/

CVE-2026-72548

Severity:

HIGH RISK

(7.5)

Information Disclosure (CWE-200) in OpenSign Tenants

CVE-2026-72548

Deferred

Vulnerability type

Category:

Information Disclosure

CWE:

CWE-200

Impact:

Unauthorized disclosure of tenant configuration and organisational data

Severity:

HIGH RISK

(7.5)

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

OpenSign through version 2.37.0 exposes organisation tenant records through the gettenant Parse cloud function without requiring authentication or authorization. An unauthenticated remote attacker who can submit a valid or discoverable contactId may retrieve configuration and other tenant data for organisations hosted by the system.

Affected context

The vulnerability occurs in OpenSign deployments where the gettenant Parse cloud function is reachable by unauthenticated clients and returns tenant records based on a supplied contactId. Risk is higher when contact identifiers can be guessed, enumerated, or obtained through other exposed application functionality, and when the returned record contains sensitive configuration or organisational information.

Exploit preconditions

  • Attack surface: The externally reachable gettenant Parse cloud function
  • Required access: No authenticated account is required
  • Trigger point: A request containing a contactId parameter
  • Why exploitation is possible: The function returns a tenant record without verifying the caller's identity or authorization to access that organisation

Why this happens

The underlying bug is an access-control failure combined with excessive data exposure. The cloud function treats a client-supplied tenant reference as sufficient authority to retrieve a complete tenant record, without enforcing authentication, object-level authorization, or response filtering. This behavior maps to CWE-200: Exposure of Sensitive Information to an Unauthorized Actor, because data is disclosed to callers who have not been established as entitled to receive it.

Potential impact

  • An attacker may enumerate organisation tenants
  • Tenant configuration data could be disclosed without authorization
  • Exposed records may support targeted attacks against organisations or administrators
  • Sensitive operational details could be used for reconnaissance
  • Privacy, compliance, and trust obligations may be affected

General mitigation guidance

Mitigation generally focuses on enforcing authorization at the cloud-function boundary and minimizing the data returned to clients.

Common defensive patterns include:

  • Require authentication before processing tenant lookup requests
  • Enforce object-level authorization for the requested organisation and record
  • Avoid trusting client-supplied identifiers as proof of access
  • Return only fields required for the caller's documented use case
  • Use opaque, non-sequential identifiers where enumeration resistance is appropriate
  • Apply rate limits, monitoring, and consistent error handling to lookup endpoints
  • Run periodic reviews of Parse cloud functions and tenant data access policies

How teams detect issues like this

Organizations can assess this class of issue through:

  • Static analysis and code review for cloud functions that return records based only on request parameters
  • Review signals such as missing authentication checks, absent object-level authorization, and unrestricted database queries
  • API tests that call tenant-related functions without credentials and compare responses across tenant identifiers
  • Data-minimization checks verifying that responses exclude fields unnecessary for the client
  • Policy enforcement requiring authorization tests for every tenant-scoped endpoint
  • Monitoring for repeated lookup requests, identifier enumeration patterns, and unusually broad tenant access
  • Neutral automated application-security review tools that identify missing access-control checks and excessive data exposure

Sources

SourceLink
github.comgithub.com/OpenSignLabs/OpenSign

Frequently asked

questions

Yes. The affected cloud function may be reachable remotely, and the described behavior does not require an authenticated session.

No account is generally required. The attacker needs network access to the function and a usable or discoverable contactId value.

Authenticate callers, enforce object-level authorization for the requested tenant, and return only the minimum necessary fields.

Treat every client-supplied identifier as untrusted, require authorization tests for tenant-scoped APIs, and continuously test for enumeration and cross-tenant data access.

Find out what's already
exploitable in your codebase.