/

CVE-2026-72882

Severity:

CRITICAL RISK

(9.9)

OS Command Injection (CWE-78) in Dokploy File Mounts

CVE-2026-72882

Received

Vulnerability type

Category:

OS Command Injection

CWE:

CWE-78

Impact:

Remote 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

Dokploy contains an OS command injection vulnerability in file mount handling, where an authenticated user with permission to create or update service mounts can supply shell metacharacters through the filePath value.

When the affected configuration is processed over SSH, attacker-controlled commands may execute on the configured remote managed server. In the default deployment model, this can result in direct remote code execution from the web interface.

Affected context

The issue affects Dokploy deployments using file mounts for services, particularly where:

  • An authenticated account can create or modify service file mounts
  • The filePath value is incorporated into a command or remote operation
  • The operation is performed on a managed server over SSH
  • Input is not safely separated from shell command construction

Risk depends on the permissions granted to the authenticated user and the privileges available to the remote managed server.

Exploit preconditions

  • Attack surface: Dokploy's authenticated web interface for service file mounts
  • Required access: An authenticated account authorized to create or update file mounts
  • Trigger point: Processing or applying a service mount configuration on the remote managed server
  • Why exploitation is possible: The filePath value can influence shell command interpretation when passed through an SSH-backed command execution path

Why this happens

The vulnerability is caused by incorporating attacker-influenced file path data into shell commands without reliably preventing shell interpretation.

When command construction treats a path as executable command text rather than as data, shell metacharacters can alter the intended command behavior. This maps to CWE-78: Improper Neutralization of Special Elements used in an OS Command, commonly known as OS command injection.

Potential impact

  • Attacker-controlled commands may execute on the configured remote managed server
  • The affected host could be compromised with the privileges of the SSH-connected account
  • Application data, deployment secrets, or service credentials may be exposed
  • An attacker could alter deployed services or disrupt hosted workloads
  • Additional systems may be affected through lateral movement or stolen credentials

General mitigation guidance

Mitigation generally focuses on ensuring that file paths remain data throughout the full execution path and that shell interpretation is not required.

Common defensive patterns include:

  • Prefer direct process execution APIs that pass arguments separately rather than invoking a shell
  • Apply strict, context-appropriate path validation and canonicalization
  • Use allowlisted mount locations and reject unexpected path formats
  • Avoid constructing remote shell commands from user-controlled values
  • Reduce SSH account and remote service privileges to the minimum required
  • Restrict which authenticated roles can create or modify service mounts
  • Use isolated deployment identities and narrowly scoped credentials
  • Treat web-interface configuration values as untrusted even when submitted by authenticated users

How teams detect issues like this

Organizations can look for this class of issue through:

  • Static analysis rules identifying shell execution APIs that receive concatenated or interpolated user input
  • Code review checks for remote command construction involving file paths or mount parameters
  • Tests using metacharacter-oriented inputs to verify that values remain inert data
  • Runtime monitoring of unusual commands executed by Dokploy's SSH identities
  • Audit logging for service mount creation and modification events
  • Policy enforcement that prohibits shell-based remote execution where direct argument passing is available

Automated security review tools may help identify command construction and insufficient input-neutralization patterns.

Sources

SourceLink
github.comgithub.com/Dokploy/dokploy/security/advisories/GHSA-qcf5-jjjp-7794

Frequently asked

questions

Yes, within the application's access model. An authenticated user who is authorized to create or update file mounts may be able to trigger execution on the configured remote managed server through the Dokploy web interface.

Direct shell access to the managed server is not necessarily required. The relevant prerequisite is authenticated Dokploy access with permission to modify the affected service mount configuration.

Treat file paths as data, avoid shell interpretation, and use direct process or SSH APIs that pass arguments separately. Strict path allowlisting and least-privilege remote accounts provide additional protection.

Enforce role-based authorization for mount changes, validate and canonicalize path inputs, review all remote command construction, and monitor commands issued by deployment identities.

Find out what's already
exploitable in your codebase.