Security is in our DNA.
OsVault is built with the same security rigor we help you enforce. Explore our architecture, compliance posture, and data handling practices.
Architecture Overview
Zero Trust Data Pipeline
Our Rust-based ingestion engine runs in isolated, stateless containers. It fetches data from NVD, OSV, EPSS, and CISA KEV without storing credentials on disk.
Edge-First Deployment
All web traffic is served through Vercel's global edge network with automatic SSL, DDoS mitigation, and geographic routing for sub-100ms response times.
Isolated GitHub App
The GitHub App is deployed as a separate, sandboxed service. It requests minimal permissions (read-only code access) and processes PR data entirely in memory.
Encrypted Database
All vulnerability data and customer metadata is stored in Supabase (PostgreSQL on AWS) with AES-256 encryption at rest and TLS 1.3 in transit.
Security Controls
A detailed breakdown of every security control across OsVault's infrastructure, data handling, and access management layers.
Data Security
Access Control
Audit & Compliance
Infrastructure
Data Handling
What We Access
- PR diff content (dependency file changes only)
- Repository file tree (for reachability analysis)
- GitHub organization metadata
What We Never Store
- Source code — analyzed in-memory, never persisted
- Personal access tokens — we use GitHub App JWTs
- Passwords or secrets from your codebase
Retention Policy
- Scan results: 90 days (configurable on Enterprise)
- Audit logs: 1 year (immutable)
- Vulnerability data: Updated daily, never deleted
Need more details?
Request our full security questionnaire, DPA, or schedule a call with our security team.
Contact Security Team →