Skip to content

Security: Hozi-AI/CVOps

Security

SECURITY.md

Security Policy

Supported versions

Version Supported
main ✅
Older releases ❌

Reporting a vulnerability

Do not open a public GitHub issue for security vulnerabilities.

Please send a report to yr0556772363@gmail.com with the subject line [CVOps Security] <short description>.

Include:

  • A clear description of the vulnerability and its potential impact
  • Step-by-step reproduction instructions (proof-of-concept if possible)
  • Any suggested mitigations or patches

Response timeline:

  • Acknowledgement within 48 hours
  • Full response or fix timeline within 7 days

We follow coordinated disclosure: we ask that you keep the vulnerability confidential until a fix is released, or 90 days have elapsed since the report, whichever comes first.

Scope

In scope:

  • Authentication or authorisation bypass (JWT, org boundaries)
  • Token leakage or replay attacks
  • SQL injection or arbitrary data exfiltration
  • Server-side request forgery (SSRF)
  • Privilege escalation across organisations or roles

Out of scope:

  • Vulnerabilities in third-party dependencies that are not directly exploitable in CVOps
  • Theoretical issues without a reproducible proof-of-concept
  • Issues that require physical access to the host machine
  • Denial-of-service attacks (rate limiting is out of scope for the current phase)

Security design notes

  • All JWTs carry a jti (JWT ID) claim; revoked tokens are blacklisted in Redis for the remainder of their lifetime
  • Refresh token rotation: the consumed refresh token is blacklisted before a new pair is issued
  • Blob storage: clients receive presigned S3 URLs (15-min TTL for GET, 60-min for PUT); raw bytes never pass through the API
  • All resources are scoped to org_id; cross-org access is blocked at the DB query level
  • Passwords are hashed with bcrypt (passlib, cost factor 12)

Preferred language

English or Hebrew.

There aren't any published security advisories