GDPR Compliance
GDPR compliance - data subject rights, lawful basis, DPIA, privacy by design, breach notification, consent management, cross-border transfers, PII masking
GDPR Compliance
Data Subject Rights
Rights Overview
| Right | Article | SLA | Implementation |
|---|---|---|---|
| Right of Access | Art. 15 | 30 gun | Data export endpoint |
| Right to Rectification | Art. 16 | 30 gun | Profile edit + audit trail |
| Right to Erasure | Art. 17 | 30 gun | Cascading delete + anonymize |
| Right to Restriction | Art. 18 | 30 gun | Processing flag on record |
| Right to Portability | Art. 20 | 30 gun | Machine-readable export (JSON/CSV) |
| Right to Object | Art. 21 | 30 gun | Opt-out mechanism |
| Automated Decision-Making | Art. 22 | 30 gun | Human review override |
Data Subject Request Handler
interface DSRRequest {
id: string;
type: 'access' | 'rectification' | 'erasure' | 'restriction' | 'portability' | 'objection';
subjectId: string;
verifiedIdentity: boolean;
receivedAt: Date;
deadline: Date; // receivedAt + 30 gun
status: 'received' | 'verified' | 'processing' | 'completed' | 'rejected';
reason?: string;
}
async function handleDSR(request: DSRRequest): Promise<DSRResponse> {
// Step 1: Identity verification ZORUNLU
if (!request.verifiedIdentity) {
return { status: 'rejected', reason: 'Identity not verified' };
}
// Step 2: Check deadline
const daysRemaining = differenceInDays(request.deadline, new Date());
if (daysRemaining <= 5) {
await alertDPO({ type: 'dsr_deadline_approaching', request });
}
// Step 3: Process by type
switch (request.type) {
case 'access':
return await generateDataExport(request.subjectId);
case 'erasure':
return await executeErasure(request.subjectId);
case 'portability':
return await generatePortableExport(request.subjectId, 'json');
case 'rectification':
return await updateSubjectData(request.subjectId, request.corrections);
case 'restriction':
return await restrictProcessing(request.subjectId);
case 'objection':
return await recordObjection(request.subjectId, request.reason);
}
}
Right to Erasure Implementation
async function executeErasure(subjectId: string): Promise<ErasureResult> {
const erasureLog: ErasureStep[] = [];
await db.transaction(async (tx) => {
// 1. Anonymize user record (yasal zorunluluklar haric)
await tx.users.update({
where: { id: subjectId },
data: {
email: `erased-${hash(subjectId)}@deleted.local`,
name: 'Erased User',
phone: null,
address: null,
dateOfBirth: null,
deletedAt: new Date(),
},
});
erasureLog.push({ table: 'users', action: 'anonymized' });
// 2. Delete personal messages
const deletedMessages = await tx.messages.deleteMany({
where: { userId: subjectId },
});
erasureLog.push({ table: 'messages', action: 'deleted', count: deletedMessages.count });
// 3. Delete sessions and tokens
await tx.sessions.deleteMany({ where: { userId: subjectId } });
await tx.refreshTokens.deleteMany({ where: { userId: subjectId } });
erasureLog.push({ table: 'sessions', action: 'deleted' });
// 4. Anonymize audit logs (log kaydi kalir, kisi bilgisi gider)
await tx.auditLogs.updateMany({
where: { actorId: subjectId },
data: { actorId: 'erased', actorEmail: 'erased' },
});
erasureLog.push({ table: 'auditLogs', action: 'anonymized' });
// 5. Notify third-party processors
await notifyProcessors(subjectId, 'erasure');
// 6. Record erasure for compliance
await tx.erasureRecords.create({
data: {
subjectHash: hash(subjectId),
erasedAt: new Date(),
systems: erasureLog,
},
});
});
return { success: true, steps: erasureLog };
}
Lawful Basis for Processing
| Basis | Article | When to Use | Example |
|---|---|---|---|
| Consent | Art. 6(1)(a) | Optional processing, marketing | Newsletter signup |
| Contract | Art. 6(1)(b) | Necessary for service delivery | Order processing |
| Legal obligation | Art. 6(1)(c) | Required by law | Tax records |
| Vital interests | Art. 6(1)(d) | Life-threatening situations | Emergency contact |
| Public interest | Art. 6(1)(e) | Public authority tasks | Government services |
| Legitimate interest | Art. 6(1)(f) | Business need, balanced with rights | Fraud prevention |
Lawful Basis Checklist
- [ ] Documented lawful basis for EACH processing activity
- [ ] Records of Processing Activities (RoPA) maintained
- [ ] Legitimate Interest Assessment (LIA) for Art. 6(1)(f)
- [ ] Special category data has Art. 9 basis
- [ ] Children's data has parental consent (Art. 8)
Data Protection Impact Assessment (DPIA)
When Required (Art. 35)
- Automated decision-making with legal effects
- Large-scale processing of sensitive data
- Systematic monitoring of public areas
- New technology with high privacy risk
- Large-scale profiling
DPIA Template
## Data Protection Impact Assessment
**Project:** [proje adi]
**Date:** [tarih]
**DPO Review:** [evet/hayir]
### 1. Processing Description
- What data: [veri turleri]
- Why: [amac]
- How: [islem yontemi]
- Who: [erisim kimlerde]
- How long: [saklama suresi]
### 2. Necessity & Proportionality
- Lawful basis: [hukuki dayanak]
- Data minimization: [minimum veri mi?]
- Purpose limitation: [amac sinirli mi?]
- Storage limitation: [saklama suresi uygun mu?]
### 3. Risk Assessment
| Risk | Likelihood | Impact | Severity | Mitigation |
|------|-----------|--------|----------|------------|
| Unauthorized access | [L/M/H] | [L/M/H] | [L/M/H] | [onlem] |
| Data breach | [L/M/H] | [L/M/H] | [L/M/H] | [onlem] |
| Purpose creep | [L/M/H] | [L/M/H] | [L/M/H] | [onlem] |
### 4. Measures
- [ ] Encryption at rest and in transit
- [ ] Access controls (RBAC)
- [ ] Audit logging
- [ ] Data minimization applied
- [ ] Retention policy configured
- [ ] Pseudonymization where possible
### 5. DPO Sign-off
Date: [tarih] | Approved: [evet/hayir]
Privacy by Design Patterns
Data Minimization
// YANLIS: Tum veriyi topla
const user = await db.users.findUnique({
where: { id },
// Returns everything including SSN, DOB, etc.
});
// DOGRU: Sadece gerekli alanlari sec
const user = await db.users.findUnique({
where: { id },
select: {
id: true,
name: true,
email: true,
// SSN, DOB, etc. secilmez
},
});
PII Masking
function maskPII(data: Record<string, unknown>): Record<string, unknown> {
const piiFields: Record<string, (val: string) => string> = {
email: (v) => v.replace(/^(.{2}).*(@.*)$/, '$1***$2'),
phone: (v) => v.replace(/^(.{3}).*(.{2})$/, '$1*****$2'),
ssn: (v) => `***-**-${v.slice(-4)}`,
creditCard: (v) => `****-****-****-${v.slice(-4)}`,
ip: (v) => v.replace(/\.\d+$/, '.xxx'),
name: (v) => `${v.charAt(0)}${'*'.repeat(v.length - 1)}`,
};
return Object.fromEntries(
Object.entries(data).map(([key, value]) => {
const masker = piiFields[key];
if (masker && typeof value === 'string') {
return [key, masker(value)];
}
return [key, value];
})
);
}
Encryption at Rest
import { createCipheriv, createDecipheriv, randomBytes } from 'crypto';
const ALGORITHM = 'aes-256-gcm';
function encryptPII(plaintext: string, key: Buffer): EncryptedData {
const iv = randomBytes(16);
const cipher = createCipheriv(ALGORITHM, key, iv);
const encrypted = Buffer.concat([cipher.update(plaintext, 'utf8'), cipher.final()]);
const authTag = cipher.getAuthTag();
return {
ciphertext: encrypted.toString('base64'),
iv: iv.toString('base64'),
authTag: authTag.toString('base64'),
};
}
function decryptPII(data: EncryptedData, key: Buffer): string {
const decipher = createDecipheriv(
ALGORITHM,
key,
Buffer.from(data.iv, 'base64')
);
decipher.setAuthTag(Buffer.from(data.authTag, 'base64'));
return decipher.update(data.ciphertext, 'base64', 'utf8') + decipher.final('utf8');
}
Data Breach Notification (72 Saat Kurali)
Notification Workflow
0h → Breach detected
→ Incident response team activated
→ Containment measures started
24h → Impact assessment completed
→ Affected data subjects identified
→ Breach severity classified
48h → Notification to DPA prepared
→ Data subject notification prepared (if high risk)
72h → DEADLINE: DPA notification submitted (Art. 33)
→ Data subject notification sent if required (Art. 34)
Breach Assessment
interface BreachAssessment {
detectedAt: Date;
nature: string; // What happened
categoriesAffected: string[]; // Data types exposed
subjectsAffected: number; // How many people
likelyConsequences: string[]; // Potential harm
measuresTaken: string[]; // Containment actions
riskLevel: 'low' | 'medium' | 'high';
notifyDPA: boolean; // Required unless low risk
notifySubjects: boolean; // Required if high risk
dpaNotificationDeadline: Date; // detectedAt + 72h
}
function assessBreachRisk(breach: BreachAssessment): string {
const highRiskFactors = [
breach.categoriesAffected.includes('financial'),
breach.categoriesAffected.includes('health'),
breach.categoriesAffected.includes('credentials'),
breach.subjectsAffected > 1000,
];
const highRiskCount = highRiskFactors.filter(Boolean).length;
if (highRiskCount >= 2) return 'high';
if (highRiskCount >= 1) return 'medium';
return 'low';
}
Consent Management
Consent Collection
interface ConsentRecord {
subjectId: string;
purpose: string; // 'marketing_email', 'analytics', 'profiling'
granted: boolean;
grantedAt: Date | null;
revokedAt: Date | null;
method: 'explicit_opt_in' | 'form' | 'api';
version: string; // Privacy policy version
ip?: string;
evidence: string; // What they agreed to (exact text)
}
async function recordConsent(consent: ConsentRecord): Promise<void> {
// Consent must be:
// - Freely given (no pre-ticked boxes)
// - Specific (per purpose)
// - Informed (clear language)
// - Unambiguous (affirmative action)
await db.consents.create({ data: consent });
await auditLog({
action: `consent.${consent.granted ? 'granted' : 'revoked'}`,
actor: consent.subjectId,
resource: consent.purpose,
details: { version: consent.version },
});
}
Consent Anti-Patterns
| Anti-Pattern | GDPR Violation | Dogru Yol |
|---|---|---|
| Pre-ticked checkboxes | Art. 7 - not freely given | Unchecked by default |
| Bundled consent | Art. 7 - not specific | Separate consent per purpose |
| Dark patterns (confusing UI) | Art. 7 - not informed | Clear, plain language |
| No withdrawal mechanism | Art. 7(3) - easy withdrawal | One-click unsubscribe |
| Consent wall (block access) | Art. 7 - not freely given | Allow access without consent |
| Implicit consent | Art. 4(11) - not unambiguous | Explicit opt-in required |
Cross-Border Data Transfer
Transfer Mechanisms
| Mechanism | When to Use | Complexity |
|---|---|---|
| Adequacy decision | EU-approved country (UK, Japan, etc.) | Low |
| Standard Contractual Clauses (SCCs) | US, India, etc. | Medium |
| Binding Corporate Rules (BCRs) | Intra-group transfers | High |
| Explicit consent | One-off transfers | Low |
| EU-US Data Privacy Framework | US companies certified | Medium |
Transfer Impact Assessment (TIA)
## Transfer Impact Assessment
**From:** [EU entity]
**To:** [Non-EU entity, country]
**Mechanism:** [SCC / BCR / Adequacy / Consent]
### Data Transferred
- Categories: [personal data types]
- Volume: [approximate records/month]
- Frequency: [continuous / batch / ad-hoc]
### Recipient Country Assessment
- Privacy laws: [adequacy level]
- Government access: [surveillance risk]
- Legal remedies: [available to EU subjects?]
### Supplementary Measures
- [ ] End-to-end encryption (keys retained in EU)
- [ ] Pseudonymization before transfer
- [ ] Access controls at destination
- [ ] Contractual prohibition on government disclosure
- [ ] Regular compliance audits
Data Minimization & Retention
Retention Policy
| Data Category | Retention Period | Legal Basis | After Expiry |
|---|---|---|---|
| Active user data | Account lifetime | Contract | Anonymize |
| Inactive user data | 2 yil inactivity | Legitimate interest | Delete |
| Transaction records | 7 yil | Legal obligation (tax) | Archive encrypted |
| Marketing consent | Until revoked | Consent | Delete |
| Support tickets | 3 yil after resolution | Legitimate interest | Anonymize |
| Access logs | 1 yil | Legitimate interest | Delete |
| Analytics (aggregated) | Indefinite | Legitimate interest | N/A (no PII) |
Automated Retention Enforcement
async function enforceRetentionPolicies(): Promise<RetentionReport> {
const report: RetentionReport = { deletedCount: 0, anonymizedCount: 0 };
// Delete inactive accounts (2 yil)
const inactiveThreshold = subYears(new Date(), 2);
const inactiveUsers = await db.users.findMany({
where: { lastActiveAt: { lt: inactiveThreshold }, deletedAt: null },
});
for (const user of inactiveUsers) {
await executeErasure(user.id);
report.deletedCount++;
}
// Anonymize old support tickets (3 yil)
const ticketThreshold = subYears(new Date(), 3);
const oldTickets = await db.tickets.updateMany({
where: { resolvedAt: { lt: ticketThreshold }, anonymized: false },
data: { userEmail: 'anonymized', userName: 'anonymized', anonymized: true },
});
report.anonymizedCount += oldTickets.count;
// Delete old access logs (1 yil)
const logThreshold = subYears(new Date(), 1);
await db.accessLogs.deleteMany({
where: { createdAt: { lt: logThreshold } },
});
return report;
}
GDPR Compliance Checklist
Organizational
- [ ] Data Protection Officer (DPO) appointed (if required)
- [ ] Records of Processing Activities (RoPA) maintained
- [ ] Privacy policy published and up to date
- [ ] Employee data protection training completed
- [ ] Vendor Data Processing Agreements (DPA) in place
- [ ] Data breach response plan documented and tested
Technical
- [ ] Encryption at rest (AES-256) for PII
- [ ] Encryption in transit (TLS 1.2+)
- [ ] Access controls with least privilege
- [ ] Audit logging for all PII access
- [ ] Data subject request API/workflow
- [ ] Consent management system
- [ ] Automated retention enforcement
- [ ] PII inventory and data flow map
- [ ] Pseudonymization where applicable
- [ ] Backup encryption and access control
No additional documents ship with this skill.
Related Skills
Data Subject Rights for AI Systems
Implements data subject rights mechanisms for AI systems including right to explanation of AI decisions, contestation procedures, human review, model…
Lawful Basis for AI Training Data
Assesses lawful basis for AI training data processing per EDPB April 2025 report on LLMs and general-purpose AI. Covers legitimate interest balancing…
Managing Consent for Analytics Cookies
Managing consent for analytics cookies and implementing privacy-preserving measurement. Covers GA4 privacy configuration, consent mode fallback behav…
Applying Privacy Design Patterns
Systematic application of the eight privacy design patterns per Hoepman: minimize, hide, separate, abstract, inform, control, enforce, and demonstrat…
User Input
[COMMUNITY] Assess EU Data Act (Regulation 2023/2854) compliance for connected products, data holders, and data processing service providers