Menu
Run style checks and tests on changed files to verify code quality before committing.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Guide NeMo Speech users through ASR fine-tuning with container setup and Lhotse training.
Get a pull request to green CI. Diagnose and fix CI failures, push fixes, re-trigger CI via the "Run CICD" label, and repeat until all checks pass. Does not post comments — this is a local developer tool.
Fix a GitHub issue in NeMo Speech (NVIDIA-NeMo/NeMo). Read the issue, reproduce the bug with a failing test, implement the fix, and verify tests pass. Only opens a PR if the user explicitly asks for it.
Debug distributed training failures (NeMo, Megatron, PyTorch) from worker stderr logs and optional AIStore daemon logs. Finds root cause across NCCL timeouts, data loading errors, and storage failures.
| name | verify |
| description | Run style checks and tests on changed files to verify code quality before committing. |
Run verification on the current changes:
Find changed files:
git diff --name-only HEAD
Also include any files you've been editing in this session.
Style check on each changed Python file or its parent directory:
python setup.py style --scope <path>
If issues are found, fix them with --fix and report what changed.
Run relevant tests based on which collection was modified:
nemo/collections/asr/ → pytest tests/collections/asr --download -m "not pleasefixme" -v --timeout=300nemo/collections/tts/ → pytest tests/collections/tts --download -m "not pleasefixme" -v --timeout=300nemo/collections/audio/ → pytest tests/collections/audio --download -m "not pleasefixme" -v --timeout=300nemo/collections/speechlm2/ → pytest tests/collections/speechlm2 -m "not pleasefixme" -v --timeout=300nemo/collections/common/ → pytest tests/collections/common -m "not pleasefixme" -v --timeout=300nemo/core/ → pytest tests/core -m "not pleasefixme" -v --timeout=300Report results: summarize passes and failures. For failures, show relevant error output and suggest fixes.