Skip to content

Security: Mihir-Rabari/inboxfm-connect

SECURITY.md

Security

We consider the security of this project a top priority. But no matter how much effort we put into it, vulnerabilities can still exist.

If you discover a vulnerability, we'd like to know about it so we can address it as quickly as possible.

Reporting a vulnerability

  • Report privately via the Security tab of this repository: Report a vulnerability. Do not open a public issue for a vulnerability.
  • Do not run automated scanners against any deployment you don't own or control.
  • Do not exploit the vulnerability beyond what's needed to demonstrate it — don't access, download, or modify data that isn't yours.
  • Use your own test account and test data when investigating; don't target other users' accounts or data.
  • If you encounter personal data belonging to others while investigating, stop immediately, don't store or share it, and mention it in your report.
  • Don't disclose the issue publicly until it's resolved.
  • Provide enough detail to reproduce the problem (affected component, steps, and impact).

Out of scope

  • Clickjacking on pages with no sensitive actions.
  • Unauthenticated/logout/login CSRF.
  • Attacks requiring MITM or physical access to a user's device.
  • Any activity that could disrupt the service (DoS).
  • Content spoofing/text injection without a demonstrated attack vector (no HTML/CSS control).
  • Email spoofing.
  • Missing DNSSEC, CAA, or CSP headers.
  • Lack of Secure/HttpOnly flags on non-sensitive cookies.
  • Dead links.
  • UNSANDBOXED execution mode — intended for trusted-operator deployments only, and blocked in EE/Cloud production.
  • Input fields accepting special characters without a demonstrated exploitable sink.
  • Capability-token endpoints (resume URLs, webhook URLs, signed file URLs) — the token is the authorization. Reports must demonstrate an actual disclosure path (logging, Referer leakage, weak entropy), not just "the token is a secret in the URL."
  • Findings whose only attack path is guessing a high-entropy identifier (e.g. a nanoid) with no demonstrated disclosure source.

What we aim to do

  • Acknowledge your report and give an initial assessment on a best-effort basis.
  • Not pursue legal action against reports made in good faith that follow the guidance above.
  • Handle reports confidentially and not share your personal details with third parties without permission.
  • Keep you informed as we work toward a fix.
  • Credit you as the discoverer in any public write-up, unless you'd rather stay anonymous.

This is a small, actively-developed fork — we don't currently run a formal bug bounty program. Good-faith security reports are still very welcome and will be taken seriously.

There aren't any published security advisories