optimize-op-preflight
Check whether a CV-CUDA operator is READY to optimize (correctness + bench coverage + captured baseline + profiling) per .agents/guidance/OPTIMIZATION_GUIDELINES.md. Use before starting an optimization campaign to confirm the readiness gate is clean.
Datos de origen
- Repositorio
- CVCUDA/CV-CUDA
- Última actividad en el origen
- 11 de agosto de 2026 a las 21:59
- Idioma detectado de SKILL.md
- inglés
- Estrellas
- 2727
- Forks
- 264
Opciones de instalación
De forma predeterminada está seleccionado el prompt que primero revisa el origen. Puedes cambiar a un comando directo o descargar una copia local.
Revisa los archivos de origen
Lee SKILL.md y los archivos complementarios que muestra SkillsMP antes de decidir si quieres instalarlo.
Mostrando SKILL.md
SKILL.md
Instrucciones de origen · Vista previa de solo lectura- name
- optimize-op-preflight
- description
- Check whether a CV-CUDA operator is READY to optimize (correctness + bench coverage + captured baseline + profiling) per .agents/guidance/OPTIMIZATION_GUIDELINES.md. Use before starting an optimization campaign to confirm the readiness gate is clean.
[//]: # "SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved."
[//]: # "SPDX-License-Identifier: Apache-2.0"
# Optimize Op — preflight
Thin entry point. Run `python3 tools/optimize_op.py <Op> --phase preflight` and interpret per
the tool's readiness (PRE-*) findings against `.agents/guidance/OPTIMIZATION_GUIDELINES.md`: correctness/regression coverage and
benchmark coverage (reused from `review_op.py` test+bench), a captured baseline, and profiling
tools. Any `GAP` means the operator is not ready — add the missing coverage **first**, in a
separate non-`perf:` commit, before starting the optimization. Resolve `MANUAL` items at the
cited locations. Findings-first.
Ver en GitHub