Anomaly Detection
- Price Move — Bar-over-bar return far outside the symbol's own recent return distribution.
- Volume Surge — Participation far above the recent volume baseline.
- Volatility Spike — 5-bar realized volatility breaking out of its longer-run regime.
- Options Flow — Put/call flow, implied vol level, term structure or open interest repricing sharply.
- Correlation Break — Decoupling from the benchmark it normally tracks.
- Liquidity Drop — Traded liquidity thinning well below its own norm.
- Overnight Gap — The move that happened while the book was closed. Price Move scores close-to-close and cannot see it: a symbol that gaps 8% and then grinds back to flat has an ordinary daily return and an extraordinary gap.
- Range Expansion — How far price travelled inside the bar, regardless of where it closed. Catches the violent two-way session that closes unchanged — invisible to every close-based detector.
- Momentum Thrust — A sustained multi-bar drift that no single bar is large enough to trip. Ten consecutive 1.2% days are a 12% move and seventeen ordinary returns.
- Envelope Break — Price standing outside its own statistical range — the shaded band on the chart above. A level, where Price Move and Momentum Thrust are both changes.
- Relative Strength — Return net of the benchmark's. Separates "the whole market fell" from "this fell". Correlation Break asks whether it still moves WITH the benchmark; this asks whether it is beating it.
- Beta Shift — How much market move the symbol now takes per unit of benchmark move. Correlation can stay at 0.9 while beta doubles — same direction, twice the size.
- Flow Imbalance — Volume signed by the direction of its own bar, summed over 5 bars and scaled by normal participation. Volume Surge sees size; this sees which way the size leaned.
- Price Impact — How much price moves per dollar traded. Rises when the tape gets thin OR when the same size starts moving price further — a different question from Liquidity Drop, which only sees the dollars.
- Persistence Streak — Runs of same-direction bars against the symbol's own history of runs. Seven straight small down bars are a distributional anomaly that no single-bar z-score can express.
- Volatility Instability — Volatility of volatility. Volatility Spike fires when vol is high; this fires when vol has stopped being stable — the regime is changing rather than merely elevated.
- Model-based — Multivariate outlier: scikit-learn IsolationForest and LocalOutlierFactor agreeing, over the joint return / gap / range / volume / volatility / correlation vector.
Ask the market a question. Get a calculated answer.
The AI is not a chatbot bolted onto a document store. It calls the same analytics engine that powers every screen on this platform — so what comes back is a number it computed from raw history, with the command that produced it.
86,000+ instruments
Global equities, ETFs, funds, options, FX, commodities, crypto, economics, filings, transcripts and news — one normalised symbol universe with adjusted history.
A real analytics engine
Screening, backtesting, technicals, options analytics, correlations, seasonality and factor models — computed on demand from raw prices, never a stale cache.
It shows its working
Answers arrive with the charts, tables and tool calls behind them, so you can check the number instead of trusting a paraphrase.
Your own documents
Upload filings, decks and research. Ask across them and the answer cites the page it came from.
Agents and workflows
Multi-step research that runs the platform's tools for you — screen, pull the history, compute, compare, then write it up.
MCP, CLI and API
The same command catalogue from Claude, your own agent, a shell or your pipeline. The answer on screen is the answer your job gets at 4am.
You ask
“How does NVDA usually trade through earnings?”
It calls
→ ka.options_expected_move(NVDA)
It answers
NVDA has averaged a 9.2% absolute move on the day after earnings and closed higher 67% of the time. Two in three reactions land between −4.2% and +16.3% — the distribution is skewed right, not symmetric.
Every figure computed live from our own history — not scraped, not summarised.
Or start with