AI & digital skillsUK · RU · EN · IT
Agent security and prompt injection
Build a testable boundary between what an agent reads and what it is authorized to do. Work from trust mapping through hostile fixtures to incident recovery.
Lessons: 618 quiz questions18 flashcards
Course outline
Free lessons: 2What you will learn
- Separate factual sources from sources of authority.
- Design narrow tools with ownership and expiry checks.
- Bind approval to a specific action version.
- Test leaks, hostile inputs and restoration without duplicate sends.
Frequently asked questions
How is learning structured?
Lessons in this course: 6. Work through explanations, complete practical assignments and check understanding with quizzes. Lesson topics are listed in the course outline.
What can I try for free?
Open lessons: 2. They are marked in the course outline. You can read open materials before registering; an account is needed to save progress across devices.
Which languages can I study in?
Ukrainian, Russian, English and Italian. Switching language preserves course progress for a registered user.
Demo course. The materials still need editorial review.