Skip to content

Security: RunOnFlux/cumulusvpn

Security

SECURITY.md

Security Policy

CumulusVPN is a privacy product — a security bug here can expose users. We take reports seriously and appreciate coordinated disclosure.

Reporting a vulnerability

Please do not open a public GitHub issue for security vulnerabilities.

Email security@cumulusvpn.com (or info@cumulusvpn.com until the security alias is live). Include:

  • what the issue is and where (component + file/endpoint if known),
  • how to reproduce it, and
  • the impact you think it has.

If you want to encrypt the report, ask for our PGP key in a first (contentless) message and we'll reply with it.

We aim to acknowledge within 72 hours and to keep you updated as we work a fix. Please give us a reasonable window to remediate before any public disclosure; we're happy to credit you (or keep you anonymous — your choice).

Scope

In scope — anything that could compromise user privacy, traffic, or funds:

  • the gateway (gateway/): WireGuard handling, the exit forwarder, the entitlement/chain scanner, the control API (enroll/status/info), rate limiting;
  • the clients (clients/): key generation and storage, config building, discovery + directory-signature verification, the payment flow;
  • the deploy tooling and app specs (deploy/): anything that could let a malicious spec or image reach the fleet.

Out of scope: findings that require a already-compromised device or OS; issues in third-party dependencies (report those upstream — e.g. wireguard-go, wireguard-apple, gVisor — though we still want to know); and the documented, labeled pre-launch gaps (see the README banner and docs/).

What we do and don't hold

By design there is no user database, no accounts, and no traffic or connection logging — so there is very little to leak or subpoena. Payment is a public on-chain fact keyed to a WireGuard public key; identity is the key itself. This posture is part of the product, and reports that undermine it are especially valuable.

Please avoid

Testing that harms users or third parties: no traffic interception of other users, no DoS/load testing against live gateways, no attacks on Flux node operators' machines, and no automated scanning that degrades the network. Use your own keys and your own test deployment.

There aren't any published security advisories