TS-2026-009: Insecure argument handling in Tailscale SSH permitted root accessWATCHING AI — hackernews