CVE-2026-45387 Open WebUI: Sharing models for others to use (read permission) also exposes model details (system prompt leakage)

What is CVE-2026-45387?

CVE-2026-45387 is a medium-severity vulnerability in Openwebui Open Webui, published on 15 May 2026. Any signed-in user can exploit it over the network, with no action from the victim.

Official description

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.5, when setting model permissions so that a group has read access to it, intending for other users to use it, those users also can read the model's system prompt. However users may consider their system prompt confidential, so this is considered a security issue. This vulnerability is fixed in 0.9.5.

Weakness: CWE-200

Which versions are affected by CVE-2026-45387?

3 open source packages ship an affected version, including open-webui. It affects 1 product.

Products include Openwebui Open Webui.

How do I fix CVE-2026-45387?

Upgrade open-webui to 0.9.5 or later.

Is CVE-2026-45387 being exploited?

It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-45387 exists.

Other vulnerabilities in Openwebui Open Webui

• CVE-2026-44551 Open WebUI: LDAP Empty Password Authentication Bypass Critical 9.1
• CVE-2026-56400 open-webui - Remote Code Execution via CORS Misconfiguration and Session Validation Critical 9.0
• CVE-2024-8017 Cross-site Scripting (XSS) in open-webui/open-webui Critical 9.0
• CVE-2026-45672 Open WebUI: Jupyter code execution works despite `ENABLE_CODE_EXECUTION=false` — feature gate bypassed High 8.8
• CVE-2026-87995 Open WebUI: Same-origin XSS to account takeover via terminal port-preview iframe hardcoding allow-same-origin High 8.7

All Openwebui Open Webui vulnerabilities

Sources

• CVE record
• NVD
• OSV
• GitHub advisory
• How this page is made

Also known as GHSA-h2cw-7qw9-56xr, PYSEC-2026-2728.