Skip to main content

Edge Model Compression

Enables efficient deployment of large ML models on edge devices using techniques like quantization and pruning.

Install this skill

or
32/100

Security score

The Edge Model Compression skill was audited on Feb 9, 2026 and we found 12 security issues across 2 threat categories, including 4 high-severity. Review the findings below before installing.

Categories Tested

Security Issues

high line 213

Eval function call - arbitrary code execution

SourceSKILL.md
213model.eval()
high line 250

Eval function call - arbitrary code execution

SourceSKILL.md
250original.eval()
high line 258

Eval function call - arbitrary code execution

SourceSKILL.md
258quantized.eval()
high line 488

Eval function call - arbitrary code execution

SourceSKILL.md
488teacher.eval()
low line 1335

External URL reference

SourceSKILL.md
1335- [TensorFlow Model Optimization](https://www.tensorflow.org/model_optimization)
low line 1336

External URL reference

SourceSKILL.md
1336- [PyTorch Quantization](https://pytorch.org/docs/stable/quantization.html)
low line 1337

External URL reference

SourceSKILL.md
1337- [ONNX Runtime](https://onnxruntime.ai/)
low line 1341

External URL reference

SourceSKILL.md
1341- [AutoKeras](https://autokeras.com/)
low line 1347

External URL reference

SourceSKILL.md
1347- [Quantization and Training of Neural Networks](https://arxiv.org/abs/1712.05877)
low line 1348

External URL reference

SourceSKILL.md
1348- [Pruning Filters for Efficient ConvNets](https://arxiv.org/abs/1608.08710)
low line 1349

External URL reference

SourceSKILL.md
1349- [Distilling Knowledge in a Neural Network](https://arxiv.org/abs/1503.02531)
low line 1350

External URL reference

SourceSKILL.md
1350- [DARTS: Differentiable Architecture Search](https://arxiv.org/abs/1806.09055)
Scanned on Feb 9, 2026
View Security Dashboard
Installation guide →