SPIP odt2spip os command injection via filenames patched in 4.4.0

An attacker with a standard author account on a SPIP site can gain remote code execution on the web server by exploiting an OS command injection flaw in the odt2spip plugin. According to VulnCheck, this vulnerability, tracked as CVE-2026-104085, allows any user with the basic redacteur role to run arbitrary system commands. The exploit requires no administrative privileges or further interaction beyond uploading a single document with a maliciously crafted filename, effectively escalating a low-privilege account to full server control.

How it unfolded

• VulnCheck’s Initial Access Intelligence team identified the OS command injection vulnerability in the SPIP odt2spip plugin.
• The National Vulnerability Database assigned CVE-2026-104085 a CVSS score of 8.7, classifying the severity as high.
• The plugin vendor released version 4.4.0 on October 7, 2026, which includes the necessary fix for the flaw.

Who is affected

The odt2spip plugin for the SPIP content management system is affected up to and including version 4.3.0. The vulnerability impacts any multi-author SPIP site where the plugin is installed and configured to use LibreOffice for document conversion. According to the source material, the attack surface includes any user holding the standard redacteur role, which is the ordinary author permission level in SPIP. This role allows users to create or modify articles, granting them access to the document import form where the vulnerability resides.

The fix

The odt2spip plugin version 4.4.0 resolves the OS command injection vulnerability. Released on October 7, 2026, this update addresses the insufficient sanitization of filenames during the file import process. According to VulnCheck, the fix has been verified in the source code. Sites running versions prior to 4.4.0 remain exposed to the risk of authenticated remote code execution via the document upload feature.

What made it possible

This incident illustrates the danger of passing user-supplied input directly into system shell commands. When applications construct command strings using filenames or other client data, they must rigorously sanitize or encode that input. Teams usually validate that no shell metacharacters can break out of the intended argument context. Check whether your systems use external binaries like LibreOffice for processing and ensure that file paths are never concatenated into executable commands without strict escaping. This weakness often serves as an initial access vector, turning low-privilege actions into full system compromise.

What to do and how to stay safe: SPIP

• Verify if your SPIP installations use the odt2spip plugin and check the current version against the affected range.
• Review server logs for unusual command execution or outbound connections originating from the web server user account.
• Restrict file upload permissions to trusted administrators only until all managed sites are updated to the patched version.
• Monitor for any new disclosures related to SPIP plugins, as similar sanitization flaws may exist in other components.

Step-by-step guide: What Is Virtual Patching: How It Works Without Code Changes

General security guidance from the RedTeam Tribune newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

What privilege level is required to exploit CVE-2026-104085?

An attacker needs a low-privileged authenticated account, specifically the standard redacteur role in SPIP.

Is there a patch available for this vulnerability?

Yes, the fix is included in odt2spip plugin version 4.4.0, released on October 7, 2026.

What is the CVSS score for this vulnerability?

The National Vulnerability Database rates CVE-2026-104085 as 8.7 high severity.