Your First Security Prompts

A prompt is just the instruction you type into an AI tool. The difference between a vague prompt and a great one is the difference between a generic, useless answer and advice you can act on immediately. In this lesson you will learn a simple formula for writing strong security prompts and practice with prompts you can reuse for the rest of your life.

What You'll Learn

• A repeatable formula for writing effective security prompts
• Ready-to-use prompt templates for common beginner tasks
• How to push AI for better, safer answers through follow-ups
• The privacy guardrails to apply to every prompt

The SAFE Prompt Formula

Most beginners type three lazy words and wonder why the answer is generic. Use this four-part formula instead. Think SAFE:

• S — Situation: describe your context. ("I am a university student using a personal Windows laptop and an iPhone.")
• A — Ask: state exactly what you want. ("Walk me through enabling two-factor authentication on my email.")
• F — Format: say how you want it. ("Give me numbered steps a beginner can follow.")
• E — Expertise level: tell it who you are. ("Assume I have zero technical background.")

Put together:

I am a university student using a personal Windows laptop and an iPhone. Walk me through enabling two-factor authentication on my Gmail account. Give me numbered steps a complete beginner can follow, and explain why each step matters.

That single prompt produces a clear, personalized walkthrough instead of a wall of jargon. The more context you give, the better the output — with one exception covered below.

Reusable Prompt Templates

Copy these into your notes. Replace the bracketed parts and you have an instant security assistant.

Explain a concept:

Explain [security term] to someone with no technical background. Use a simple analogy, give one real-world example, and tell me one thing I should do about it.

Assess a suspicious message:

I received this message. Assess whether it looks like phishing or a scam, list the specific warning signs, and tell me the safe action to take. Here is the message: [paste message with personal details removed].

Harden an account:

Give me a beginner checklist to secure my [service, e.g. Instagram] account. Order the steps from most to least important and explain each in one sentence.

Build a study plan:

Create a 4-week beginner study plan to learn cybersecurity fundamentals, assuming 3 hours per week and no prior experience. Include free resources and a small weekly exercise.

Quiz yourself:

Act as a cybersecurity tutor. Ask me one beginner question at a time about online safety, wait for my answer, then tell me if I am right and explain why. Start now.

Follow-Ups: Where the Real Value Is

The first answer is rarely the best one. Treat AI like a conversation, not a vending machine. Powerful follow-ups:

• "Explain that last point more simply — I did not follow it."
• "What could go wrong if I do this incorrectly?"
• "Is there a free tool for this instead of a paid one?"
• "Now check your own answer — is anything you said outdated or risky?"

That last one is surprisingly effective. Asking the model to critique itself often surfaces caveats it left out the first time. Remember the course rule: AI advises, you verify.

The Privacy Guardrail on Every Prompt

There is one rule that overrides "give more context": never include real secrets. Before you hit enter, scan your prompt for:

• Real passwords or PINs — never, under any circumstances.
• Full account numbers, card numbers, or government ID numbers.
• Other people's private information.
• Multi-factor authentication codes.

You can almost always get the same help with placeholders. Instead of pasting a real account number, write "my bank account number" or use "1234". The AI understands the structure without seeing your secrets.

A Quick Hands-On Exercise

Open ChatGPT or Claude and run this exact prompt, filling in your own details:

I am a beginner with zero security background. Act as my personal security coach. Ask me 5 quick questions about how I currently use passwords, email, and my phone, one at a time. After my answers, give me a prioritized 5-step action plan to improve my security this week. Keep it simple and encouraging.

This turns a generic chatbot into a personalized coach in under two minutes. Notice how giving it a role ("act as my personal security coach") and a format ("one at a time", "5-step action plan") changes the quality of the help.

Your Homework for This Lesson

Save the five prompt templates above into a note titled "My AI Security Prompts." Then run the "Harden an account" template on whichever account matters most to you. Complete at least the top two steps it recommends today. You have just turned a prompt into real protection.

Key Takeaways

• The SAFE formula (Situation, Ask, Format, Expertise) turns vague prompts into actionable answers.
• Reusable templates for explaining, assessing, hardening, planning, and quizzing make AI an instant security assistant.
• Follow-up questions — especially asking the AI to critique itself — unlock the best answers.
• The privacy guardrail overrides everything: never include real passwords, full account numbers, IDs, or MFA codes.
• Giving AI a role and a format dramatically improves the usefulness of its output.