Crypto Scam Detector

Real-time cryptocurrency scam detection with database-first architecture. Protects users from phishing, honeypots, rug pulls, and ponzi schemes. No external API calls during checks!

🔍 Crypto Scam Detector v2.0

Database-first cryptocurrency scam detection for OpenClaw

Analyzes crypto addresses for phishing, honeypots, rug pulls, and ponzi schemes using a local database with background sync from Etherscan. Zero external API calls during user checks = instant results!

✨ What's New in v2.0

🚀 Major Architecture Upgrade

• ✅ Database-first design - All checks query local SQLite database
• ✅ Instant results - No API latency during checks (<5ms)
• ✅ No rate limits - User queries never hit Etherscan API
• ✅ Background sync worker - Separate process pulls from Etherscan
• ✅ Transaction message analysis - Decodes and analyzes hex data
• ✅ Auto-queue system - Unknown addresses automatically queued for sync
• ✅ Deep scanning - Detects suspicious keywords in transaction data

🔍 Enhanced Detection

Now catches scams the old version missed:

• ✅ "Lazarus Vanguard" hacking group references
• ✅ "Orbit Bridge Hacker" mentions
• ✅ Private key phishing attempts
• ✅ Exploit recruitment messages
• ✅ And much more...

📦 What's Included

text

crypto-scam-detector/
├── SKILL.md                    # This file
├── DATABASE_ARCHITECTURE.md    # Technical documentation
├── database.py                 # SQLite database layer
├── crypto_check_db.py          # Database-only checker (instant)
├── sync_worker.py              # Background Etherscan sync worker
├── secure_key_manager.py       # Encrypted API key storage
├── install.sh                  # Auto-installer
├── setup.sh                    # API key setup wizard
├── check_address.sh            # Convenience script (sync if needed)
├── requirements.txt            # Python dependencies
└── venv/                       # Virtual environment (created on install)

🚀 Quick Start

1. Install

bash

cd ~/.openclaw/workspace/skills/crypto-scam-detector
bash install.sh

2. Configure Etherscan API Key (Optional but Recommended)

Option A: Interactive Setup (Encrypted storage)

bash

./setup.sh
# Follow the wizard to encrypt your API key

Option B: Environment Variable

bash

export ETHERSCAN_API_KEY="your_key_here"

Get free API key: https://etherscan.io/myapikey

3. Check an Address

bash

# Check address (instant, database-only)
python3 crypto_check_db.py 0x1234567890abcdef1234567890abcdef12345678

4. Run Background Sync Worker

Manual mode:

bash

python3 sync_worker.py
# Runs continuously, processes queue

Batch mode:

bash

python3 sync_worker.py --max-jobs 20
# Process 20 addresses then exit

Cron schedule (recommended):

bash

# Add to crontab
*/10 * * * * cd ~/.openclaw/workspace/skills/crypto-scam-detector && source venv/bin/activate && ETHERSCAN_API_KEY="key" python3 sync_worker.py --max-jobs 30

💡 How It Works

Architecture Flow

text

User checks address
       ↓
┌──────────────────┐
│ crypto_check_db  │ ← Queries local database ONLY
└────────┬─────────┘   (No external API calls)
         │
         ↓
┌──────────────────────┐
│ Local SQLite DB      │
│ ~/.config/crypto-    │
│  scam-detector/      │
│                      │
│ • Addresses          │
│ • Transactions       │
│ • Risk scores        │
│ • Scam indicators    │
│ • Sync queue         │
└────────▲─────────────┘
         │
         │ Background sync
         │
┌────────┴─────────────┐
│ sync_worker.py       │ ← Pulls from Etherscan
│                      │   (Uses your API key)
│ • Reads queue        │
│ • Calls Etherscan    │
│ • Decodes TX data    │
│ • Analyzes messages  │
│ • Stores in DB       │
└──────────────────────┘

User Flow

1. Check address: python3 crypto_check_db.py 0x...
2. If in database: Instant results with full analysis
3. If NOT in database:
   • Returns "unknown" status
   • Automatically adds to sync queue
   • Shows: "⏳ Check again in a few minutes"
4. Background worker syncs it (next cron run or manual trigger)
5. Check again: Full analysis now available

🔍 Detection Capabilities

Scam Types Detected

Type	Detection Method
Phishing	Keyword analysis: "private key", "seed phrase", "verify wallet"
Honeypot	Contract code analysis (unverified contracts)
Rug Pull	Transaction pattern analysis
Exploit Groups	Keywords: "Lazarus", "hack", "exploit", "breach"
Social Engineering	Keywords: "urgent", "claim reward", "airdrop winner"

Risk Scoring

Algorithm factors:

• Suspicious transaction count (+25 per TX, max +50)
• Account age (new addresses: +10)
• Balance patterns (large balance + suspicious TX: +20)
• Contract verification (unverified: +30)

Risk Levels:

• 0-19: ✅ Low Risk
• 20-49: ℹ️ Medium Risk
• 50-79: ⚠️ High Risk
• 80-100: 🚨 Critical Risk

📋 Commands Reference

Check Address

bash

# Human-readable output
python3 crypto_check_db.py 0x...

# JSON output
python3 crypto_check_db.py 0x... --json

Sync Worker

bash

# Add address to queue
python3 sync_worker.py --add-address 0x...

# Run worker (continuous)
python3 sync_worker.py

# Process N addresses then stop
python3 sync_worker.py --max-jobs 20

# Custom delay between addresses
python3 sync_worker.py --delay 2.0

# Show database stats
python3 sync_worker.py --stats

Convenience Script

bash

# Check and auto-sync if needed
./check_address.sh 0x...
# Automatically syncs if not in DB, then shows results

🎯 Example Output

Critical Risk Address

text

🚨 Analysis for 0x098b716b8aaf21512996dc57eb0615e2383e2f96

Risk Score: 100/100 - CRITICAL RISK
Last Updated: 2026-02-20 07:14:32

🚨 KNOWN SCAM DETECTED!

⚙️ Smart Contract
⚠️ NOT VERIFIED on Etherscan
   Transactions: 38
   Balance: 101.802430 ETH

🚨 5 Scam Indicator(s) Detected:
   • Suspicious keyword detected: 'lazarus' (confidence: 80%)
   • Suspicious keyword detected: 'hack' (confidence: 80%)
   • Suspicious keyword detected: 'exploit' (confidence: 80%)
   • Suspicious keyword detected: 'private key' (confidence: 80%)

⚠️ 5 Suspicious Transaction(s):
   • 0x74f7fbfe5a0bd3...
     Reason: Suspicious keyword detected: 'lazarus'
     Message: "Greetings Lazarus Vanguard..."

📋 Recommendations:
  🚫 DO NOT send funds to this address
  ⚠️ This address has been flagged as high risk
  📞 Report the source that gave you this address

Unknown Address (Not Yet Synced)

text

⏳ Analysis for 0xnew_address_not_in_db

Risk Score: 0/100 - UNKNOWN
Last Updated: N/A

⏳ Address not yet in database
   Address not in database. Added to sync queue.

📋 Recommendations:
  ⏳ This address will be analyzed soon
  🔍 Check again in a few minutes
  ⚠️ Exercise caution until analysis completes

⚙️ Configuration

Database Location

Default: ~/.config/crypto-scam-detector/crypto_data.db

Etherscan API Rate Limits

• Free tier: 5 calls/second, 100,000 calls/day
• Each address: 4 API calls (balance, TX count, TX list, code)
• Default delay: 1.5 seconds between addresses (safe for free tier)

Recommended Cron Schedule

bash

# Every 10 minutes, process 30 addresses
*/10 * * * * cd ~/.openclaw/workspace/skills/crypto-scam-detector && source venv/bin/activate && ETHERSCAN_API_KEY="key" python3 sync_worker.py --max-jobs 30 --delay 2.0

# Handles ~4,320 addresses per day

🛡️ Security

• ✅ Encrypted API key storage - AES-256 with PBKDF2
• ✅ No third-party sharing - API key only sent to Etherscan
• ✅ Local processing - All analysis happens on your machine
• ✅ No telemetry - Zero data collection
• ✅ Open source - Fully auditable code

📊 Database Schema

Tables

• addresses - Address info, risk scores, balances, metadata
• transactions - Suspicious transactions with decoded messages
• scam_indicators - Individual red flags per address
• sync_queue - Addresses waiting to be synced

See DATABASE_ARCHITECTURE.md for full technical details.

🔄 Sync Frequency

Default behavior:

• First check → address queued for sync
• Worker processes queue (manual or cron)
• Subsequent checks → instant from database

Recommended: Run worker via cron every 5-10 minutes

💻 OpenClaw Integration

Via Chat

text

"Check if 0x1234... is a scam"
"Is this address safe: 0xabc..."
"Verify 0xdef... before I send ETH"

Automatic Detection

When you check an address, OpenClaw:

1. Runs crypto_check_db.py
2. If not in DB → queues for sync
3. Returns current status
4. Suggests checking again after sync

🐛 Troubleshooting

"Address not in database"

Solution: Wait for background worker to sync it, or manually trigger:

bash

python3 sync_worker.py --add-address 0x...
python3 sync_worker.py --max-jobs 1

"Etherscan API key not configured"

Solution: Set API key via environment or setup wizard:

bash

./setup.sh  # or
export ETHERSCAN_API_KEY="your_key"

Rate limit errors

Solution: Increase delay between addresses:

bash

python3 sync_worker.py --delay 3.0

📈 Performance

• ✅ Check latency: <5ms (database query)
• ✅ Sync time: ~2 seconds per address (4 API calls)
• ✅ Database size: ~1KB per address
• ✅ Capacity: Handles millions of addresses

🆚 Comparison: v1 vs v2

Feature	v1.1.3 (Old)	v2.0.0 (New)
Check speed	2-5 seconds (API calls)	<5ms (database)
Rate limits	Yes (every check)	No (checks only query DB)
TX message analysis	❌ Not analyzed	✅ Fully analyzed
False negatives	High (missed scams)	Low (deep analysis)
Architecture	Direct API calls	Database + background worker
API key usage	Every check	Only background worker

📜 License

MIT License - Free and open source

🤝 Support

• GitHub: https://github.com/trustclaw/crypto-scam-detector
• Issues: Report bugs or request features
• ClawHub: https://clawhub.com/crypto-scam-detector
• Hackathon: NeoClaw Hackathon 2026

🏆 Credits

Developed by Trust Claw Team for NeoClaw Hackathon 2026

Built with:

• SQLite - Local database
• Etherscan API - Blockchain data
• ChainAbuse API - Community scam reports
• Python asyncio - Async operations

---

🔐 Stay safe in crypto! Always verify addresses before sending funds.