Skills › Cybersecurity

AI Security

Understand and defend against prompt injection, data poisoning, and LLM exploits.

intermediate 🔐 Cybersecurity
0%
Confidence · no data yet
Sign in to track

After this skill you can…

  • Identify and patch prompt injection vulnerabilities
  • Test LLM apps for data exfiltration risks
  • Apply sandboxing and output validation

Prerequisites

Watch (10 videos)

Episode 9: Automating Single-Turn Attacks with PyRIT | AI Red Teaming 101
Microsoft Developer · beginner hands-on
→ Automate single-turn attacks with PyRIT→ Configure datasets for AI security testing
Secure Agent Authorization with OAuth 2.0 | Amazon Bedrock AgentCore | Amazon Web Services
Amazon Web Services · advanced hands-on
→ Secure AI agents using OAuth 2.0→ Configure AgentCore for authorization
Duet AI in Google Cloud - security
Google Cloud Tech · beginner hands-on
→ Identify vulnerabilities with Duet AI in Google Cloud→ Provide recommendations to address security risks
Security Operations & Threat Intelligence | Google Gemini
Google Cloud Tech · beginner hands-on
→ Detect and respond to cloud threats with Gemini→ Investigate security incidents with threat intelligence
Mimecast fights AI-powered cyber threats using AWS innovation | Amazon Web Services
Amazon Web Services · advanced hands-on
→ Develop cutting-edge security solutions using AWS→ Combat AI-powered email threats
Getting Control over Security and Observability Data | Amazon Web Services
Amazon Web Services · advanced hands-on
→ Process security data with AWS Pipelines→ Implement observability in cloud environments
Hunting Malware with Velociraptor (YARA & Memory Forensics)
John Hammond · intermediate hands-on
→ Hunt malware with Velociraptor→ Analyze memory with YARA
Secure Your AI Lifecycle with AI Protection in Security Command Center
Google Cloud Tech · intermediate hands-on
→ Protect AI data→ Detect AI threats→ Respond to AI incidents
Cryptography for Beginners - Full Python Course (SHA-256, AES, RSA, Passwords)
freeCodeCamp.org · beginner hands-on
→ Build a command-line cryptography tool→ Implement hashing and encryption algorithms→ Secure passwords and data
How to implement secure MCP patterns: Identity, authorization, and runtime protection
Google Cloud Tech · intermediate hands-on
→ Implement secure MCP patterns→ Mitigate risks like token theft and data exfiltration

Read (10 articles)

📄
Proof-of-Behavior: The Missing Trust Layer for AI Agents
Dev.to · arian gogani · 2026-04-12
📄
MCP Security in 2026: How to Protect Your AI Agents from Prompt Injection
Dev.to · nexus-api-lab.com · 2026-04-20
📄
Why I spent 14 months building a firewall for AI agents
Dev.to · Alexander Paris · 2026-04-22
📄
AI Could Never Fully Take Over Cybersecurity
Dev.to · Somnath Das · 2026-04-22
📄
AI Security Is Broken — And We’re Testing the Wrong Things
Dev.to · Crucible Security · 2026-04-29
📄
Why Traditional Security Testing Misses 70% of AI Attack Surface
Dev.to · Hernan Huwyler · 2026-04-30
📄
How attackers hijack LLM agents — and how to stop them
Dev.to · Guruprasad J Rao · 2026-04-30