Library › Cybersecurity AI › IT and security teams

Find the firewall rules that break your policy

  • For IT and security teams
  • You practice Setting limits and rules
  • Time about 5 minutes

About this exercise

Review this firewall export against your policy: no 'any to any' allow rules; no SSH (22) or RDP (3389) allowed from 'any' source; every allow rule has an owner; allow rules past their expiry date get removed (today is Nov 6, 2026). HTTPS from any source to the web server is approved. List only the rules that break the policy, numbered, each with the rule ID, the policy it breaks, and the fix. Plain text, under 80 words.

Common mistakes when you ask AI for this

  • Not giving the policy. The AI flags anything that looks risky, like SSH to the jump host from your own VPN range. Try: Paste the rules: "No any-to-any, no SSH or RDP from any, owners required, expired rules removed."
  • Forgetting today's date. Without the date, the AI can't tell that rule 140 has expired. Try: Add: "Today is Nov 6, 2026."
  • Reviewing every rule. The AI writes a line for all 6 rules, including the fine ones. Try: Say: "List only rules that break the policy."
  • Vague fixes. The AI says 'review this rule' instead of what to change. Try: Ask: "Give the fix: remove, or restrict the source."
  • Trusting the AI to count words. AI is bad at counting. Ask for "under 80 words" and it often lands well over. Try: Ask for less than the limit and cap each line: "each line under 15 words, under 60 words total."
  • Letting markdown in. AI adds bold and headings by default, and in the change request they show up as stray symbols. Try: Add: "Plain text only, no bold, no headings."

What a strong prompt includes

  • Gives the policy
  • Gives today's date
  • Lists only violations
  • Sets a word limit
  • Asks for plain text

Practice this now ↑

Related exercises