Studies: named collections of runs that differ only in their parameters (#157)
Config folder: --config/SBT2_CONFIG, venue profiles out of the package (#151)
Spec: venue models are named by kind, not import path (#155)
CLI, Results, Spec: sbt2 report, the run summary table, stored_run and from_document Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Strategy, Spec, Results, Run, Glossary: the risk table and the drawdown guard as built (#98) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
M4 scope: nautilus sizing and RiskEngine, one drawdown guard in the base's order gate Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Spec, Glossary, Decisions: bars name a BarSource
Spec: consumers ask the resolved spec instead of reading its dicts
Spec, Library, Glossary, Decisions: FractionSplit and DateSplit replace Split; SpecError (#69)
Drop provenance, code identity, result identity and the data fingerprint
Spec: expansion of lists and parts; CLI: sbt2 run runs in sequence until batch
Spec, Results: splits as built in #55 (required split, rounding, split JSON in the summary)
Candles: 1-minute candles stored, bars built from trades or candles per run (#49)
Liquidation off without quotes, a spec key for it, and the faster run numbers
Run pre-flight: source symbol, known gaps in the summary, new errors
Splits: train / validation / test, with only test locked The optimiser tunes on train and ranks on validation; validation runs as often as you like; the lock becomes the test lock. Walk-forward is deferred until M10. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Library: settle the open questions and bring every page in line Two parts (in-sample / out-of-sample) replace train / validation / test; Holdout by fraction or date, no embargo; only the final out-of-sample part is locked, per lineage by date overlap; every run stored; code identity replaces the git SHA; one root replaces data/ and config/. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Library: target design for sbt2 as a published package, open questions Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Spec: capital as nautilus money, seed goes to the fill model, unknown spec key error
Spec: name instruments by full nautilus InstrumentId
Spec: pass nautilus arguments through, no wrapper types; importable venue models
docs: thin nautilus strategy base; asset profiles keyed by nautilus Strategies are nautilus Strategy subclasses on a thin sbt2 base that submit their own orders; the plain-Python pipeline (intents, reconciler, adapter) is gone (PR #18). Asset profiles are keyed by nautilus's AssetClass + InstrumentClass and venue profiles name them with nautilus enum names (issue #19). Results notes that annualization must be passed as nautilus's period (default 252) and that benchmark-relative stats come from nautilus. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Funding streaming verified on the rc6 develop wheel; funding schema from nautilus; explicit fee model required Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Move to nautilus v2: native funding, portfolio equity snapshots, run-time bars, public catalog writers, nautilus tearsheets Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Move to nautilus: run config, venue profiles, mark-price valuation, brackets, RiskEngine, bar aggregators, instrument provider, reports Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Drop lockbox and access log, keep strategies in this repo, remove host OS from deployment Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Simplify architecture to six modules: data, spec, strategy, run, results, cli Merge download/ingest/data into data; spec/runner/engine into spec + run; results/metrics/analysis into results. Drop core and venue as modules. Access log moves to spec. ResultStore and output sink interfaces kept. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Refine architecture: core and metrics modules, output sink, split safety, strategy pipeline - core leaf module for shared types; pure metrics module shared by runner and analysis - full dependency graph including plug-ins; drop research module until the optimizer lands - engine writes through an output sink; runner commits staging on success - venue profile names venue + asset class; resolve maps symbols to InstrumentIds - splits frozen by access log; new test windows may not overlap seen data - notebook loaders lock every split's test window - strategy pipeline: strategy -> sizing -> risk guards -> pure reconciler -> thin adapter - fresh process per run; result identity; valuation price; per-source known gaps - benchmark moved out of the run spec; perp benchmark uses the perp price without funding Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Apply review decisions: runner/results/data modules, resolved-spec hashing, lockbox, splits, intents
Make data sources pluggable; move Bybit specifics to their own page