CVE-2026-87994: Open WebUI message authorship bypass in channel completions

CVE-2026-87994 · Severity: medium · CVSS 4.3 · Published 2026-09-09

Open WebUI is a user-friendly interface for AI chat that supports collaborative channels where team members can post and edit messages. A vulnerability allows any channel member to overwrite another member's message through the chat completions endpoint, causing the modified text to appear as if written by the original author. This enables fraud, reputation damage, and data integrity loss without leaving a clear audit trail.

Technologies: open-webui (PyPI). Vendors: PyPI.

Quick answers

Is CVE-2026-87994 exploited in the wild?
No. Junglewise Threat Intelligence has no report of CVE-2026-87994 being exploited in the wild.
How severe is CVE-2026-87994?
Medium severity, with a CVSS score of 4.3, published 9 September 2026.
Which products and versions are affected?
open-webui (PyPI) >= 0.9.5, < 0.11.1, ECOSYSTEM:introduced=0.9.5,fixed=0.11.1.
Is there a fix for CVE-2026-87994?
Yes. open-webui (PyPI): upgrade to 0.11.1 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-87994 attacked?
It is attacked over the network.
Am I affected by CVE-2026-87994?
Check whether you run open-webui (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

How to fix

  • open-webui (PyPI): upgrade to 0.11.1 or later
    pip install "open-webui>=0.11.1"

Technical details

The vulnerability is an authorization bypass (CWE-639, CWE-862) in the channel message editing flow. The chat completions endpoint (`POST /api/chat/completions`) checks whether the caller has write access to the channel and whether the target message belongs to that channel, but fails to verify that the caller authored the message being edited. In contrast, the dedicated channel message update route (`POST /api/v1/channels/<channel_id>/messages/<message_id>/update`) correctly enforces authorship checks and rejects cross-member edits with a 403 response. An authenticated channel member with write access can exploit this inconsistency by submitting a message ID belonging to another user to the completions endpoint, causing the target message text to be replaced via AI completion while retaining the original author's identity in the stored record. The vulnerability requires channels to be enabled (disabled by default in `ENABLE_CHANNELS`), attacker membership with write permissions to the target channel, and knowledge of the target message ID (which is visible to all channel members). Fixed in version 0.11.1 via commit 7d392bedc, which adds author identity verification to the channel completions branch matching the dedicated edit route logic.

Affected products

  • Open WebUI open-webui >= 0.9.5, < 0.11.1

Identifiers

  • CVE-2026-87994
  • GHSA-fmqh-xp37-5hr8
  • PYSEC-2026-4122

Timeline

  • 2026-09-10: disclosed
  • 2026-09-10: patched: Fixed in version 0.11.1
  • 2026-08-31: advisory: Published to GitHub Advisory Database

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.