القائمة
Install Python dependencies before running verification or test steps. Choose the correct path depending on whether the agent builds OpenVINO from source or not.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
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 | python-bootstrap |
| description | Install Python dependencies before running verification or test steps. Choose the correct path depending on whether the agent builds OpenVINO from source or not. |
Install packages only when they are actually needed for a step. Do not pre-install everything upfront.
Use this path when the agent does not compile OpenVINO from source.
The openvino release wheel provides the runtime needed for tests and
conversion checks.
pip install openvino optimum-intel torch \
--extra-index-url https://download.pytorch.org/whl/cpu
For frontend agents that also need ONNX or TensorFlow:
pip install onnx onnxruntime
pip install tensorflow # only if testing TF frontend
Use this path when the agent builds OpenVINO from source.
Do NOT
pip install openvinowhen a source build is present in the same environment. Installing the release wheel alongside a dev build creates twoopenvinopackages that shadow each other, producing confusing import errors and incorrect test results.
Install only the packages that are not produced by the OV build:
pip install torch \
--extra-index-url https://download.pytorch.org/whl/cpu
pip install optimum-intel pytest
# Do NOT add 'openvino' here — use the build output instead
After building OpenVINO, add the build's Python bindings to the path:
# Point Python at the locally built openvino package
export PYTHONPATH="<build_dir>/src/bindings/python:$PYTHONPATH"
# or install the wheel produced by the build
pip install <build_dir>/wheels/openvino-*.whl
agent-results/<agent>/session.md
and continue — do not abort the entire pipeline over a missing optional dep.