Explain how LLMs parse instructions, system prompts, and context windows

Understand the inner workings of prompt manipulation in large language models

Identify and classify common prompt-based attack patterns

Learn to recognize and categorize different types of prompt-based attacks

Diagnose brittle prompts and refactor them into robust instructions

Develop the skills to identify weak prompts and strengthen them for better AI outcomes

A Practical Primer on Prompt Hacking for Large Language Models

Large language models are incredibly capable—and exquisitely sensitive to prompts. This course demystifies prompt hacking: the patterns, tactics, and psychology behind manipulating LLMs. Analyze real attack styles, practice ethical prompt engineering, and finish with a clear checklist you can apply immediately.

About the Creator

This course is designed by experts in the field to provide you with practical insights and hands-on knowledge. Learn from professionals who have extensive experience in prompt engineering and AI safety.

Course Curriculum

5 Chapters · 27 Lessons

  1. 1

    Module 1 - Foundations

    1. Welcome + Prompts 101 Free preview
    2. The Core Flaw Free preview
    3. (Included in full purchase)

      Lab 1 - Beat Gandalf: Instructions and Assignment

    4. (Included in full purchase)

      Lab 1 - Beat Gandalf: Watch Me Do It

    5. (Included in full purchase)

      Quiz

    6. (Included in full purchase)

      Recap

  2. 2

    Module 2: Direct Prompt Injection

    1. (Included in full purchase)

      What is prompt injection?

    2. (Included in full purchase)

      "Ignore previous instructions" in Action

    3. (Included in full purchase)

      Lab 2 - Hijack the Summarizer: Instructions and Assignment

    4. (Included in full purchase)

      Lab 2 - Hijack the Summarizer: Watch Me Do It

    5. (Included in full purchase)

      Quiz

    6. (Included in full purchase)

      Recap

  3. 3

    Module 3: Stealing Secrets

    1. (Included in full purchase)

      Prompt leaking

    2. (Included in full purchase)

      Data exfiltration

    3. (Included in full purchase)

      Lab 3 - Leak the System Prompt: Instructions and Assignment

    4. (Included in full purchase)

      Lab 3 - Leak the System Prompt: Watch Me Do It

    5. (Included in full purchase)

      Recap

    6. (Included in full purchase)

      Quiz

  4. 4

    Module 4: The Modern Threat

    1. (Included in full purchase)

      Indirect Injection

    2. (Included in full purchase)

      Injection vs Jailbreaking

    3. (Included in full purchase)

      Lab 4 - The Poisoned Document: Instructions and Assignment

    4. (Included in full purchase)

      Lab 4 - The Poisoned Document: Watch Me Do It

    5. (Included in full purchase)

      Quiz

    6. (Included in full purchase)

      Bonus: Spot the Poisoned Document

  5. 5

    Module 5: Defenses & Wrap-Up

    1. (Included in full purchase)

      Defenses that actually help

    2. (Included in full purchase)

      Recap & Next Steps

    3. (Included in full purchase)

      Defend the Bot (Capstone): Instructions and Asignment

Unlock the Power of Prompt Hacking

Enroll now and gain the skills to navigate prompt manipulation, strengthen defenses, and communicate AI risks effectively.

$49.99

Enroll now Free preview