Is my AI tool safe?Assistants and personal agentsOpen WebUI
Is Open WebUI safe?
No tool is safe or unsafe on its own: it depends on what it can reach and how it is set up. The record for Open WebUI: 0 incidents in this center's library and 1 CVE in its advisory feed (1 high), the newest published Nov 7, 2025. Below: what went wrong, what was fixed, and how to use it safely.
What to watch for
The default front end for self-hosted models, usually deployed with sign-up open and the admin account holding every provider key. The feed carries a high-severity flaw; the bigger risk is the exposure pattern.
How to use it safely
Close sign-up, put it behind SSO or a reverse proxy with authentication, and keep provider keys in a secret store rather than the admin UI.
What it can reach
- Often exposed to the internet
- Holds API keys and tokens
- Installs plugins, skills or tools from registries
Known CVEs
- CVE-2025-64496 High Open WebUI code injection from an external model server (Direct Connections)
How many people use it
- 153k GitHub stars · GitHub
- 1.4M PyPI downloads in a month (to Oct 1, 2026) · PyPI Stats
A high CVE count can mean heavy scrutiny as much as weak engineering, so read the incidents and advisories, not only the number. The counts are what this center has recorded, not every flaw ever found; usage figures are the vendors' and registries' own published numbers.