Skip to content

Security: OpenSelfHosting/openkey_docs

Security

SECURITY.md

Security Policy

Reporting a vulnerability

If you believe you have found a security issue in OpenKey (server, extension, CLI, docs, or the app), please report it privately.

Do not open a public GitHub issue for exploitable vulnerabilities.

Preferred contact:

Please include:

  1. Affected package (openkey_server, openkey_extension, openkey_cli, official app binaries, …)
  2. Version / commit hash if known
  3. Steps to reproduce
  4. Impact (e.g. ciphertext disclosure, auth bypass, local privilege)

We aim to acknowledge reports within 7 days and to ship a fix or mitigation for confirmed issues as quickly as practical.

Scope

In scope:

  • Cryptographic design flaws that weaken zero-knowledge guarantees
  • Authentication / authorization bypass on the sync API
  • Plaintext vault leakage from the extension, native messaging bridge, or CLI beyond intentional fill/save flows
  • Remote code execution or privilege escalation in supported packages

Out of scope (unless you can show a practical exploit chain):

  • Denial of service against a self-hosted instance
  • Phishing / social engineering of master passwords
  • Compromised devices with an unlocked vault (the vault key is in memory by design)
  • Issues that require physical access plus an unlocked session

Security model (summary)

  • The sync server stores ciphertext only. Master passwords and plaintext vault keys must never leave the client.
  • Access JWTs are short-lived; refresh tokens are hashed at rest and rotated on use.
  • Soft-deleted vault items remain as tombstones until peers sync; LWW uses per-item revision.

See openkey_docs/guide/security.md for the full threat model.

There aren't any published security advisories