cvcuda
- 12 skills
- 0 followers
- 17 hours ago last updated
- ▌ Make Op · cvcudaAdd a new CV-CUDA operator end-to-end per .agents/guidance/MAKE_OP_GUIDELINES.md, with a deterministically-enforced definition-of-done. Use when asked to create/add a new operator, scaffold one, or verify that a new operator is complete (approved spec, wired scaffold, bit-exact regression tests across the declared support matrix, required layout parity, complement negatives, docs + relnote, benched dtypes, tests that run and pass, optimization-ready).
- ▌ Review Op · cvcudaReview a CV-CUDA operator end-to-end (support / test / bench / docs coverage). Use when the user asks to review an operator, audit its input-type/layout/dtype support, test coverage, benchmark coverage, or docs/API, or to find & fix per-operator coverage gaps. Produces a deterministic findings-first report.
- ▌ Optimize Op · cvcudaDrive a single-operator optimization campaign per .agents/guidance/OPTIMIZATION_GUIDELINES.md, with a deterministically enforced definition-of-done and versioned MR summary. Use when asked to optimize an operator, run or finish a performance campaign, generate or refresh its MR performance summary, or verify that a perf MR satisfies correctness, benchmark, baseline, lead-exhaustion, memory, and API/ABI gates.
- ▌ Refactor Op · cvcudaFind and safely apply per-operator refactoring / redundancy-reduction opportunities in a CV-CUDA operator (near-duplicate Tensor/VarShape kernels, reinvented shared utilities, dead code). Use when asked to reduce code duplication, de-duplicate or unify an operator's kernels/bindings, remove dead code, or verify a refactor changed nothing observable. Produces a deterministic findings-first report; the apply path is gated by a strict bit-exact / feature-set / test-coverage parity check.
- ▌ Make Op Verify · cvcudaVerify a new CV-CUDA operator against the deterministic final regression checklist (the /make-op done-gate). Use when asked whether a new operator is complete/done, or to gate it before merge - gold reference, bit-exact coverage across the declared support matrix, required layout parity, complement negatives, docs + relnote, benched dtypes, tests that run and pass, and optimization-readiness.
- ▌ Make Op Scaffold · cvcudaScaffold a new CV-CUDA operator — a complete, wired, building skeleton — and delegate the implementation to a human or another AI. Use when asked to set up / stub out a new operator without implementing it, or to bootstrap one for someone else to finish. Supports --bare (skip the spec when the definition itself is delegated).
- ▌ Review Op Support · cvcudaReview a CV-CUDA operator's input-type, layout, dtype, and channel support matrix. Use when asked which inputs an operator supports, whether required layouts are complete, or whether support is enforced consistently across C, C++, and Python APIs.
- ▌ Optimize Op Verify · cvcudaVerify a CV-CUDA optimization campaign's deterministic definition-of-done and concise versioned MR summary per .agents/guidance/OPTIMIZATION_GUIDELINES.md. Use to gate whether a perf campaign or performance MR is ready, including reference-SKU statistics, hard checklist evidence, baseline validation, lead exhaustion, memory checks, and API/ABI stability.
- ▌ Review Op Docs API · cvcudaReview a CV-CUDA operator's DOCS & API artifacts — operator_list row, Python autofunction (fn + _into), Limitations-table-vs-code consistency, docstrings, and SPDX headers. Use when asked whether an operator's docs/API surface is complete and consistent.
- ▌ Optimize Op Preflight · cvcudaCheck 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.
- ▌ Review Op Test Coverage · cvcudaReview a CV-CUDA operator's test coverage, including C++ correctness, required cross-layout parity, correctness rigor, and the Python API surface. Use when asked whether an operator is adequately tested or to find and fill test gaps.
- ▌ Review Op Bench Coverage · cvcudaReview a CV-CUDA operator's BENCHMARK coverage — drivers, layout axis, baselines, the basic-tier floor, row counts, and coverage statistics. Use when asked whether an operator's benchmarks/baselines are complete or to find/fill bench gaps.