and price events come from daily-bar changes, not intraday ticks. Signals are swing- and position-trade oriented — not built for day-trading.
All 38 rules are human-authored and auditable — every signal traces back to its exact events and rule. Confidence scores are formula-based, not yet backtested; there is no ML model making opaque calls. The Trade Book's technical rules were backtested on the S&P 500; the news layer on top of them has not been backtested yet.
APEX does not connect to a brokerage or place orders for you: it proposes, you decide. The Trade Book tracks the engine's own portfolio, so you can see how its rules trade. Automated trading on the signals is next on the roadmap.