CVE-2026-45387

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

Published May 15, 2026 CVSS 3.1 · 4.3 Medium EPSS < 1% KEV · Not listed
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.

Analysis

AI-generated

Technical impact

Open WebUI versions prior to 0.9.5 can allow users with read access to a shared model to view the model's system prompt. This unintentionally exposes text that users may consider confidential, leading to an information disclosure vulnerability identified as CWE‑200. No exploitation of code or denial of service is possible; the impact is purely data leakage.

Affected Systems

The product affected is Open WebUI from the vendor open‑webui. All releases before 0.9.5 are vulnerable; the fix is included in version 0.9.5 and later.

Risk and Exploitability

The CVSS score of 4.3 indicates a medium severity level. Exploitation requires that an attacker already has read permission to a shared model, so the attack vector is local to the platform's permission model. EPSS information is unavailable, and the vulnerability is not listed in CISA’s KEV catalog, suggesting that widespread exploitation has not been reported. Overall, the risk is moderate, primarily due to potential disclosure of sensitive prompt data.

Remediation

No solution or workaround provided in the CVE record.

OpenCVE Recommended Actions

  • Upgrade Open WebUI to version 0.9.5 or later, which contains the fix for the system prompt leakage.
  • If an upgrade cannot be performed immediately, restrict read permissions to only trusted or internal users and avoid sharing models that contain sensitive system prompts.
  • Validate that the system configuration does not expose the system prompt field to users with read access, and monitor forthcoming releases for additional security updates.

Generated by OpenCVE AI on May 15, 2026 at 21:21 UTC.

Tracking

Sign in to view the affected projects.

Advisories

Source ID Title
Github GHSA Github GHSA GHSA-h2cw-7qw9-56xr Open WebUI: Sharing models for others to use (read permission) also exposes model details (system prompt leakage)
History
Type Values Removed Values Added
First Time appeared Openwebui
Openwebui open Webui
CPEs cpe:2.3:a:openwebui:open_webui:*:*:*:*:*:*:*:*
Vendors & Products Openwebui
Openwebui open Webui
Type Values Removed Values Added
First Time appeared Open-webui
Open-webui open-webui
Vendors & Products Open-webui
Open-webui open-webui
Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'partial'}, 'version': '2.0.3'}

Type Values Removed Values Added
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.
Title Open WebUI: Sharing models for others to use (read permission) also exposes model details (system prompt leakage)
Weaknesses CWE-200
References
Metrics cvssV3_1

{'score': 4.3, 'vector': 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:N'}

Weaknesses
  • CWE-200

    Exposure of Sensitive Information to an Unauthorized Actor

Sources

Status
PUBLISHED
Assigner
GitHub_M
Published
Updated
May 15, 2026, 9:08 PM
Reserved
May 12, 2026, 12:51 AM
Status
Analyzed
Published
May 15, 2026, 9:16 PM
Modified
Nov 11, 2026, 8:10 PM