Skip to content

Security: OneHillAI/OKGF

Security

SECURITY.md

Security Policy

OKGF is a specification plus a small reference validator. The security surface is limited, but two areas matter and we take reports on both.

Scope

  • The reference validator (tools/okgf_validate.py) - parsing untrusted bundles (YAML, tar-free but file I/O), and signature verification. A bug that lets a malformed bundle crash, hang, or mislead the validator is in scope.
  • The trust model in the spec - the x-okgf-signature scheme and the conformance rules. If you find a way a page could be treated as trusted when it should not be (or vice versa), that is a spec-level security issue and we want to hear it.

Out of scope: vulnerabilities in tools that implement OKGF (report those to the implementation's own project).

Reporting

Email dev@onehill.org with a description and, if possible, a minimal reproducing bundle. Please do not open a public issue for a security report.

You can also use GitHub's private vulnerability reporting on this repository.

We aim to acknowledge a report within a few business days and will keep you updated as we assess and fix.

There aren't any published security advisories