Settings

Defaults that apply to every backtest you fire — sizing conventions, the Indian cost model, AI prompt preferences, deployment guardrails.

Coming Soon

Default sizing rules used when a strategy doesn't specify its own. The sweep engine will also vary these as a swept parameter.

Live once the engine reads from a persistent settings store (Phase B2).

Coming Soon

Brokerage, STT, exchange charge, SEBI, stamp duty, GST, slippage assumption, bid-ask spread treatment. All line-itemed in results.

Default cost model ships as defaults you can override; Phase B1.

Coming Soon

Walk-forward split (default 70/30), bootstrap fold count (default k=5), survivor-bias controls.

Per-strategy override available; defaults configurable here in Phase B2.

Coming Soon

Model selection (Claude Opus 4.6 / Sonnet 4.6), prompt templates, confidence thresholds, max tokens per call.

Wired with Phase B3 — needs the Anthropic API key to be set as an env var on this Vercel project.

Coming Soon

Auto-pause gates: block live deployment when current regime fails the strategy's tested regimes (e.g. VIX > 22).

Lands with the Phase B4 live linkage to UNIVERSE.

Coming Soon

Optional layer: post-tax P&L using Indian options STCG (20%), surcharge, cess, loss carryforward.

Toggle-on in Phase B2 results UI.