Retrospective / Time-Machine analysis
Rewind a name to a moment, see it as it was THEN, attribute the move to
catalysts, and test whether a tradeable edge actually existed. The whole value
is honesty: no lookahead (never use a price the moment didn't yet know) and
no stale data (never report yesterday's close as "now").
alice analysis snapshot provides dated market summaries; the alice-analysis
skill describes quantitative queries. Choose historical news sources with
coverage of the period being studied.
The freshness gate — DO THIS FIRST, every time
Every snapshot/quant result carries a freshness contract:
asOf, isLatestActual, staleTradingDays, and a freshnessWarning when the
data does not reach the anchor. Before you state any "current" number, check
it.
isLatestActual: false → the close you're holding is STALE. Do not call it
the current price. Say "as of , N trading days behind" and, for
anything live, pull a realtime broker source.- A free vendor (yfinance) lags a day or two and a free broker tier (alpaca
SIP) may not have today yet. An overnight catalyst can land in exactly that
blind spot — the classic trap is reporting a flat green close while the real
reaction already happened after the bar you can see.
snapshot --query <SYM> auto-picks the freshest source (realtime broker >
delayed vendor). Prefer it over hand-fetching from a delayed vendor.
Procedure
Snapshot the anchor (no lookahead). Reconstruct the moment with
asOf — bars never run past it.
alice analysis snapshot --query XLE --asOf 2026-04-03 # summary
alice analysis snapshot --query XLE --asOf 2026-04-03 --bars 30 # + dated path
Read the latest print (close, vs-prevClose, day high/low, amplitude —
a sleepy vs-prevClose number hides an intraday plunge-and-recover) and
levels (sma20/50, rsi14, distance from the period high — the "how far off
the top" feel). The snapshot is summary-only by default (the dated path
can be large); add --bars N when you actually need the per-day series.
windowBars tells you how many are available.
Align the catalysts to the price. Build a dated event timeline from
relevant sources and compare it with the bars. Distinguish evidence of a
catalyst from a coincident headline, and note material gaps in coverage.
Test the hypothesis in Workspace research code. If the question needs
a hypothetical trade, save the dated inputs and write an inspectable script
with the native Coding Agent. Pin the source and evaluation window. State
signal timing, executable entry/exit prices, position sizing, fees and
slippage explicitly. Do not use a closing-price signal to assume an
executable fill at that same close. Exclude pre-entry price extremes from
holding-period MFE/MAE. Save the script, inputs and results together so the
assumptions can be changed and the calculation reproduced.
Map the index to dates when you need the path in a quant script. Most
reads are covered by snapshot; when you must compute over the series and want
the date axis, add --dates to alice analysis quant (it returns
dates[barId] for one interval, or dates["barId@interval"] when the same
barId appears at multiple intervals, so you can map each value to its day).
Write it down honestly
A retro is only worth as much as its weakest assumption. State, every time:
- the asOf and that the analysis used no later data,
- the source + freshness of every "current" number,
- what you couldn't see (feeds didn't cover it / cookie-gated / SIP didn't
have the latest day) — name the gap rather than paper over it.
The failure mode this skill exists to prevent: a confident call built on a
stale or future-leaking price. When in doubt, distrust the data before the
market.
1---2name: retrospective3description: Subjective retrospective / time-machine analysis: rewind a name to a past point, reconstruct what it looked like THEN (no future knowledge), align the news catalysts to the price path, and pressure-test "if I'd entered there, would it have worked?". Use when the question is about a past moment or a hypothetical entry: "rewind XLE to early April", "what did NVDA look like before earnings", "if we'd bought energy after the Iran headline, easy money?", "replay the SMH spike — policy or earnings?", "was there an entry signal at the time", "would an 8% trailing stop have saved me", "event study on the Hormuz escalation". It strings together the as-of snapshot, the date-windowed news, and Workspace research code into one honest replay — and it is ruthless about data freshness, because a retro built on a stale or future-leaking price is worse than no retro.4---56# Retrospective / Time-Machine analysis78Rewind a name to a moment, see it as it was THEN, attribute the move to9catalysts, and test whether a tradeable edge actually existed. The whole value10is honesty: **no lookahead** (never use a price the moment didn't yet know) and11**no stale data** (never report yesterday's close as "now").1213`alice analysis snapshot` provides dated market summaries; the `alice-analysis`14skill describes quantitative queries. Choose historical news sources with15coverage of the period being studied.1617## The freshness gate — DO THIS FIRST, every time1819Every snapshot/quant result carries a freshness contract:20`asOf`, `isLatestActual`, `staleTradingDays`, and a `freshnessWarning` when the21data does not reach the anchor. **Before you state any "current" number, check22it.**2324- `isLatestActual: false` → the close you're holding is STALE. Do not call it25 the current price. Say "as of <date>, N trading days behind" and, for26 anything live, pull a realtime broker source.27- A free vendor (yfinance) lags a day or two and a free broker tier (alpaca28 SIP) may not have today yet. An overnight catalyst can land in exactly that29 blind spot — the classic trap is reporting a flat green close while the real30 reaction already happened after the bar you can see.31- `snapshot --query <SYM>` auto-picks the freshest source (realtime broker >32 delayed vendor). Prefer it over hand-fetching from a delayed vendor.3334## Procedure35361. **Snapshot the anchor (no lookahead).** Reconstruct the moment with37 `asOf` — bars never run past it.38 ```bash39 alice analysis snapshot --query XLE --asOf 2026-04-03 # summary40 alice analysis snapshot --query XLE --asOf 2026-04-03 --bars 30 # + dated path41 ```42 Read the `latest` print (close, vs-prevClose, day high/low, **amplitude** —43 a sleepy vs-prevClose number hides an intraday plunge-and-recover) and44 `levels` (sma20/50, rsi14, distance from the period high — the "how far off45 the top" feel). The snapshot is **summary-only by default** (the dated path46 can be large); add `--bars N` when you actually need the per-day series.47 `windowBars` tells you how many are available.48492. **Align the catalysts to the price.** Build a dated event timeline from50 relevant sources and compare it with the bars. Distinguish evidence of a51 catalyst from a coincident headline, and note material gaps in coverage.52533. **Test the hypothesis in Workspace research code.** If the question needs54 a hypothetical trade, save the dated inputs and write an inspectable script55 with the native Coding Agent. Pin the source and evaluation window. State56 signal timing, executable entry/exit prices, position sizing, fees and57 slippage explicitly. Do not use a closing-price signal to assume an58 executable fill at that same close. Exclude pre-entry price extremes from59 holding-period MFE/MAE. Save the script, inputs and results together so the60 assumptions can be changed and the calculation reproduced.61624. **Map the index to dates when you need the path in a quant script.** Most63 reads are covered by snapshot; when you must compute over the series and want64 the date axis, add `--dates` to `alice analysis quant` (it returns65 `dates[barId]` for one interval, or `dates["barId@interval"]` when the same66 barId appears at multiple intervals, so you can map each value to its day).6768## Write it down honestly6970A retro is only worth as much as its weakest assumption. State, every time:71- the **asOf** and that the analysis used no later data,72- the **source + freshness** of every "current" number,73- what you **couldn't see** (feeds didn't cover it / cookie-gated / SIP didn't74 have the latest day) — name the gap rather than paper over it.7576The failure mode this skill exists to prevent: a confident call built on a77stale or future-leaking price. When in doubt, distrust the data before the78market.
TraderAlice