كشف تسرب الخصوصية تلقائياً — فحص أمني أسبوعي في ١٥ دقيقة

#!/usr/bin/env bash

# scan-repo.sh — امسح المستودع بحثاً عن تسريبات محتملة

# الاستخدام: ./scan-repo.sh /path/to/project



TARGET="${1:-.}"

REPORT="/tmp/scan-$(date +%Y%m%d).json"

RESULTS='{"scan_date":"'$(date -Iseconds)'","target":"'$TARGET'","findings":[]}'

echo "$RESULTS" > "$REPORT"



add_finding() {

    local severity="$1" category="$2" file="$3" line="$4" match="$5"

    python3 -c "

import json

with open('$REPORT') as f: r = json.load(f)

r['findings'].append({'severity':'$severity','category':'$category','file':'$file','line':$line,'match':'$match'})

with open('$REPORT','w') as f: json.dump(r, f, indent=2)

"

}



echo "🔍 Scanning $TARGET ..."



# 1. عناوين البريد (محتملة — تجاهل npm/yarn/pip cache)

grep -rnI --include='*.py' --include='*.js' --include='*.ts' --include='*.json' \

    --include='*.md' --include='*.yml' --include='*.yaml' --include='*.env' \

    --include='*.config.*' --include='*.txt' \

    -E '[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}' "$TARGET" \

    2>/dev/null | grep -vE '(node_modules|vendor|__pycache__|\.git|package-lock|yarn\.lock)' \

    | grep -vE '@types|@babel|@eslint|@testing-library|examples?\.com|domain\.com|your@email|test@' \

    | while IFS=: read -r file line match; do

        add_finding "low" "email" "$file" "$line" "$match"

done



# 2. مفاتيح API — Amazon, Google, GitHub, OpenAI, Stripe, Slack

grep -rnI --include='*.py' --include='*.js' --include='*.ts' --include='*.json' \

    --include='*.yml' --include='*.yaml' --include='*.env' --include='*.config.*' \

    -E '(AKIA[0-9A-Z]{16}|sk-[a-zA-Z0-9]{32,}|gh[pousr]_[a-zA-Z0-9]{36,}|AIza[0-9A-Za-z_-]{35}|whsec_[a-zA-Z0-9]{32,}|xox[baprs]-[0-9a-zA-Z-]{10,}|sk_live_[a-zA-Z0-9]{24,})' "$TARGET" \

    2>/dev/null | grep -vE '(node_modules|vendor|__pycache__|\.git)' \

    | while IFS=: read -r file line match; do

        add_finding "critical" "api_key" "$file" "$line" "$match"

done



# 3. أرقام هواتف (أنماط دولية)

grep -rnI --include='*.py' --include='*.js' --include='*.json' --include='*.md' \

    --include='*.txt' --include='*.yml' \

    -E '\+[0-9]{1,3}[0-9]{6,14}' "$TARGET" \

    2>/dev/null | grep -vE '(node_modules|vendor|__pycache__|\.git)' \

    | grep -vE '\+[0-9]{1,3}[0\*]{6,}' \

    | while IFS=: read -r file line match; do

        add_finding "medium" "phone" "$file" "$line" "$match"

done



# 4. عناوين IP داخلية

grep -rnI --include='*.py' --include='*.js' --include='*.json' --include='*.yml' \

    --include='*.yaml' --include='*.env' --include='*.md' \

    -E '\b(10\.|172\.(1[6-9]|2[0-9]|3[01])\.|192\.168\.)[0-9]{1,3}\.[0-9]{1,3}\b' "$TARGET" \

    2>/dev/null | grep -vE '(node_modules|vendor|__pycache__|\.git)' \

    | grep -vE '(example\.|0\.0\.0\.0|127\.0\.0)' \

    | while IFS=: read -r file line match; do

        add_finding "medium" "internal_ip" "$file" "$line" "$match"

done



# 5. أسماء الشركات (قائمة مخصصة)

COMPANIES=("Acme Corp" "Example Inc" "Test Company" "Client Name")

for company in "${COMPANIES[@]}"; do

    grep -rnI --include='*.py' --include='*.js' --include='*.json' --include='*.md' \

        --include='*.yml' --include='*.yaml' --include='*.txt' \

        -F "$company" "$TARGET" 2>/dev/null \

        | grep -vE '(node_modules|vendor|__pycache__|\.git)' \

        | while IFS=: read -r file line match; do

            add_finding "low" "company_name" "$file" "$line" "$match"

    done

done



echo "✅ Scan complete. Report: $REPORT"

python3 -c "

import json

with open('$REPORT') as f: r = json.load(f)

print(f'  Total findings: {len(r[\"findings\"])}')

for level in ['critical','medium','low']:

    count = sum(1 for f in r['findings'] if f['severity']==level)

    if count: print(f'    {level}: {count}')

"