Please do not open a public issue for security reports.
Use GitHub's private vulnerability reporting on this repository (Security → Report a vulnerability). That opens a private advisory where we can coordinate a fix and disclosure without exposing users in the meantime.
If private reporting is unavailable to you, contact @KitsuneKode on GitHub to arrange a channel.
Reports are handled on a best-effort basis — this is a community-maintained project — but a clear reproduction (what you ran, what happened, what you expected) gets triaged fastest.
Only the latest release is supported. Kunai ships as a rolling CLI; fixes land
on main and go out in the next release rather than being backported to older
tags. kunai doctor reports the version; kunai upgrade updates it.
Kunai launches external tools (mpv, curl/curl-impersonate, yt-dlp,
ffprobe), fetches remote stream metadata and media URLs, and stores local
state (SQLite history/cache, config, an install manifest). The kinds of reports
most relevant here:
- Argument or command injection into a spawned process — e.g. a provider response or URL reaching argv unescaped.
- SSRF or unsafe redirects in stream resolution, the relay, or update downloads (redirects are screened; bypasses matter).
- Local privilege/integrity issues: installer or updater writing outside its install root, unsafe file permissions on credentials or locks, lock ownership confusion between concurrent processes.
- Secrets exposure: tokens, cookies, or private URLs leaking into logs,
support bundles, the
--debuglog, or analytics payloads. Analytics must stay strictly opt-in — a path that sends anything without an explicit user action is a bug worth reporting. - Supply chain: tampered or mismatched release/update artifacts, or the checksum verification being skippable.
- Vulnerabilities in
mpv,curl,yt-dlp, or providers' upstream sites themselves — report those to their own projects. - Content returned by third-party provider sites (ads, tracking on their end).
- Attacks requiring the ability to modify Kunai's own files or environment — if an attacker already controls that layer, the threat model is different.
- The public docs site's static hosting provider.
- Acknowledgement of the report.
- A fix or a documented decision, coordinated in the private advisory.
- Credit in the release notes if you want it.
Please give us reasonable time to ship a fix before public disclosure.