Roadiz has Server-Side Request Forgery (SSRF) in roadiz/documents

Description

This vulnerability allows an authenticated attacker to read any file on the server's local file system that the web server process has access to, including highly sensitive environment variables, database credentials, and internal configuration files.

Field Details
Vulnerability Class Server-Side Request Forgery (SSRF) & Local File Inclusion (LFI)
Affected Component RZ\Roadiz\Documents\DownloadedFile::fromUrl()
Prerequisites Authenticated user with ROLE_ACCESS_DOCUMENTS

Technical Description

The Roadiz backend features tools for importing external media, such as compiling cover art from Podcast RSS Feeds or OEmbed providers. This feature is handled by various MediaFinders, which ultimately pass the extracted media URLs to the DownloadedFile::fromUrl(string $url) parsing mechanism.

Inside fromUrl(), the application uses PHP's native fopen() function to fetch the remote resource and copy it into the local temporary directory before injecting it into the Flysystem Documents storage.

The Flaw

The $url parameter is passed to fopen without any schema validation or sanitization. In PHP, when stream wrappers are enabled, functions like fopen do not restrict operations to HTTP streams. If a file:// scheme is supplied, PHP seamlessly converts the operation into a local file system read. Because an attacker tightly controls the XML feed (e.g., from a Podcast integration), they can inject a file:// URI, forcing the CMS to "download" internal system files directly into the publicly accessible Media Library.


Proof of Concept (PoC)

To reliably reproduce this vulnerability without requiring a live external URL, the attacker simply mimics the behavior of the Podcast importer manipulating the internal system.

Step 1: Craft the Malicious Payload

The attacker creates a standard Podcast RSS XML feed (podcast.xml) and hosts it externally (or on an internal network reachable by the CMS). Inside this XML, the href attribute for the podcast thumbnail is weaponized to target a sensitive system file:

<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd">
  <channel>
    <title>Roadiz LFI Exploit</title>
    <!-- Payload triggers the local filesystem fetch via PHP streams -->
    <itunes:image href="file:///app/.env" /> 
  </channel>
</rss>

Step 2: Exploit the CMS

  1. Authenticate to the Roadiz Backoffice.
  2. Navigate to Documents (Media Manager).
  3. Select Add a document -> Import from URL (or trigger a Podcast sync).
  4. Supply the URL of the malicious podcast.xml file.

Step 3: Extract the Data

  1. The AbstractPodcastFinder processes the XML and feeds file:///app/.env directly into DownloadedFile::fromUrl().
  2. The Roadiz application silently reads its own .env file, creating a new "Document" arrayed with the contents of the file.
  3. The file manifests in the Media Manager grid as a broken image icon.
  4. The attacker actively downloads the newly generated Document from the dashboard, successfully extracting the framework's internal API keys, database credentials, and APP_SECRET.

Impact Analysis

Exploitation of this vulnerability results in a total loss of Confidentiality for the web application and underlying operating system.

  • Application Compromise: An attacker can retrieve .env, security.yaml, or database .sqlite files, leading to complete horizontal and vertical privilege escalation.
  • System Enumeration: The attacker can read /etc/passwd, enumerating system users in preparation for lateral movement.
  • Cloud Environment Compromise: If deployed within AWS, Azure, or GCP, the SSRF vector can be pivoted to read internal cloud metadata endpoints (e.g., http://169.254.169.254/latest/meta-data/), allowing the attacker to steal Root IAM roles globally compromising the victim's infrastructure.

Basic information

Type
reviewed
Severity
medium
Advisory on GitHub
Open advisory ↗
Repository advisory
Open repository advisory ↗
Source code
Browse source ↗
Published (advisory)
2026-03-23 21:43:14 UTC
Updated
2026-03-27 21:00:11 UTC
GitHub reviewed
2026-03-23 21:43:14 UTC
NVD published
2026-03-26

EPSS Score

Score Percentile
0.03% 8.64%

CVSS Scores

Base score Version Severity Vector
6.8 3.1
CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:C/C:H/I:N/A:N Click to expand
Attack vector (AV:N)
Could be attacked over the internet or any normal routed network—not just someone sitting at the machine.
Attack complexity (AC:L)
Once they can reach the bug, pulling it off is straightforward—no weird race conditions or rare setup.
Privileges required (PR:H)
They need powerful rights—admin, root, or similar—before this pays off.
User interaction (UI:N)
Nobody has to click “OK” or open a trap file; it can work without a victim helping.
Scope (S:C)
Breaking this can reach past the original component and bite other resources—bigger blast radius.
Confidentiality (C:H)
Serious risk that confidential data gets exposed in a big way.
Integrity (I:N)
Data isn’t meaningfully altered or forged.
Availability (A:N)
Service keeps running; no real outage angle.

Identifiers

CWEs

CWE id Name
CWE-918 Server-Side Request Forgery (SSRF)

Credits

  • ROCmertakdag (finder)
  • ambroisemaupate (coordinator)

Affected packages (4)

Vulnerable version ranges and first patched releases as published by GitHub.

Ecosystem Package Vulnerable range First patched Vulnerable functions
composer roadiz/documents >= 2.7.0, < 2.7.9 2.7.9
composer roadiz/documents >= 2.6.0, < 2.6.28 2.6.28
composer roadiz/documents >= 2.4.0, < 2.5.44 2.5.44
composer roadiz/documents < 2.3.42 2.3.42

References

cvelogic Threat Intelligence