メニュー
Analyze a HuggingFace model and attempt OpenVINO conversion — probe properties, run strategy matrix, classify failures, and produce a structured routing report.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Write unit tests for OpenVINO ov::Model graph transformations (passes). Use when the user asks to write, add, or refactor tests for a transformation pass (MatcherPass, ModelPass), or to modernize legacy transformation tests.
Troubleshooting all sorts of failures, crashes, exceptions and errors using debug capabilities. Analyze accuracy, performance, model compilation, or memory issues. Dump tensors and intermediate blobs. Serialize and visualize IRs, execution graphs. Enable verbose, logging. Profile execution. Compare layer outputs. Inspect, trace or dump transformations. Identify executed operations, nodes, primitives, kernels.
Upgrade the PyTorch version used by OpenVINO tests (torch / torchvision / torchaudio) and resolve fallout — missing operator translators, new functionalized `*_copy` aten ops, decomposition changes, FX-only tests failing in TorchScript mode, and accuracy regressions caused by stricter typing. Use when the user asks to "bump torch", "update pytorch to X.Y", "upgrade torch tests", or when pytorch_tests / model_hub pytorch tests fail after a torch version change. Do not use for: enabling a single new PyTorch operator unrelated to a version bump, GenAI / Optimum upgrades, or plugin-level numerical bugs unrelated to the frontend.
Adds a core operator to the OpenVINO toolkit. Use when asked to implement a new operation into OpenVINO.
Add a new operation to the OpenVINO CPU plugin — node registration, JIT/oneDNN executors (AVX2/AVX-512/AMX), and functional tests.
Adds a new operation to OpenVINO Frontend pipelines with translator updates, registration, and tests.
| name | analyze-and-convert |
| description | Analyze a HuggingFace model and attempt OpenVINO conversion — probe properties, run strategy matrix, classify failures, and produce a structured routing report. |
When you have a model ID (or local path) and need to determine whether it converts to OpenVINO IR, identify why it fails, and emit routing signals for the next agent in the pipeline.
Execute in strict order — each step produces files consumed by the next.
| Step | File | Purpose |
|---|---|---|
| 1 | probe-model.md | Gather model profile (architecture, task, op types) without downloading weights |
| 2 | try-conversion.md | Run conversion strategy matrix (optimum-cli / ovc / convert_model), capture all outputs |
| 3 | classify-failure.md | Map errors to taxonomy, extract missing_op / shape_inference / accuracy signals (skip on full success) |
| 4 | build-report.md | Assemble structured report, emit <!-- agent-result --> marker for orchestrator |
