Execute compliance and security auditing for Cursor usage. Triggers on "cursor compliance", "cursor audit", "cursor security review", "cursor soc2", "cursor gdpr".
Compliance and security auditing framework for Cursor IDE usage. Covers SOC 2, GDPR, and HIPAA assessment with audit checklists, evidence collection, and remediation guidance.
| Certification | Status | Notes |
|---|---|---|
| SOC 2 Type II | Certified | Annual audit, report available on request |
| Penetration testing | Annual | Results shared under NDA (Enterprise) |
| Encryption at rest | AES-256 | All stored data |
| Encryption in transit | TLS 1.2+ | All API communications |
| Zero data retention | Available | Via Privacy Mode |
| GDPR compliance | Yes | EU data processing supported |
| HIPAA BAA | Not available (as of early 2026) | See HIPAA section |
Developer Machine
│
├─► Cursor Client ──► Cursor API (US/EU) ──► Model Provider
│ (local) (routing + auth) (OpenAI/Anthropic)
│ │
│ └─► Zero retention agreement
│
├─► Codebase Index ──► Embedding API ──► Turbopuffer (vectors)
│ (no plaintext stored)
│
└─► Local Settings (API keys, preferences)
(never transmitted)
[ ] SSO (SAML/OIDC) configured and enforced
[ ] MFA enabled at Identity Provider level
[ ] RBAC roles assigned: Owner, Admin, Member
[ ] Inactive users deprovisioned (SCIM or manual)
[ ] Access review completed (quarterly)
Evidence:
- SSO configuration screenshot from admin dashboard
- IdP MFA policy documentation
- User list export from Cursor admin
- SCIM sync logs (if applicable)
[ ] Privacy Mode enforced at team level
[ ] .cursorignore configured for sensitive files
[ ] Data classification aligned with .cursorignore patterns
[ ] Model provider data retention agreements documented
[ ] BYOK configuration documented (if applicable)
Evidence:
- Privacy Mode enforcement screenshot
- .cursorignore file contents (committed to git)
- Cursor data use policy acceptance
- API key provider agreements
[ ] All Cursor API calls use TLS 1.2+
[ ] Corporate proxy configured with valid certificates
[ ] No self-signed certificates or TLS bypasses
[ ] Network firewall rules documented
Evidence:
- Network architecture diagram showing Cursor data flows
- Firewall rules for cursor.com domains
- Proxy configuration settings
[ ] Admin dashboard usage analytics reviewed monthly
[ ] Anomalous usage patterns investigated
[ ] Seat utilization tracked for access reviews
Evidence:
- Monthly usage report screenshots
- Incident response log for anomalies
- User activity summary
Data Category: Source code snippets
Processing Purpose: AI-assisted code generation
Legal Basis: Legitimate interest (developer productivity)
Data Location: In-transit only (zero retention with Privacy Mode)
Sub-processors: OpenAI, Anthropic, Turbopuffer (embeddings)
Retention: None (Privacy Mode) or per provider policy (no Privacy Mode)
| Right | Cursor Support |
|---|---|
| Right to access | Account settings at cursor.com/settings |
| Right to erasure | Account deletion removes all server-side data |
| Right to portability | Settings export (settings.json) |
| Right to restriction | Privacy Mode limits processing |
| Right to object | Privacy Mode + .cursorignore |
[ ] Data Processing Agreement (DPA) signed with Cursor (Enterprise)
[ ] Privacy Mode enabled for all EU team members
[ ] Sub-processor list reviewed (cursor.com/privacy)
[ ] Data protection impact assessment (DPIA) completed
[ ] Team briefed on not pasting PII into Chat/Composer
Evidence:
- Signed DPA
- Privacy Mode enforcement confirmation
- DPIA document
- Team training records
Current status: Cursor does not offer a Business Associate Agreement (BAA) as of early 2026.
If your organization handles PHI:
1. Enable Privacy Mode (mandatory)
2. Configure .cursorignore to exclude ALL PHI-containing files:
.cursorignore:
**/patient-data/
**/medical-records/
**/hl7/
**/fhir-resources/
**/*.hl7
**/*.ccda
3. Consider BYOK through Azure with BAA:
- Azure OpenAI has HIPAA BAA option
- Route Cursor AI requests through Azure
- Azure handles data governance
4. Train developers: NEVER paste PHI into Chat or Composer
5. Code review policy: verify no PHI in AI-generated code
6. CRITICAL: Consult your compliance team before any Cursor
usage with systems that process PHI
Severity: High
Risk: Code may be retained by model providers for training
Remediation:
1. Admin Dashboard > Privacy > Enable enforcement (immediate)
2. Notify all team members (email)
3. Verify enforcement: check each member's status in dashboard
4. Document: date of enforcement, approval authority
Severity: Medium
Risk: Sensitive files may be included in AI context
Remediation:
1. Create .cursorignore at project root
2. Add patterns for: .env*, secrets/, credentials/, PII directories
3. Commit to git (PR review required)
4. Verify: Cursor Settings > Codebase Indexing > View included files
5. Confirm sensitive files absent from indexed list
Severity: Medium
Risk: Shared or unrotated API keys
Remediation:
1. Audit which team members use BYOK keys
2. Verify keys are personal (not shared team keys)
3. Implement quarterly key rotation schedule
4. Document key management policy
5. Consider centralizing through Azure gateway (Enterprise)
Severity: Medium
Risk: Former employees retaining Cursor access
Remediation:
1. Export current member list from admin dashboard
2. Cross-reference with HR active employee list
3. Deactivate accounts for departed employees
4. Enable SCIM for automatic deprovisioning
5. Schedule quarterly access reviews