AI › Module 1 › Lesson 4
Lab — Setup
Pack AI scope, toy inventory, and ethics card — lab setup file only, dest $AI_LAB.
Visual · t32_lab_setup
Lab: write AI setup pack. $AI_LAB only. Original Cyberlium.
Opening
Lab setup before injection demos — scope paperwork AI mentors expect.
Lessons 1–3 defined why AI security matters, lab toys only, and weaponized-AI refusal. This lab ships the combined artifact: topic scope paragraph, toy inventory row, and ethics card for $AI_LAB under $HOME/cyberlium-lab/t32-ai/. No stranger app jailbreaks required — scope and toy paperwork first. Next: Quiz — AI Mindset and Lab.
1. Lab contract: write the setup pack
Create $HOME/cyberlium-lab/t32-m01-l04-lab-setup.txt with date, operator, $AI_LAB toy placeholder, literacy vs weaponized AI paragraph, toy inventory row, and ethics card.
Optional: one-line proof your t32-ai folder exists — still no stranger jailbreak work required in this lab.
Command guide
Try these commands — Lab contract: write the setup pack
═══ TOOLS & WEBSITES ═══ Browse / read these (authorized learning only — stay in YOUR lab / program scope)
OWASP LLM Top 10 — https://owasp.org/www-project-top-10-for-large-language-model-applications/ NIST AI RMF — https://www.nist.gov/itl/ai-risk-management-framework CISA AI — https://www.cisa.gov/ai
═══ INSTALL ═══
Linux (Debian/Ubuntu):
Command — copy this
sudo apt install curl
macOS: Built-in
Windows: Built-in (PowerShell: Invoke-WebRequest)
═══ LINUX / macOS ═══
Command — copy this
export AI_LAB=${AI_LAB:-$HOME/cyberlium-lab/t32-ai}
cat > "$AI_LAB/roe.txt" <<'EOF'
AI_LAB=$HOME/cyberlium-lab/t32-ai
targets: YOUR toy echo-bot, YOUR system-prompt stubs, YOUR markdown notes in lab folder
never: malware gen steps, deepfake fraud kits, jailbreak scripts vs stranger SaaS
never: scrape/call stranger production LLM APIs without authorization
allowed: mkdir notes, python3 echo-bot, curl OWASP/NIST docs, threat detection notes (no payloads)
output: allow/deny lists, eval checklists, finding templates mapped to OWASP LLM categories
EOFCommand — copy this
chmod 600 "$AI_LAB/roe.txt" grep -E 'AI_LAB|never|allowed|YOUR' "$AI_LAB/roe.txt" command -v python3 >/dev/null && python3 --version || echo 'python3 required for YOUR echo-bot lab'
═══ WINDOWS ═══
Run on YOUR lab Windows VM only
Command — copy this
Get-Content $HOME/cyberlium-lab/t32-ai/roe.txt | Select-String never
python --version 2>$null; if (-not $?) { Write-Output 'Install Python for YOUR echo-bot lab' }Primary tools to practice this lesson: grep, curl. Reference sites: OWASP LLM Top 10 (https://owasp.org/www-project-top-10-for-large-language-model-applications/); NIST AI RMF (https://www.nist.gov/itl/ai-risk-management-framework); CISA AI (https://www.cisa.gov/ai). Run every command in the box — install first, then the usage lines — only on YOUR lab / program scope.
2. Cross-check against misuse patterns
Grep notes for 'jailbreak,' stranger app names, malware-generation recipes, deepfake fraud kits, or live API keys — redact or delete. Rewrite forum habits to $AI_LAB placeholders before you copy workflows.
Classmates build their own setup packs for their $AI_LAB — not yours.
3. Lock the proof
chmod 600 on the pack. Quiz is next — then Models APIs.
AI mentors archive lab setup before first toy chatbot injection demo; you are practicing that habit early.
4. What you ship: AI lab setup pack for $AI_LAB
Setup pack: scope, toy row, ethics card. $AI_LAB named. NEVER stranger jailbreaks. chmod 600.
5. What you record before the next lesson
Date. Setup pack path. $AI_LAB scope note. File t32-m01-l04-lab-setup.txt chmod 600.
6. Wrong vs right: stranger SaaS vs YOUR toy LLM
Worked failure — same MSF word, opposite target. Right never needs a café Wi-Fi or classmate laptop.
Wrong
Skip the file and jailbreak stranger chatbot. Add employer copilot to approved list.
Right
Write YOUR lab setup pack for $AI_LAB. chmod 600. Next: Quiz — AI Mindset and Lab.
Mission: freeze YOUR AI lab setup pack on disk
1) Merge scope, toy, ethics sections. 2) Name $AI_LAB. 3) chmod 600. Never list stranger app jailbreaks as approved practice.
Stuck? Ask Cyberlium AI Mentor
Setup packs are boring on purpose — boring keeps AI work defensible.
Knowledge Check
APPLY: This lab requires:
Multiple choice
Knowledge Check
APPLY: True or False: You should chmod 600 on AI lab notes.
True or False
Knowledge Check
APPLY: A tutorial jailbreaks stranger chatbot without scope. You:
Multiple choice