Agent Skills by Nec
Showing 17 vetted skills indexed across 1 GitHub repositories.
artifact-pack
Package repository artifacts into shareable packets with regenerated document formats.
paper-sync
Reconcile manuscript claims with experimental evidence and generate inconsistency reports.
gradient-check
Diagnose gradient flow issues in hybrid quantum-classical training paths.
log-result
Log verified experimental results into docs/EXPERIMENTS.md and update related narratives.
reconcile-results
Aggregate evidence from experiments, papers, notebooks, JSON logs, and checkpoints to identify and resolve study metric discrepancies.
status
Aggregate git state, checkpoints, and artifacts into a concise project status report.
roadmap-sync
Reconcile benchmark truth and publication strategy into a next-phase roadmap.
codex-audit
Compare repository Codex integration against official docs to build a usage matrix.
architecture
Design quantum-classical model architectures for Ottoman-Turkish character recognition.
benchmark-triage
Reconcile experiment metrics and benchmark docs to recommend the next training action.
setup-env
Automate environment setup and validation for quantum ML projects.
compare
Compare model versions and circuit variants from experimental evidence.
train
Orchestrate hybrid model training workflows with checkpoint handling and platform-aware execution.
sync-colab
Synchronize local repository work with Google Colab while preserving checkpoints and experiment metadata.
experiment
Plan and execute structured quantum-classical model experiments with ablations and diagnostics.
review-circuit
Review quantum circuit code for ansatz suitability, gradient flow, and barren plateau risk.
performance-debug
Diagnose training slowdowns and bottlenecks in quantum-aware PyTorch/PennyLane pipelines.