/

CVE-2026-76836

Severity:

HIGH RISK

(8.8)

Improper Code Generation (CWE-94) in AzuraCast Config

CVE-2026-76836

Deferred

Vulnerability type

Category:

Improper Code Generation

CWE:

CWE-94, CWE-863

Impact:

Operating-System Command Execution

Severity:

HIGH RISK

(8.8)

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

AzuraCast exposes Liquidsoap custom configuration fields through a profile-editing endpoint that may not enforce the permission intended to protect them. A station manager with profile access could potentially cause operating-system commands to run when generated Liquidsoap configuration is processed and the backend restarts the station.

Affected context

The issue occurs in AzuraCast station profile management when backend configuration fields are included in a broadly authorized deserialization group. Those values may then be written verbatim into a generated Liquidsoap script, while the dedicated Liquidsoap configuration endpoint applies a stronger broadcasting permission boundary. Risk depends on the attacker's station permissions, configuration mutability, and whether the affected backend restart and synchronization behavior are active.

Exploit preconditions

  • Attack surface: The station profile-edit API endpoint and its configuration deserialization path
  • Required access: A user able to edit a station profile, potentially with profile permission but without broadcasting permission
  • Trigger point: Processing generated Liquidsoap configuration during a backend restart or synchronization task
  • Why exploitation is possible: Public configuration properties are assigned without field-level authorization, and selected values are emitted into an executable Liquidsoap script without sufficiently reducing their input trust

Why this happens

The vulnerability combines an authorization boundary failure with unsafe code generation. A deserialization group exposes backend configuration fields through an endpoint guarded only by profile permissions, while generic property assignment does not apply separate permission checks to each field. The configuration writer then places attacker-influenced values into a Liquidsoap script, where command-execution functionality may interpret them as executable logic. This behavior maps primarily to CWE-94: Improper Control of Generation of Code, with CWE-863: Incorrect Authorization describing the missing field-level permission enforcement.

Potential impact

  • An authorized station manager could cause operating-system commands to execute in the backend environment
  • The station or associated application process could be compromised
  • Sensitive configuration data may be accessed or altered
  • Other services or hosts could be affected if the backend has excessive privileges or network access
  • Broadcast content, station data, or operational workflows could be disrupted

General mitigation guidance

Mitigation generally focuses on preserving the intended authorization boundary and preventing user-controlled values from becoming executable configuration.

Common defensive patterns include:

  • Applying field-level authorization checks during deserialization and property assignment
  • Excluding privileged Liquidsoap configuration fields from general profile-edit operations
  • Requiring the dedicated broadcasting permission for all custom configuration changes
  • Treating generated script content as a code boundary and using structured, allowlisted configuration APIs where possible
  • Validating configuration values against strict schemas and rejecting executable constructs when they are not required
  • Running backend and Liquidsoap processes with least privilege and restricted filesystem and network access
  • Separating station-management permissions from configuration and process-control permissions

How teams detect issues like this

Detection can combine authorization review, data-flow analysis, and runtime monitoring.

Useful approaches include:

  • Static analysis for generic property assignment from request data into privileged entities
  • Searches for deserialization groups that expose sensitive configuration fields
  • Code review checks ensuring endpoint permissions match field sensitivity
  • Taint analysis tracking request-controlled values into generated scripts or command-execution APIs
  • Policy checks requiring allowlisted configuration schemas for script-generating components
  • Audit logging and alerting for profile updates that modify backend or Liquidsoap configuration fields
  • Monitoring unexpected process creation, configuration rewrites, and backend restarts associated with station changes

Sources

SourceLink
github.comgithub.com/AzuraCast/AzuraCast
github.comgithub.com/AzuraCast/AzuraCast/blob/0.23.8/backend/src/Entity/Station.php
github.comgithub.com/AzuraCast/AzuraCast/blob/0.23.8/backend/src/Radio/Backend/Liquidsoap/ConfigWriter.php
github.comgithub.com/AzuraCast/AzuraCast/security/advisories/GHSA-q8wg-3qg7-8pc7
vulncheck.comvulncheck.com/advisories/azuracast-through-liquidsoap-configuration-write-via-profile-edit-serialization-group-bypass

Frequently asked

questions

It may be exploitable through the authenticated station-management API when an account has profile-edit permission and the affected configuration fields are exposed to that permission scope.

Not necessarily. The described authorization gap allows a user with the narrower profile permission to reach fields intended to require broadcasting-level access.

Enforce authorization at the individual field and operation levels, keep privileged configuration out of general profile updates, and ensure user-controlled values cannot become executable script content.

Use least-privilege permission models, strict schemas and allowlists, security-focused review of serialization groups, and monitoring for configuration changes that trigger process restarts or unexpected command execution.

Find out what's already
exploitable in your codebase.