/

CVE-2026-72563

Severity:

HIGH RISK

(8.1)

Broken Access Control (CWE-284) in Handesk Lead Records

CVE-2026-72563

Deferred

Vulnerability type

Category:

Broken Access Control

CWE:

CWE-284

Impact:

Unauthorized modification of cross-team lead data

Severity:

HIGH RISK

(8.1)

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

A broken access control vulnerability in BadChoice Handesk allows an authenticated agent to update lead records owned by other teams. The risk arises when the agent can reach the LeadsController@update endpoint and the application fails to enforce team-level authorization.

Affected context

The vulnerability occurs in Handesk deployments where authenticated agents can access the lead update workflow and authorization is not enforced before records are changed. The risk is amplified when the Lead model permits mass assignment of all columns, allowing submitted attributes to be applied without sufficiently restricting the update scope.

Exploit preconditions

  • Attack surface: The authenticated lead-management web endpoint
  • Required access: Any valid Handesk agent account
  • Trigger point: Processing an update through LeadsController@update
  • Why exploitation is possible: The endpoint does not adequately verify team ownership or record-level permissions, while the Lead model permits unrestricted mass assignment

Why this happens

The underlying bug is a missing authorization check at a cross-team resource boundary. The update operation relies on authentication without confirming that the requesting agent may modify the specific lead, which maps to CWE-284: Improper Access Control. An unrestricted guarded configuration further broadens the update surface by allowing request-controlled attributes to be assigned to the model.

Potential impact

  • Attackers may overwrite lead records belonging to other teams
  • Business data integrity may be damaged or made unreliable
  • Unauthorized changes may disrupt sales and support workflows
  • Sensitive lead attributes could be altered or exposed through workflow manipulation
  • Audit trails and downstream processes may become difficult to trust

General mitigation guidance

Mitigation generally focuses on enforcing authorization at the resource and team boundaries before any update is performed.

Common defensive patterns include:

  • Applying server-side policy checks that verify record ownership or permitted team scope
  • Treating authentication and authorization as separate controls
  • Replacing unrestricted mass assignment with explicit allowlists of updateable attributes
  • Rejecting client-controlled ownership, team, and privilege-related fields
  • Using least-privilege roles and narrowly scoped agent permissions
  • Preserving consistent authorization checks across controllers, APIs, background jobs, and bulk-update paths

How teams detect issues like this

Organizations can identify similar weaknesses through:

  • Static analysis and policy checks for missing authorization calls in update and delete handlers
  • Code review signals such as controller actions that load records without ownership or policy validation
  • Searches for empty or overly broad mass-assignment protection
  • Tests that attempt cross-team access with authenticated users from different teams
  • Runtime monitoring for unusual cross-team update patterns and unexpected changes to ownership-related fields
  • Automated security review tools that flag authorization gaps and unrestricted model assignment

Sources

SourceLink
github.comgithub.com/BadChoice/handesk

Frequently asked

questions

Yes, where the Handesk application is reachable and an attacker possesses a valid agent account with access to the lead-management workflow.

Generally no. The described issue requires authentication as an agent, but any agent account may be sufficient if team-level authorization is absent.

Enforce server-side, record-level authorization before updates and restrict model assignment to an explicit list of permitted fields.

Combine centralized authorization policies, least-privilege roles, strict mass-assignment controls, and automated cross-tenant access tests.

Find out what's already
exploitable in your codebase.