/

CVE-2026-93985

Severity:

CRITICAL RISK

(9.9)

Code Injection (CWE-94) in OpenPanel Webhooks Validator

CVE-2026-93985

Deferred

Vulnerability type

Category:

Code Injection

CWE:

CWE-94

Impact:

Sandbox Escape and Arbitrary Code Execution

Severity:

CRITICAL RISK

(9.9)

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

OpenPanel js-runtime contains a sandbox escape in its JavaScript webhook template validator because computed member access can reach constructor chains that should be blocked. A project user with write access may create a crafted webhook template that executes arbitrary code in the worker process.

Affected context

The vulnerability occurs in OpenPanel's JavaScript webhook template validation and execution workflow when user-controlled templates are evaluated in a constrained runtime. Deployments are at risk when project members with write access can define or modify webhook templates and the validator does not comprehensively restrict computed property access or constructor traversal.

Exploit preconditions

  • Attack surface: JavaScript webhook template creation and validation
  • Required access: Project write access sufficient to create or modify webhook templates
  • Trigger point: Processing or evaluating a crafted webhook template in the worker process
  • Why exploitation is possible: The sandbox permits computed member access that can traverse constructor chains and reach dynamic code execution capabilities

Why this happens

The underlying bug is incomplete sandbox enforcement in a JavaScript template validator. Although direct access to dangerous objects or properties may be restricted, equivalent access through computed members and constructor relationships remains available. This allows untrusted template expressions to escape intended object boundaries and invoke dynamic code-generation behavior, mapping to CWE-94: Improper Control of Generation of Code (Code Injection).

Potential impact

  • An attacker could execute arbitrary code in the worker process
  • The application runtime could be compromised
  • Webhook credentials, environment variables, or accessible project data may be exposed
  • The attacker could alter jobs, templates, or downstream processing
  • Further compromise could occur if the worker has excessive permissions or network access

General mitigation guidance

Mitigation generally focuses on reducing the trust placed in user-authored templates and enforcing a complete, deny-by-default execution boundary.

Common defensive patterns include:

  • Prefer declarative, non-evaluated webhook templates over executable JavaScript
  • Use a purpose-built expression language with an allowlisted grammar and object model
  • Block constructor traversal, computed access to sensitive members, dynamic code generation, and reflective APIs
  • Validate and normalize template syntax before execution, while avoiding validation as the sole security boundary
  • Run worker processes with least privilege, isolated credentials, restricted filesystem access, and limited network reachability
  • Separate template validation from privileged application operations
  • Treat project write access as a security-sensitive capability and limit it according to need

How teams detect issues like this

Detection can combine static analysis, runtime policy enforcement, and focused review of sandbox boundaries.

Useful approaches include:

  • Searching for template evaluators, dynamic JavaScript execution, constructor access, computed member resolution, and reflective APIs
  • Reviewing denylist-based sandbox logic for alternate property-access paths
  • Testing policy enforcement against equivalent direct and computed access patterns
  • Monitoring worker processes for unexpected child processes, filesystem activity, network connections, or environment access
  • Enforcing code-review and CI policies that prohibit dynamic evaluation unless an approved isolated runtime is used
  • Using automated security review tools to identify incomplete JavaScript sandbox restrictions

Sources

SourceLink
github.comgithub.com/Openpanel-dev/openpanel/security/advisories/GHSA-6f7h-cvp6-w9w5
vulncheck.comvulncheck.com/advisories/openpanel-js-runtime-javascript-template-sandbox-escape-rce

Frequently asked

questions

It may be exploitable through the application's project-management interface when a user can create or modify webhook templates. The practical exposure depends on how project write access is granted and where workers are reachable.

The described attack requires project write access or an equivalent permission to submit webhook templates. It does not necessarily require host-level access to the worker.

Replace executable JavaScript templates with a declarative, allowlisted template model, or enforce a genuinely isolated runtime that blocks constructor traversal, reflective access, and dynamic code generation.

Treat sandboxing as a defense-in-depth boundary, test alternate access paths during security review, minimize worker privileges, and restrict template-authoring permissions.

Find out what's already
exploitable in your codebase.