출력
기대 결과
- public call
- representative return shape
- verification result
엔진 역할
quant engine application skill for 전략: 사용자 정의 boolean rule 백테스트 (Rule + sizing + stop).
공개 호출 방식
import dartlab
result = dartlab.quant("strategy", "005930") 호출 동작
Runs a 전략 DSL quantitative calculation. Confirm period, benchmark, target requirements, and for strategy/backtest axes separate rule and cost assumptions.
대표 반환 형태
Returns a DataFrame or dict. Core fields are target, period, priceDate/latestAsOf, benchmark, metric, value, score/signal/rank, assumptions, and flags.
기본 실행 순서
- target, period, and source data are fixed first.
- Run the public call exactly as documented.
- Check latestAsOf/date, missing values, flags, and assumptions.
- Bind numeric claims to tableRef/valueRef/dateRef/executionRef.
- Hand off multi-axis narrative composition to story or the parent report skill.
기본 검증
This skill is a public execution document. If this axis call, representative return keys, error behavior, or runtime limits change, update this file in the same change.
런타임
실행 환경별 호환성
| 환경 | 상태 | 비고 / 제한 |
|---|---|---|
| Local Python | supported | — |
| Server | supported | — |
| MCP | supported | — |
| Web AI | supported | — |
| Pyodide | limited | — |
실패 회피
흔한 실패 · 절대 금지
절대 금지
- Do not guarantee performance.
- Do not cite returns without period, benchmark, and assumptions.
- Do not present a quantitative signal as causal financial analysis.