WebAssessment

Intelligent bug bounty program tracking with two-tier detection strategy using GitHub Compare API.

🎯 Overview

Automatically tracks new bug bounty programs across HackerOne, Bugcrowd, Intigriti, YesWeHack, and other platforms. Uses a smart two-tier detection system to minimize bandwidth while maximizing discovery speed.

🏗️ Architecture

Two-Tier Detection Strategy

Tier 1: Fast Detection

• Checks domains.txt commits (500KB vs 10MB+ JSON)
• Zero changes = instant exit (sub-second)
• Changes detected = proceed to Tier 2

Tier 2: Precise Analysis

• Uses GitHub Compare API (gets ONLY the diff)
• Extracts new programs and scope changes
• Caches lightweight metadata (1-2KB per program)

Data Sources

• Primary: arkadiyt/bounty-targets-data (updates every 30 minutes)
• Platforms: HackerOne, Bugcrowd, Intigriti, YesWeHack, Federacy
• Update frequency: Check as often as needed without hitting rate limits

📦 Installation

cd ~/.claude/skills/hacking/bug-bounties
bun install
./bounty.sh init

🚀 Usage

Initialize (First Time)

./bounty.sh init

Establishes baseline by recording current commit SHAs for all tracked files.

Update Programs

./bounty.sh update

Checks for:

• 🆕 New programs launched
• 📈 Scope expansions on existing programs
• 💰 VDP → Paid bounty upgrades

Show Recent Discoveries

# Last 24 hours (default)
./bounty.sh show

# Last 7 days
./bounty.sh show --last 7d

# Last 48 hours
./bounty.sh show --last 48h

# All cached programs
./bounty.sh show --all

Search Programs

# Search by name
./bounty.sh search "stripe"

# Search by platform
./bounty.sh search "hackerone"

📊 Output Examples

Update Command

🔍 Checking for new bug bounty programs...

📊 TIER 1: Fast change detection
🆕 Changes detected! 3 commits since last check

🔬 TIER 2: Detailed analysis of platform changes
  Checking hackerone...
    🔄 2 commits found
  Checking bugcrowd...
    ✓ No changes
  Checking intigriti...
    🔄 1 commits found
  Checking yeswehack...
    ✓ No changes

⏱️  Completed in 2.3s

============================================================
📊 UPDATE SUMMARY
============================================================
🆕 New programs:        3
📈 Scope expansions:    2
💰 Upgraded to paid:    1
✅ Platforms checked:   5
⏱️  Duration:            2.3s
============================================================

🆕 NEW PROGRAMS:

1. [HACKERONE] Example Corp
   URL: https://hackerone.com/example
   Bounty: 💰 Yes
   Max Severity: critical
   Scopes: *.example.com, api.example.com, admin.example.com...

Show Command

📋 Bug bounty programs discovered in the last 24h

1. [HACKERONE] Robinhood Markets (new_program)
   URL: https://hackerone.com/robinhood_markets
   Bounty: ❌ VDP only
   Max Severity: 🔴 CRITICAL
   Scopes (8):
     - *.robinhood.com
     - *.robinhood.net
     - *.1integrations.com
     ... and 5 more
   Discovered: 10/20/2025, 10:30:00 PM

2. [BUGCROWD] Acme Corp (upgraded_to_paid)
   URL: https://bugcrowd.com/acme
   Bounty: 💰 Paid
   Max Severity: 🟠 HIGH
   Scopes (3):
     - *.acme.com
     - api.acme.io
   Discovered: 10/20/2025, 6:15:00 PM

Total: 2 program(s)

💡 Tip: Use "initiate-recon <number>" to start testing a program

🗂️ File Structure

bug-bounties/
├── README.md                          # This file
├── package.json                       # Dependencies
├── bounty.sh                          # CLI wrapper
├── state.json                         # Tracking state
├── src/
│   ├── types.ts                       # TypeScript types
│   ├── config.ts                      # Configuration
│   ├── state.ts                       # State management
│   ├── github.ts                      # GitHub API client
│   ├── tracker.ts                     # Main tracker logic
│   ├── init.ts                        # Initialize command
│   ├── update.ts                      # Update command
│   └── show.ts                        # Show/search command
├── cache/
│   ├── programs_metadata.json         # Lightweight program cache
│   └── recent_changes.json            # Last 30 days of changes
└── logs/
    └── discovery.jsonl                # Audit trail

🎨 Integration with Hacking Skill

This system integrates with the main hacking skill to enable:

1. Automatic Discovery: "Update bug bounties" → finds new programs
2. Quick Browsing: "Show new bounties from last 24h" → instant results
3. Automated Recon: "Initiate recon on #2" → launches pentester agent

🔧 Technical Details

API Usage

• GitHub API rate limits: 60/hour (unauth), 5000/hour (auth)
• Our usage: ~2-3 calls per update check
• Check frequency: Can check every 10-15 minutes without issues

Caching Strategy

• Metadata only: ~1-2KB per program (vs 50-100KB in full JSON)
• Auto-expiry: Removes entries >30 days old
• Total size: <1MB for hundreds of programs

Change Detection

Detects three types of changes:

1. New programs: Program didn't exist in cache
2. Scope expansions: More domains/scopes added
3. Upgraded to paid: offers_bounties: false → true

📝 State File Format

{
  "last_check": "2025-10-20T22:00:00.000Z",
  "tracked_commits": {
    "domains_txt": "abc123...",
    "hackerone": "def456...",
    "bugcrowd": "ghi789...",
    "intigriti": "jkl012...",
    "yeswehack": "mno345..."
  },
  "initialized": true
}

🎯 Next Steps

After discovering interesting programs, you can:

1. Review program details: Check scopes, rules, and severity levels
2. Initiate reconnaissance: Launch automated recon workflow
3. Set up monitoring: Track changes to specific programs
4. Start testing: Begin systematic vulnerability assessment

🔗 Resources

• Data source: arkadiyt/bounty-targets-data
• Alternative: projectdiscovery/public-bugbounty-programs
• Platforms tracked:
  • HackerOne: https://hackerone.com
  • Bugcrowd: https://bugcrowd.com
  • Intigriti: https://intigriti.com
  • YesWeHack: https://yeswehack.com

🎓 Design Philosophy

This system implements Option D: Intelligent Two-Tier Hybrid from the deep thinking analysis:

• Speed: Fastest for both "no changes" and "changes detected" cases
• Efficiency: Minimal bandwidth and storage
• Precision: Exact diffs without downloading full datasets
• Scalability: Handles all platforms effortlessly
• Maintainability: Clean architecture, easy to debug

---

Version: 1.0.0 Last Updated: 2025-10-20 Maintained By: PAI System