These are small, deterministic artificial qfq OHLCV histories used to verify ordinary and wide-limit amplitude parameters without importing the legacy project at test time. The rows are calendar-spaced only to keep the fixture readable; the formula treats them as ascending trading observations.

multi_signal.json documents the independent-mask orchestration case. The unit test forces all seven masks on one prepared target row, which is intentional: the legacy result set did not contain a trustworthy historical same-day multi-hit sample, while the new contract requires retaining all hits.