/

CVE-2026-75031

Severity:

CRITICAL RISK

(9.8)

Code Injection (CWE-94) in Interchange Quick Question

CVE-2026-75031

Awaiting Analysis

Vulnerability type

Category:

Code Injection

CWE:

CWE-94

Impact:

Remote Code Execution

Severity:

CRITICAL RISK

(9.8)

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 critical code-injection vulnerability in Interchange's "quick question" administrative feature allows unauthenticated users to submit arbitrary Perl code for server-side execution. The code is typically evaluated inside a Safe container, but deployments that enable the non-default AllowGlobal directive for an affected catalog may expose broader server capabilities.

Affected context

The vulnerability occurs in Interchange installations that expose the "quick question" administrative feature and process user-controlled input as Perl code. Default deployments may constrain execution through a Safe container, while catalog configurations using the non-default AllowGlobal directive can substantially increase the potential impact.

Exploit preconditions

  • Attack surface: The Interchange "quick question" administrative feature
  • Required access: No authentication may be required when the feature is reachable
  • Trigger point: Server-side processing of attacker-controlled quick-question input
  • Why exploitation is possible: Input is interpreted as Perl code rather than being treated strictly as data; Safe-container restrictions may limit impact unless broader global access is enabled

Why this happens

The underlying flaw is the evaluation of user-controlled input as executable Perl code. This behavior maps to CWE-94: Improper Control of Generation of Code, because data from an external request can influence or become executable program logic. Although a Safe container may restrict available operations, relying on execution isolation does not remove the unsafe code-generation pattern, particularly when broader global access is configured.

Potential impact

  • An attacker may execute Perl code within the Interchange server process
  • The attacker could access or alter application data available to that process
  • The vulnerability may enable compromise of catalog functionality or administrative workflows
  • Deployments with broader global access could experience more extensive host or system impact
  • The application could be used for further compromise of connected services

General mitigation guidance

Mitigation generally focuses on ensuring that user input is never evaluated as executable Perl code and on reducing the privileges available to the application process.

Common defensive patterns include:

  • Replacing dynamic code evaluation with fixed operations selected from an allowlisted set
  • Treating quick-question input strictly as data and applying context-appropriate validation
  • Keeping restrictive execution boundaries enabled and avoiding broad global-access configuration
  • Limiting the application process to only the files, services, and permissions it requires
  • Restricting unauthenticated access to administrative features through authentication and authorization controls
  • Using isolated service accounts and segmented runtime environments where practical

How teams detect issues like this

Detection can combine source review, runtime monitoring, and configuration policy checks.

Useful approaches include:

  • Static analysis for dynamic Perl evaluation, string-to-code conversion, and execution of request-derived values
  • Code review checks for administrative endpoints that pass untrusted input into interpreters
  • Configuration audits identifying enabled global-access directives or weakened Safe-container boundaries
  • Web and application logs reviewed for unusual requests to the quick-question feature and unexpected server-side errors
  • Policy enforcement requiring authentication for administrative functionality and prohibiting unauthenticated code-evaluation paths
  • Automated security review tools that flag user-controlled input reaching dynamic execution APIs

Sources

SourceLink
github.comgithub.com/interchange/interchange/commit/65b6ea9d3761dd1fd2071c962819562afa335e4e
interchangecommerce.orginterchangecommerce.org/i/dev/news?mv_arg=00071

Frequently asked

questions

Yes. Where the quick-question feature is reachable without authentication, an attacker may be able to submit crafted input remotely.

No. The feature may still evaluate attacker-controlled Perl code within the Safe container by default. Enabling AllowGlobal can remove important restrictions and increase the potential impact.

Ensure external input is handled only as data, eliminate dynamic evaluation of request values, and keep restrictive execution and authorization boundaries enabled.

Use allowlisted operations instead of generated code, require authentication for administrative features, audit interpreter boundaries, and enforce least-privilege runtime configurations.

Find out what's already
exploitable in your codebase.