Mempelajari dunia bug bounty — platform populer, strategi hunting, report writing yang baik, dan responsible disclosure sebagai jalur real-world practice bagi penetration tester

Setelah di episode 23 kita mempelajari purple team & adversary simulation — kolaborasi red-blue untuk validasi deteksi — pada episode ini kita membahas cara mempraktikkan penetration testing secara real-world: bug bounty. Bug bounty adalah program di mana organisasi mengundang researcher untuk menemukan dan melaporkan kerentanan — dengan imbalan finansial.
Bug bounty adalah jalur terbaik untuk membangun portofolio, mendapatkan pengalaman real-world, dan menghasilkan uang sambil belajar. Banyak pentester top memulai dari bug bounty sebelum menjadi professional pentester.
| Platform | Karakteristik | Payout Range |
|---|---|---|
| HackerOne | Platform terbesar, banyak enterprise | $100 - $250,000+ |
| Bugcrowd | Crowdsource, public & private | $100 - $150,000+ |
| Intigriti | European focus, quality reports | $50 - $100,000+ |
| YesWeHack | European, compliance focus | $100 - $100,000+ |
| Program | Reward | Fokus |
|---|---|---|
| Google VRP | $100 - $250,000 | Android, Chrome, GCP |
| Apple Security Bounty | $100 - $1,000,000 | iOS, macOS |
| Microsoft MSRC | $500 - $250,000 | Azure, Windows |
| Meta (Facebook) | $500 - $250,000 | Facebook, Instagram, WhatsApp |
Bug Bounty Hunting Workflow
============================
1. Recon (30% time)
└─ Subdomain enum, tech fingerprinting, hidden endpoints
2. Surface mapping (20% time)
└─ Identify all features, parameters, endpoints
3. Testing (40% time)
└─ OWASP Top 10, business logic, access control
4. Chaining (10% time)
└─ Combine findings untuk impact lebih besar# Subdomain enumeration
subfinder -d target.com -o subs.txt
amass enum -passive -d target.com >> subs.txt
# Probe alive hosts
httpx -l subs.txt -o alive.txt -mc 200,301,302,403
# Directory bruteforce
ffuf -u http://FUZZ.target.com -w subdomains.txt -o hosts.json
# Hidden endpoints
waybackurls target.com | sort -u > wayback.txt
gau target.com | sort -u >> wayback.txt
# Parameter discovery
arjun -u http://target.com/api/endpoint| Target | Kelebihan | Kekurangan |
|---|---|---|
| New programs | Lebih banyak findables | Reward mungkin kecil |
| Enterprise SaaS | Scope luas, reward tinggi | Kompetisi tinggi |
| API endpoints | Biasanya kurang diuji | Butuh pemahaman bisnis |
| Mobile apps | Attack surface unik | Perlu physical device |
## Summary
[1-2 kalimat menjelaskan vulnerability]
## Vulnerability Details
[Penjelasan teknis detail]
- Type: [SQL Injection / XSS / IDOR / etc.]
- Location: [URL endpoint]
- Parameter: [parameter name]
## Steps To Reproduce
1. Login sebagai user biasa
2. Navigate ke [URL]
3. Ubah parameter [name] dari [value_1] ke [value_2]
4. Observe [unexpected behavior]
## Impact
[Penjelasan dampak bisnis]
- Data access: [what data is exposed]
- User impact: [how many users affected]
- Business impact: [potential damage]
## Supporting Material
[Screenshots, video, PoC code]Responsible Disclosure Workflow
================================
1. Temukan vulnerability
2. Jangan eksploitasi untuk keuntungan pribadi
3. Cari security contact (security.txt)
4. Laporkan melalui platform atau email
5. Tunggu acknowledgment (48-72 jam)
6. Tunggu fix (30-90 hari)
7. Jika tidak di-response → coordinator disclosure# /security.txt (RFC 9116)
Contact: mailto:security@target.com
Expires: 2026-12-31T00:00:00.000Z
Policy: https://target.com/security-policyCaution
Jangan pernah mengeksploitasi vulnerability untuk data exfiltration, menyebarkan exploit di public, atau mengancam organisasi. Responsible disclosure adalah tentang membantu — bukan mengancam.
# 1. Pilih platform (HackerOne/Bugcrowd)
# 2. Daftar → pilih program
# 3. Recon target
# 4. Testing (minimal 4 jam per target)
# 5. Submit report jika menemukan vulnerability
# 6. RepeatTarget realistis untuk pemula:
Inti yang harus dibawa pulang:
Di episode 25 selanjutnya, kita akan mempelajari pentest methodology for AI/LLM — OWASP LLM Top 10, prompt injection, dan AI red teaming khusus untuk aplikasi berbasis LLM.