Authenticated system-prompt disclosure via model-detail endpoint
Published May 15, 2026 · Updated May 15, 2026
Information disclosure in Open WebUI before 0.9.5 allows remote authenticated users with model read access to retrieve its system prompt. The /api/v1/models/model endpoint returns params.system for a supplied model identifier even when that model is absent from the caller's editable-model list. Group-level read permission and knowledge of the model identifier are required; saving changes is blocked, so the demonstrated impact is limited to prompt confidentiality.
Summary
What happened
Information disclosure in Open WebUI before 0.9.5 allows remote authenticated users with model read access to retrieve its system prompt. The /api/v1/models/model endpoint returns params.system for a supplied model identifier even when that model is absent from the caller's editable-model list. Group-level read permission and knowledge of the model identifier are required; saving changes is blocked, so the demonstrated impact is limited to prompt confidentiality.
The record
- CVE
- CVE-2026-45387
- Published
- May 15, 2026
- Updated
- May 15, 2026
- Vendor
- Open WebUI
- Product
- Open WebUI
- Classifications
- CWE-200, T1213
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- May 15, 2026CVE publishedPublication date reported by the CVE source.
- May 15, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 0.9.5
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- GHSA model-detail endpoint proof of conceptproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo