/

CVE-2026-59714

Severity:

HIGH RISK

(7.1)

Authorization Bypass (CWE-862) in Open WebUI Channels

CVE-2026-59714

Deferred

Vulnerability type

Category:

Missing Authorization

CWE:

CWE-862

Impact:

Unauthorized Message Modification and Data Integrity Loss

Severity:

HIGH RISK

(7.1)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Open WebUI contains an authorization enforcement weakness that allows any authenticated user to overwrite messages in channels they do not belong to, including private and direct-message channels. The issue becomes risky when a chat completion request can reference a channel and message identifier without verifying that the caller has access to both resources.

Affected context

The vulnerability affects Open WebUI deployments using channel chat completion functionality in versions from 0.9.5 before 0.10.0. It occurs when pipeline output is processed through the channel emitter and the message update operation accepts a caller-supplied message identifier without confirming that the identified message belongs to an authorized channel.

Exploit preconditions

  • Attack surface: Authenticated chat completion and channel messaging functionality
  • Required access: A valid Open WebUI user account
  • Target condition: A message exists in a channel, private channel, or direct-message conversation the user is not authorized to access
  • Trigger point: Processing channel-prefixed completion requests that reference an existing message
  • Why exploitation is possible: Authorization is not adequately bound to the requested message, channel, and caller before the database update occurs

Why this happens

The underlying issue is an object-level authorization failure mapped to CWE-862: Missing Authorization. The channel output path uses a caller-controlled message identifier when writing to the Messages table, but does not enforce a relationship between that message, the requested channel, and the authenticated user's membership or permissions. This creates an insecure direct object reference pattern in which resource ownership and access checks are separated from the update operation.

Potential impact

  • An authenticated user may overwrite messages outside their authorized channels
  • Private or direct-message content may be altered
  • Conversation integrity and auditability could be reduced
  • Users may be misled by unauthorized changes to channel history
  • Modified messages can cause reputational, operational, or moderation impacts

General mitigation guidance

Mitigation generally focuses on enforcing authorization at the resource-operation boundary rather than trusting identifiers supplied by clients.

Common defensive patterns include:

  • Verify channel membership and permissions before every message read or write
  • Bind message ownership and channel relationships in the authorization query or transaction
  • Treat message and channel identifiers as references only, not as proof of access
  • Apply centralized, deny-by-default authorization checks consistently across API and pipeline paths
  • Use least-privilege service permissions for database operations
  • Record and review authorization failures and sensitive message updates

How teams detect issues like this

Organizations can assess for this class of issue through:

  • Static analysis searches for database updates using client-supplied object identifiers without nearby authorization checks
  • Code review of channel, message, and direct-message access-control paths
  • Tests that attempt cross-channel reads and writes using authenticated accounts with different memberships
  • Policy checks requiring authorization decisions to include both the caller and target resource
  • Audit monitoring for message updates where the actor is not a member of the associated channel
  • Automated application-security review tools that identify insecure direct object reference patterns

Sources

SourceLink
github.comgithub.com/open-webui/open-webui/commit/ac3449cac91e62b08a7c28e54fcd044d14dea791
github.comgithub.com/open-webui/open-webui/pull/26385
github.comgithub.com/open-webui/open-webui/releases/tag/v0.10.0
github.comgithub.com/open-webui/open-webui/security/advisories/GHSA-x2ff-v5v8-m75m

Frequently asked

questions

An authenticated Open WebUI user may be able to exploit it when the deployment exposes the affected channel completion path and the user can reference a message outside their authorized channels.

Yes. The described authorization gap can affect private channels and direct-message conversations when the message identifier is not properly bound to an authorized channel or conversation.

Enforce server-side, object-level authorization before every message update, checking the caller's permissions, the target channel, and the relationship between the message and channel within the same protected operation.

Use centralized deny-by-default authorization, avoid trusting client-supplied identifiers, add cross-tenant and cross-channel authorization tests, and monitor sensitive resource modifications.

Find out what's already
exploitable in your codebase.