# Adversarial Ml

> Adversarial machine learning, robustness, and security in ML

- Skill: `neuralblitz/adversarial-ml` (Agent Skill)
- Install (CLI): `npx skillmds@latest add neuralblitz/adversarial-ml`
- Raw SKILL.md: https://api.skillmd.com/api/skills/neuralblitz/adversarial-ml/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- License: MIT
- Author: NeuralBlitz (https://skillmd.com/u/neuralblitz)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/neuralblitz/adversarial-ml

---


## What I do
- Understand adversarial attacks
- Build robust models
- Test model security
- Apply adversarial training

## When to use me
When building secure or robust ML systems.

## Key Concepts
- Adversarial examples
- FGSM, PGD attacks
- Adversarial training
- Defensive methods
- Certified robustness
- Model extraction
- Backdoor attacks