CVE-2026-54054 in Transmuteinfo

Summary

by MITRE • 09/11/2026

Transmute is a free, open-source, self-hosted file conversion and compression tool. Prior to version 1.3.0, Transmute's URL import endpoint, `POST /api/files/url`, is vulnerable to Server-Side Request Forgery (SSRF). The HTTP downloader used by this endpoint fetches user-supplied URLs with redirects enabled and does not validate whether the target resolves to a public, external address. As a result, an authenticated user (or guest user if they are enabled) may be able to cause the Transmute server to make HTTP requests to internal or cloud-local resources from the server's network position. Because downloaded content is stored and can later be retrieved through `GET /api/files/{id}`, this issue can result in full-read SSRF rather than blind SSRF. This is fixed in version 1.3.0.

Once again VulDB remains the best source for vulnerability data.

Analysis

by VulDB Data Team • 09/11/2026

The vulnerability identified in the Transmute file conversion and compression tool constitutes a critical Server-Side Request Forgery (SSRF) flaw within its URL import functionality, specifically targeting the POST /api/files/url endpoint prior to version 1.3.0. This security defect arises from insufficient validation of user-supplied URLs during the HTTP download process. The application's downloader is configured with redirects enabled and lacks a mechanism to verify whether the target address resolves to a public or external network resource. Consequently, an attacker can manipulate the input parameters to direct the server to fetch content from internal networks, cloud-local metadata services, or other restricted resources that are not intended to be accessible via this interface. This misconfiguration allows the application to act as a proxy for unauthorized requests originating from the server's own network position, bypassing standard perimeter defenses and access controls designed to isolate external-facing applications from sensitive backend infrastructure.

The operational impact of this vulnerability is significantly amplified by the architectural design of Transmute regarding how downloaded content is handled. Unlike many SSRF vulnerabilities that result in blind responses where the attacker cannot see the outcome of the server-side request, this flaw enables full-read SSRF capabilities. Because the application stores the fetched content and makes it retrievable via a subsequent GET /api/files/{id} endpoint, an authenticated user can observe the response body returned by the internal or cloud-local resources. This capability transforms a potentially limited injection point into a powerful data exfiltration vector. An attacker could leverage this to read sensitive configuration files from local servers, access cloud provider instance metadata such as temporary security credentials, or probe internal network services like databases and administrative panels that are typically hidden behind firewalls. The requirement for authentication means the attack surface is restricted to users with valid accounts, although guest user access if enabled would further expand the potential attacker pool.

From a classification perspective, this vulnerability aligns closely with CWE-918, which defines Server-Side Request Forgery (SSRF) flaws where the server makes requests based on unvalidated user input. The specific failure to validate that the target resolves to a public address corresponds to CWE-623, specifically regarding the use of dangerous functions or APIs without proper validation of inputs for security implications. In terms of offensive tactics, this behavior maps directly to MITRE ATT&CK technique T1571, which covers Non-Standard Ports and protocols often used in SSRF attacks to bypass network restrictions, as well as T1046, Network Service Discovery, where the attacker uses the server to map out internal networks. The ability to retrieve data via a secondary endpoint also relates to T1078, Valid Accounts, if authentication is required, highlighting how valid credentials can be abused for lateral movement and information gathering within an organization's network perimeter.

Mitigation strategies must prioritize immediate patching of the Transmute application to version 1.3.0 or later, where this validation logic has been corrected. For environments where upgrading is not immediately feasible, defensive measures should include implementing a strict allowlist approach for URL parsing rather than relying on blocklists, ensuring that only explicitly permitted domains and protocols are processed by the downloader. Additionally, network-level controls such as egress filtering can be employed to restrict outbound HTTP connections from the Transmute server's host or container environment to known external destinations, thereby blocking requests intended for internal IP ranges or cloud metadata endpoints like 169.254.169.254. It is also advisable to disable guest user access if not strictly necessary and to implement robust input validation that checks both the domain name against a whitelist and resolves DNS records to ensure they do not point to private, loopback, or link-local IP addresses before initiating any network connection.

Responsible

GitHub M

Reservation

06/11/2026

Disclosure

09/11/2026

Moderation

accepted

CPE

ready

EPSS

0.00000

KEV

no

Activities

very low

Sources

Might our Artificial Intelligence support you?

Check our Alexa App!