UTC
High

Any authenticated user can steal another user's session token via script embedded in a DOCX preview

CVE-2026-109290 · GHSA-f9xp-mfmq-x6cg · published 2026-09-27 · updated 2026-10-28

Severity
High
CVSS score
8.7
Chance of exploitation (EPSS)
Not scored
Fixed in
0.11.4
  • Exploitation Public proof of concept
  • Automatable No
  • Technical impact Total
  • CISA SSVC
What it is

The embedded script executes in the Open WebUI origin in the viewer's authenticated session and can read the session token. A stolen administrator token replays directly against the API without a password, giving the attacker the full administrator capability set (enumerating all users, changing roles, reading all chats and documents, creating API keys, changing instance settings). A low-privileged attacker thereby escalates to full instance control through one previewed document. Instances where DOCX preview is never used by higher-privileged users are not affected in practice.

What to do

Upgrade open-webui to 0.11.4 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.

Fixed in v0.11.4 by #29699. The DOCX preview no longer renders embedded HTML parts (the renderer is told not to inline document-supplied HTML sub-documents), and after rendering any link target that does not resolve to http, https, mailto or tel is removed. Upgrading to v0.11.4 or later fully resolves the issue with no further configuration.

Affected
PackageRegistryAffectedFixed in
open-webuiPyPI>= 0.11.1, <= 0.11.30.11.4

The impact, patch and workaround text comes from the advisory, shortened (published by Open WebUI). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].