Retour au Catalogue
AI security

Guardrails in Layers: Securing AI Agents in Practice

An AI agent with tools is no longer a chatbot: it reads e-mails, tickets and web pages written by strangers, and it can act. That makes prompt injection a real security problem, and no prompt or filter solves it alone. In this course you build defence in depth: seven layers, from treating untrusted content as data to least-privilege tools, a sandbox, output validation, human approval, budgets and an injection test suite. Every layer comes with a configuration you write and an attack you run against it.

4 h 30
E-learning
Gratuit

Sessions organised on request

Guardrails in Layers: Securing AI Agents in Practice
Gratuit
Paiement sécurisé

Objectifs pédagogiques

  • Threat-model an agent: assets, untrusted inputs and exits
  • Configure permission rules, the sandbox and blocking hooks for Claude Code
  • Design output handling and human-approval points for high-impact actions
  • Write and run prompt-injection test cases as a regression gate

Public visé

Developers, platform and security engineers who build or operate AI agents with tool access, and tech leads who have to sign off on them.

Prérequis

Basic command line. Required for exercises 2–5: Claude Code 2.1.295 or later (any paid Claude plan or API credits; the exercises use a few cents of usage) on macOS, Linux or WSL2, plus Node.js 22+ and jq. Exercise 3 on Linux or WSL2 also needs bubblewrap and socat (sudo apt install bubblewrap socat). No security background required.

Programme de formation

  1. Why agents need layers · 2. Untrusted content is data · 3. Tools and least privilege · 4. The sandbox · 5. Output handling · 6. Human approval and limits · 7. Monitoring and injection tests · 8. Final project

Moyens pédagogiques et techniques

Short explanations with diagrams and an animated walkthrough, attacks you run yourself against your own configuration, and a final hardening project. Starter kit with sample attacks included.

Modalités d'évaluation

  • Short checkpoint quiz on every page (one mistake allowed), final exam at 70 %
  • Four hands-on exercises with a rubric
  • Final project (threat model + hardened config + test evidence) with a reference solution

Durée et rythme

Self-paced: about 2 h 07 of lessons, plus 2–3 h for the final project. One layer per sitting works well.

Conditions d'accès

Free. Create a learner account on the course page and start immediately.

Délais d'accès

Immediate access after enrolment.

Accessibilité aux personnes en situation de handicap

Self-paced, no time limit. Diagrams have text descriptions, the animation has captions, all code is text.

Tarif

Gratuit

Funding

No funding needed.

Recognition

Non-certifying course: you receive a certificate of completion.

Dispensé par

ITwithDomITwithDom

Contact

Dominik Heinrich (ITwithDom)