How We Protect Your Codebase
Vibe Code Detector is engineered to audit codebases securely. Deep analysis is powered by leading AI models operating under strict zero-retention and zero-training terms. Source files are never retained or trained on.
1. Read-Only Repository Access
GitHub permissions require strictly read-only access to repository contents. We never request write access to your source code or repository settings.
2. Server-Side Secret Redaction
Detected API keys, tokens, and credentials are automatically redacted on the server before report payloads are generated. Sensitive credentials are never logged or transmitted.
3. Zero AI Model Training
Code analyzed by our third-party AI provider is processed under strict zero-retention terms. Code is used solely to generate your launch readiness audit, is never stored after analysis, and is never used to train any model.
4. No Untrusted Code Execution
We perform passive static AST and AI pattern analysis only. We never run customer build scripts, package lifecycle hooks, test suites, or binary executables from scanned repos.
Our Security & Data Guarantees
- •Encrypted in Transit: All data is transmitted over HTTPS/TLS, with HMAC SHA-256 webhook signature verification.
- •Zero Model Training: Code is analyzed under strict zero-retention terms and is never used for AI model training.
- •Fail-Closed Entitlements: Unverified webhook requests or unauthenticated scan requests automatically fail-closed without accessing internal code state.