Summary
Wekan versions 8.31.0 through 8.33 expose all global webhook integrations, including sensitive URL and token fields, through the globalwebhooks publication without any server-side authentication.
Any DDP client, including unauthenticated ones, can subscribe to this publication and retrieve webhook configuration data, potentially enabling unauthorized use of those webhooks and access to connected external services.
Affected context
This vulnerability arises in Wekan deployments running versions 8.31.0 through 8.33 that use global webhook integrations.
The issue is specifically in the Meteor globalwebhooks publication, which is meant to power the admin settings UI but lacks server-side access control.
In such setups, any client capable of speaking DDP to the Wekan server can subscribe to this publication, regardless of whether they are logged in or have admin privileges, and receive all configured global webhook URLs and tokens.
Exploit preconditions
- Attack surface: The Meteor DDP API exposed by the Wekan server (typically the same endpoint used by the web UI).
- Required access: Network reachability to the Wekan instance; no authentication or valid user account is required.
- Trigger point: A client issues a subscription request to the
globalwebhookspublication over DDP. - Why exploitation is possible: The server-side publication does not perform authentication or authorization checks before returning the full set of global webhook documents, including sensitive fields, effectively treating all subscribers as trusted administrators.
Why this happens
The root cause is a missing authorization check on a server-side data publication that returns sensitive configuration information.
In Meteor, publications define which data is sent to subscribing clients, and they must explicitly enforce access control. Here, the globalwebhooks publication exposes all global webhook documents—including URL and token fields—without checking if the requester is authenticated or has admin privileges.
This maps to CWE-200: Exposure of Sensitive Information to an Unauthorized Actor, with a related aspect of CWE-306: Missing Authentication for Critical Function, because the publication behaves as a critical administrative data channel that is accessible without authentication.
At an abstract level, the insecure pattern is: a backend data endpoint that returns high-sensitivity fields is implemented without server-side authentication/authorization logic, relying solely on intended client-side usage (admin page) for protection.
Potential impact
- An unauthenticated attacker could retrieve all configured global webhook URLs.
- Attackers may obtain authentication tokens or secrets used by those webhooks.
- Unauthorized invocation or replay of webhooks to external services using stolen URLs and tokens.
- Impersonation of the Wekan instance to third-party services that trust those webhook calls.
- Potential data exfiltration from connected external systems if webhooks include sensitive payloads.
- Abuse of integrations to trigger unintended actions in automation or CI/CD pipelines.
General mitigation guidance
Effective mitigation focuses on enforcing robust server-side access control for any endpoint or publication that returns sensitive data:
- Require authentication for all administrative publications and APIs, ensuring that only logged-in users can access them.
- Add explicit authorization checks that restrict global webhook data to administrator or equivalent privileged roles.
- Separate sensitive fields (such as tokens and secret URLs) into dedicated data structures that are never exposed directly to clients; use server-side invocations instead of client-owned secrets.
- Apply the principle of least privilege in data exposure: only return the minimal fields necessary for UI rendering, masking tokens where possible.
- Regularly review publication and API definitions to ensure they do not depend on client-side routing or UI constraints for security.
- Centralize authorization logic so that sensitive publications must opt in to strict checks rather than rely on ad-hoc conditions.
How teams detect issues like this
Detection can leverage both code-level and runtime approaches:
- Static analysis / code scanning: Search for Meteor
Meteor.publish(or equivalent publication definitions) that return collections containing sensitive fields (e.g.,token,secret,url) without checkingthis.userIdor role-based permissions. - Code review signals: Flag any server-side data endpoints designed for admin UIs that do not include explicit authentication and authorization logic; pay attention to assumptions that "only the admin page calls this."
- Security testing: Use automated or manual DDP clients to enumerate available publications and attempt unauthenticated subscriptions, verifying which data is returned.
- Policy enforcement: Establish guidelines that all publications and APIs must validate authentication/authorization and that sensitive configuration data is never directly sent to clients.
- Logging and monitoring: Monitor access patterns to administrative publications and webhook configuration resources, alerting on unexpected unauthenticated requests or access from unusual IP ranges.

