📢 Advertisement
T

Tapehouse

Live technical scanner · Binance Futures
Hunters
5 AI agents on deskstandby
P
Pulse
Tape
F
Forge
Flow
A
Atlas
Book
Æ
Aegis
Risk
N
Nyx
Critic
Five AI agents work the tape, flow, book, and risk on your chosen universe — then Nyx weighs all four and prints one verdict: candidate, watch, or no trade. Default is no trade.
Agents hunt BTCUSDT only — one print
~6 days of context
Stop = 1.5× ATR(14)
How the scoring works
This is a plain, rules-based scanner — not an opaque model, and not a prediction of what will happen. For every coin in the chosen universe it computes three scores from real Binance Futures candle data:

Momentum — how far RSI(14) sits from neutral. Only coins with RSI ≤ 40 (long candidates) or RSI ≥ 60 (short candidates) are considered at all.
Volume — the latest candle's volume vs. its 20-period average. Confirms the move isn't happening on thin activity.
Structure — how close price sits to its own recent swing high/low, in the direction of the candidate trade.

These three combine into one confidence score (0–100). Entry is the latest price; the stop is 1.5× ATR(14) away; the target is set by your chosen risk:reward multiple.

For the top 3 picks only, a fuller TA breakdown is computed too — market structure, ADX/DI, MACD, EMA stack, Supertrend, Parabolic SAR, Aroon, CMF, Ichimoku, ATR percentile, prior-session range, and relative strength vs BTC — every value calculated live from the same candle data, nothing simulated. One thing you won't see: order-flow/CVD. That needs trade-by-trade tape data this scanner doesn't pull, so rather than fake that number it's just left out. A CONCERNS section also auto-flags real conflicts — low volume, an extended price, weak trend strength, and so on — so the confidence score isn't the only thing you're looking at.

None of this is a guarantee. Treat every result with the same scrutiny you'd give a backtest.
Not financial advice. Technical/statistical output only, based on live public market data.
Past scans do not predict future ones — treat every result with the same scrutiny you'd give a backtest.
📢 Advertisement