Overview
Krewvio is SOC 2 Type II certified and undergoes annual third-party penetration testing. All customer data is encrypted at rest with AES-256 and in transit with TLS 1.3. Production systems are isolated, access is logged, and incident response is rehearsed quarterly.
Our latest SOC 2 report and penetration test letter are available under NDA — email security@krewvio.com.
Infrastructure
- 01Hosting — Amazon Web Services, us-east-1, multi-AZ for high availability.
- 02Network — production VPC is fully private. No public-facing databases. All ingress through a managed WAF.
- 03Compute — containerized workloads on managed Kubernetes. No long-running shell access to production hosts.
- 04Backups — automated, encrypted, retained for 30 days. Quarterly restore drills.
- 05Disaster recovery — RTO of 4 hours, RPO of 15 minutes for the production database.
Encryption
Every byte of customer and homeowner data is encrypted at rest with AES-256, keys managed in AWS KMS with annual rotation. Traffic between every client and Krewvio is encrypted with TLS 1.3; legacy ciphers and SSL 3.0 are disabled.
Database backups are encrypted with a separate key. Disk volumes and S3 buckets are encrypted at the storage layer. Production secrets live in AWS Secrets Manager, never in source control.
Access control
- 01Every Krewvio employee has a hardware-key MFA token (Yubikey) required for production access.
- 02Production access is gated through a bastion host with full session recording.
- 03Access is granted on a least-privilege basis and reviewed quarterly.
- 04New hires complete security training before any production access is granted.
- 05Offboarding revokes all access within 4 hours of termination.
Application security
- 01Every code change is reviewed by a second engineer before merge.
- 02Automated dependency scanning blocks known-vulnerable libraries.
- 03Static analysis runs on every pull request — secret scanning, SAST, license compliance.
- 04Dynamic analysis runs nightly against staging.
- 05Annual third-party penetration testing; remediation priority tracked publicly inside the company.
- 06Coordinated disclosure: security@krewvio.com. 24-hour acknowledgement, 90-day remediation target.
AI Agent data handling
The AI Agent sends conversation content to model providers (Anthropic and OpenAI) under zero-data-retention agreements. Providers do not store homeowner messages after a response is generated, and they do not use customer or homeowner data to train models.
Drafts are reviewable in your dashboard before they go live unless you explicitly enable auto-reply. The AI Agent never sends a homeowner message that contains another customer's data.
Incident response
- 01On-call engineer paged within 5 minutes of any production alert.
- 02Security incidents triaged by the security lead within 30 minutes.
- 03Customer notification for confirmed data incidents within 72 hours, in plain language, with the scope and the fix.
- 04Quarterly tabletop drills covering ransomware, credential compromise, and subprocessor breach scenarios.
- 05Annual external red-team exercise.
Subprocessors
A full list of current subprocessors and their roles is at krewvio.com/subprocessors. We give 30 days' written notice before adding a new one with access to customer data.
Compliance and audits
- 01SOC 2 Type II — annual audit, available under NDA.
- 02GDPR — Data Processing Addendum available for any EU-based customer (or any customer with EU-resident homeowners).
- 03CCPA — homeowner data-subject requests honored within 45 days.
- 04TCPA — every SMS sent through Krewvio respects consent, opt-out, and quiet-hours rules.
Reporting a vulnerability
Email security@krewvio.com. Include reproduction steps and a PoC if you have one. We respond within 24 hours, fix critical issues inside 7 days, and credit reporters publicly with permission. Krewvio does not pursue legal action against good-faith researchers operating within the scope at krewvio.com/security/scope.