Skip to content

Security: one-more-refactor/flick

Security

SECURITY.md

Security policy

Supported versions

The latest release (and the current main branch). There are no backport branches; fixes ship as a new release.

Reporting a vulnerability

Please report vulnerabilities privately via GitHub security advisories (the "Report a vulnerability" button under the repository's Security tab). Do not open a public issue for anything exploitable.

You can expect an acknowledgment within a few days. This is an indie project — there is no bug bounty — but reporters get credit in the advisory and release notes (or anonymity, if preferred).

Scope notes

Areas worth attention, roughly in order of interest:

  • Auth: session cookies (flick_session, HttpOnly, SameSite=Lax, Secure behind TLS), the email-first login flow, email login codes, OAuth/OIDC callback handling, guest-session minting.
  • Uploads & parsers: PDF, EPUB, TXT, and Kindle-clippings parsing of attacker-supplied files (25 MB limit), plus HTML extraction for imports.
  • URL import / SSRF: POST /api/import/url fetches user-supplied URLs. The guard resolves DNS, rejects non-global addresses (loopback, RFC1918, CGNAT, link-local, ULA, v4-mapped v6), and pins the connection to the vetted IP to prevent rebinding. Bypasses of this guard are squarely in scope.
  • Rate limiting: the X-Forwarded-For trust rule (only honored from private/loopback peers) and anything that lets a client dodge the auth endpoint limits.
  • Access control: all book/stats routes are scoped to the session user; any cross-user read or write is a serious bug.

Out of scope: denial of service via volume alone, reports about missing security headers on a bare (unproxied) HTTP deployment, and anything requiring a compromised host.

There aren't any published security advisories