AdminNeo can run attacker code in the admin browser via a rogue database server
If someone tricks you into connecting AdminNeo to a fake database server, that server can send back a specially crafted reply that makes your browser run code chosen by the attacker. This can happen without you logging in to anything.
| Severity | HighCVSS 8.2 of 10 |
|---|---|
| Being exploited? | UnclearNo reports either way yet |
| Fix available? | Yes5.5.1 · since 30 Oct 2026 |
| How sure is this? | Official sourceBacked by an official source |
How the attack works simplified
- 1Attacker sets up serverFake MySQL server returns crafted version string
- 2Admin connectsAdminNeo connects to the rogue database server
- 3Version string insertedUnescaped into a script tag with valid nonce
- 4Script runsCSP nonce lets injected code execute
- ResultAttacker runs code in the admin's browser
What happened
AdminNeo places the database server's version string unescaped inside a JavaScript script tag that has a valid CSP nonce, so a malicious version string can break out of the JavaScript context and execute arbitrary code, and the CSP does not block it.
Are you affected?
You are affected if you run AdminNeo versions earlier than 5.5.1 and the admin can be made to connect to a database server the attacker controls or can reach.
| Package / product | Affected versions | Fixed in |
|---|---|---|
| adminneo-org/adminneo · composer | < 5.5.1 | 5.5.1 |
What to do
- Update AdminNeo (composer adminneo-org/adminneo) to version 5.5.1 or later
- Only connect AdminNeo to database servers you trust until the update is applied
- Log out of AdminNeo sessions and avoid using it while browsing other sites
Timeline
- 30 Oct 2026, 22:52 UTCPublicly disclosed
- 30 Oct 2026, 22:52 UTCReported by GitHub Advisory · GitHub Advisory
- 30 Oct 2026, 22:53 UTCPicked up by Zero Radar
Sources 1 report, oldest first
Official vendor, government CERT or advisory database
- AdminNeo Vulnerable to XSS via Rogue MySQL Server Version String (CSP Nonce Bypass)GitHub Advisory · 30 Oct 2026, 22:52 UTC
Technical details
- ID
- CVE-2026-96632 · GHSA-rx7m-wxwq-vxr9
- CVSS
- 8.2 / 10
- Disclosed
- 30 Oct 2026, 22:52 UTC
- On Zero Radar
- 30 Oct 2026, 22:53 UTC (1 min after disclosure)
- Last change
- 30 Oct 2026, 22:53 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