Security
Permission first. Always.
Sentily is a security product, so we hold ourselves to the controls we check for. This page describes our current MVP practices — it is not a certification.
Safe scanning policy
- Only verified assets with an explicit, recorded, non-expired authorization are scanned.
- Authorization and verification are rechecked by the worker before every operation.
- Passive checks only: no exploit payloads, credential attacks or intrusive crawling.
- Workers block private, loopback, link-local, multicast and cloud-metadata addresses and pin resolved IPs.
Data protection
- Tenant isolation is enforced in the API, database queries and workers.
- Passwords hashed with Argon2id; session cookies are HTTP-only, Secure and SameSite.
- Integration tokens and secrets are encrypted at rest and never exposed to the browser.
- Administrative and privileged actions are recorded in an immutable audit log.
AI remediation
- Only minimal, redacted code context is sent to the AI provider.
- Repository content, scan output and AI output are treated as untrusted input.
- Patches are validated in an isolated sandbox without production credentials.
- Pull requests are opened as drafts after explicit approval and are never merged automatically.
Responsible disclosure
- Report suspected vulnerabilities to [email protected].
- Please give us reasonable time to investigate and fix before public disclosure.
- Do not access other customers' data, degrade service or use social engineering.
- We do not currently hold ISO 27001, SOC 2 or similar certifications.