promotion-gate

name: promotion-gate description: Evaluate promotion readiness from flywheel outputs using statistical and economic gates, then produce a concise go/no-go decision note.

Promotion Gate

Use this skill when deciding whether a strategy candidate moves from research iteration to deeper validation.

What this skill does

1. Reads flywheel output (summary, promotion, costs).
2. Verifies required gates:
   • realized Sharpe
   • net Sharpe
   • alignment guard
   • event ordering guard
3. Produces a decision summary with blocking reasons.

Core APIs

• openquant.research.run_flywheel_iteration(...)
• openquant.pipeline.summarize_pipeline(...)

Decision contract

Required outputs:

• promote_candidate (bool)
• failed gate list
• key metrics (realized_sharpe, net_sharpe, turnover, estimated_cost)
• short rationale

Artifact source

• decision.md and metrics.parquet in each experiment run folder.

Quality checks

uv run --python .venv/bin/python pytest python/tests/test_experiment_scaffold.py -q