A strategy is a spec, not a black box.
The library stores StrategySpec JSON, not opaque binaries. You can read it, audit it, copy a clause out for a teammate.
Versioned. Diffable.
Auditable.
- Auto-versionedEvery save is a new version. Roll back to any prior version in one click.
- Tagged metadataSymbol, asset class, RR, win-rate target, expected DD, author, parent fork.
- Inline diffSee exactly which clause changed between v7 and v8. Approve or reject.
- Saved backtestsEach version keeps its own backtest report. Old reports never get overwritten.
- Full text searchSearch by symbol, pattern, RR, session, or any free-text in the spec.
{
"id": "strat_a82c3f...",
"name": "NAS100 EA · v9 PA",
"symbol": "NAS100",
"version": 9,
"tags": ["price-action","NY-open","FVG"],
"expected": {
"winRate": "62-67%",
"tradesPerMonth": "20-26",
"avgRR": "1.2",
"maxDrawdownEst": "<3%"
},
"parent": "strat_a82c3f@v8",
"author": "[email protected]",
"createdAt": "2026-04-22T14:11:00Z",
"lastBacktest": {
"from": "2025-11-05", "to": "2026-05-05",
"trades": 146, "winRate": 0.658,
"netReturnPct": 69.09, "maxDDPct": 1.80,
"calmar": 5.24, "profitFactor": 2.55
}
}Premium card grid. Cards that pull their weight.
Each card surfaces win-rate, PF, Calmar, max DD, trades / mo at a glance.
Open the spec editor, edit, save → new version. Old version still backtestable.
Fork a strategy with a new name. Tweak risk per-trade, sessions, kill-switches.
Side-by-side spec diff between any two versions. Promote or roll back.
By symbol, asset class, win-rate, last-backtest date, deployed status.
Tag with anything — "ICT", "NY-open", "ATR-trail", "client-X". Filter on tags.
Inline expand: full spec, KPI strip, monthly returns table, drawdown chart.
One toggle pushes a version to a connected broker account. Same spec, same engine, live.
Every backtest, every deployment, every kill-switch event — fully audited.
From draft to deployed in five steps.
Quant-AI emits a v1 spec from your prompt. It lands in the library as a draft.
- Tagged automatically
- Pre-flagged risks
Run a 6-month tick-accurate backtest. Inspect the report, reject or accept.
- Saved with the version
- Drawdown + monthly grid
Tweak risk, session, RR. Each save creates a new version. Diff against v1.
- Inline JSON editor
- No-merge-conflict workflow
Push the version to a demo account. Watch it run for a week alongside backtest.
- Same spec, real broker
- Reconcile gap report
Promote to live. Risk envelope re-armed. Engine enforces every cap.
- One toggle
- Audit log entry
Lend a strategy. Don't give the keys.
Share a link that exposes the spec + backtest report. Receiver can clone, not edit yours.
Team plan: shared library, per-strategy permissions, per-strategy comment threads.
Strategies are private until you explicitly share. No marketplace push without consent.
When a strategy is forked from yours, the fork tree is preserved. Clean lineage.
Risk shows up where strategies live.
Each card shows daily DD used, max DD used, consecutive losses — colour-coded.
If a strategy is paused by the engine, the card shows the reason and time.
Drill into a strategy to see its own equity curve, separate from portfolio.
Where the library starts mattering.
A strategy library that thinks like a senior trader.
Versions. Diffs. Backtests. Replay. Audit-ready by default.
Join the waitlist