ÆGIS was the legendary shield carried by Athena. Embedded with Medusa's head, it could paralyze anyone who dared face it.
ÆGIS is a gamified offensive-defensive cybersecurity training platform focused on serverless environments.
Instead of passive lessons and static quizzes, an active APT known as Loki's Shadows attacks the player in real time using realistic payloads while they study.
▸ Command Injection ▸ SSRF
▸ IDOR ▸ Supply Chain Attacks
▸ Broken Authentication ▸ Privilege Escalation
▸ Cloud attack vectors ▸ Defensive mitigation strategies
Under pressure.
Most learning platforms separate theory from practice.
ÆGIS does the opposite.
While reading documentation, solving missions, or reviewing study material, the platform continuously launches simulated attacks against the player.
OBJECTIVES
──────────────────────────────────────
→ real pressure
→ faster pattern recognition
→ offensive-defensive thinking
→ practical security instincts
You are not preparing for attacks. You survive them while learning.
| SYS | MODULE | DESCRIPTION |
|---|---|---|
☽ |
Loki's Shadows | Real-time attack engine with adaptive escalation |
⊞ |
Mission System | Progressive security scenarios with vulnerable code and defensive fixes |
⚡ |
Flashcards | Mission-based spaced repetition system |
▣ |
Simulations | Timed security exams and challenge modes |
◈ |
Study Module | Field files, theory, payload analysis, and code walkthroughs |
✦ |
Leaderboard | Real-time global ranking powered by Supabase |
◈ |
ÆGIS-BOT | Context-aware AI tutor |
☠ |
Loki | Dynamic taunts and adaptive attack behavior |
Loki dynamically scales attack intensity according to the active mission.
MISSION THEME INTERVAL TIMER BURST
─────────────────────────────────────────────────────────────
01 Command Injection 30–45s 15s —
02 IDOR 22–35s 13s —
03 Broken Auth 16–28s 11s 15% double
04 SSRF 12–22s 9s 25% double
05 Supply Chain 9–16s 8s 35% double
06 Final APT 7–13s 7s 50% triple ◀ MAX
Thematic payload examples:
→ IDOR object access manipulation
→ Broken Auth forged JWT payloads
→ SSRF AWS metadata exploitation
→ Supply Chain poisoned dependencies
Attack distribution:
70% mission-specific attacks
30% global attack pool
// every blocked attack → +XP +HP recovery
// every mistake → ÆGIS shield damage
// HP reaches 0% → system collapse
Each mission contains:
[1] Vulnerable scenario
[2] Attack explanation
[3] Defensive implementation
[4] Final checkpoint
| MISSION | THEME |
|---|---|
| Mission 01 | Command Injection |
| Mission 02 | IDOR |
| Mission 03 | Broken Authentication |
| Mission 04 | SSRF |
| Mission 05 | Supply Chain |
| Mission 06 | Final APT |
The platform includes a contextual AI tutor capable of understanding:
→ active section
→ mission progress
→ player HP
→ guardian level
→ current challenge state
Common topics are answered locally with near-zero latency.
All progression flows through:
window._grantXP({
xp: 60,
hp: -20,
blocks: 1,
fails: 1,
label: 'missions:loki',
});| EVENT | RESULT |
|---|---|
| Block Loki attack | +XP +HP recovery |
| Complete mission | Massive XP reward |
| Correct simulation answer | +XP |
| Fail defense | -HP |
| Timeout | Critical HP damage |
ÆGIS/
├── index.html
├── main.js
├── style.css
├── config.js
└── js/
├── ai-router.js
├── missions-attacks.js
├── missions-data.js
├── missions-ui.js
├── personalities.js
├── estudos.js
├── estudos_content.js
├── ranking.js ← STATE wrapped in Proxy → auto Supabase sync
├── nick-screen.js
└── aegis-mobile-nav.js
Frontend → HTML + CSS + Vanilla JavaScript
Backend → Node.js + Express
Database → PostgreSQL (Supabase)
Auth → Supabase Auth
Deploy → Railway
Build → No bundlers / zero dependencies
express ^4.18.2 → HTTP server
cors ^2.8.5 → Cross-origin requests
dotenv ^16.4.5 → Environment variables
node-fetch ^2.7.0 → Server-side fetch
nodemon ^3.1.0 → Dev auto-reload
# install dependencies
npm install
# configure environment
cp .env.example .env
# fill in MISTRAL_KEY, SUPABASE_URL, SUPABASE_KEY
# run dev server
npm run dev
# run production
npm starthttp://localhost:3000
const SUPABASE_URL = 'your-url';
const SUPABASE_KEY = 'your-key';launchAttack();
window._grantXP({ xp: 500, hp: 50 });
window.STATE.aegisHp = 0;
triggerAegisDeath();
onMissionCompleted(1, 200);
navigate('ranking');
console.log(window.STATE);Current
- Real-time Loki attacks
- Dynamic escalation system
- Thematic payload engine
- 6 progressive missions
- Flashcard system
- Study module
- AI tutor integration
- Real-time leaderboard
- Mobile responsive layout
- Guardian login screen
For the full roadmap including Season 01.5 and Season 02 planned features, see ROADMAP.md.
Contributions are welcome. ÆGIS is open source and community-driven.
Before submitting a pull request, please read:
- CONTRIBUTING.md — how to set up the project, branch conventions, and PR guidelines
- CODE_OF_CONDUCT.md — expected behavior in all project spaces
→ fork the repo
→ create a feature branch
→ open a pull request with a clear description
All contributions — code, documentation, missions, translations — are appreciated.
Security is not learned passively.
Recognition comes from exposure.
Instinct comes from repetition.
Discipline comes from pressure.
ÆGIS was built to simulate that pressure.
Distributed under the MIT License.
> LOKI'S SHADOWS IS MONITORING THIS FREQUENCY
> 01000111 01110101 01100001 01110010 01100100 01101001
> Guardian — ÆGIS depends on you


