Kener forges admin login cookies using a hardcoded key
If the server's secret key is left empty or unset, an outsider can fake a login cookie and take full control of the admin area without a password. They only need to know an administrator's email address.
| Severity | CriticalCVSS 9.2 of 10 |
|---|---|
| Being exploited? | UnclearNo reports either way yet |
| Fix available? | YesKener 4.1.10 or later · since 26 Oct 2026 |
| How sure is this? | Official sourceBacked by an official source |
How the attack works simplified
- 1AttackerKnows an admin email address
- 2Hardcoded keyUses the built-in DUMMY_SECRET fallback
- 3Forged cookieSigns a kener-user session cookie
- 4Admin APIServer accepts the cookie at /manage/api
- 5Full admin accessCan also forge password-reset tokens
- ResultAttacker gains full administrative control
What happened
Kener versions 3.0.0 before 4.1.10 fall back to a hardcoded DUMMY_SECRET key to sign session cookies when KENER_SECRET_KEY is empty or unset. Attackers can sign their own kener-user cookie with that key to authenticate to /manage/api.
Are you affected?
You are affected if you run Kener 3.0.0 up to (but not including) 4.1.10 and KENER_SECRET_KEY is empty or unset.
Kener 3.0.0 before 4.1.10, when KENER_SECRET_KEY is empty or unset
What to do
- Update Kener to version 4.1.10 or later
- Set KENER_SECRET_KEY to a long, random value, even if you are not yet updating
- Check admin accounts and active sessions for signs of unexpected access
Timeline
- 26 Oct 2026, 22:16 UTCPublicly disclosed
- 26 Oct 2026, 22:16 UTCReported by NVD · NVD
- 26 Oct 2026, 22:19 UTCPicked up by Zero Radar
Sources 1 report, oldest first
Official vendor, government CERT or advisory database
- CVE-2026-116747NVD · 26 Oct 2026, 22:16 UTC
Technical details
- ID
- CVE-2026-116747
- CVSS
- 9.2 / 10
- EPSS
- 0.6% chance of exploitation within 30 days
- Disclosed
- 26 Oct 2026, 22:16 UTC
- On Zero Radar
- 26 Oct 2026, 22:19 UTC (3 min after disclosure)
- Last change
- 26 Oct 2026, 22:19 UTC
The plain-language parts and the attack diagram were written by an AI model (anthropic/claude-haiku-5.5) from the sources above. Facts like versions, exploitation and dates come from the sources themselves. How Zero Radar works