Gogs Flaw Attackers Overwrite Large File Storage Data
A critical security vulnerability has been identified in Gogs, a widely used open-source self-hosted Git service. /
Tracked as CVE-2026-25921, this flaw allows unauthenticated attackers to silently overwrite Git Large File Storage (LFS) objects across any repository.
By exploiting a lack of content verification, threat actors can conduct stealthy software supply-chain attacks, replacing legitimate project files with malicious backdoors.
Git LFS is designed to replace large files, such as audio, datasets, and compiled binaries, with text pointers inside Git, storing the actual file contents on a remote server.
The vulnerability in Gogs stems from two dangerous architectural decisions in how these remote files are handled:
An attacker can exploit this flaw with very low complexity over the network. The attack begins when a threat actor identifies the OID of a target LFS object in a victim’s repository.
The attacker then initiates a file upload to their own repository but intentionally uses the victim’s OID.
During this upload, the attacker submits a tampered file, such as a compromised binary or a script embedded with malware.
Gogs blindly accepts the file and overwrites the legitimate LFS object in the global storage path.
Because there is no integrity check, users downloading the target LFS object from the original repository will silently receive the attacker’s malicious payload.
There are no warnings presented on the webpage or the command line, making this an incredibly dangerous vector for software supply-chain compromises.
Following the disclosure of CVE-2026-25921 by security researcher zjuchenyuan, the maintainers of Gogs released an official fix.
Organizations relying on Gogs must implement the following defensive measures:
Follow us on Google News, LinkedIn, and X to Get Instant Updates and Set GBH as a Preferred Source in Google.
Attackers abused compromised ConnectWise Cloud Backup credentials to access Microsoft 365 customer environments, prompting ConnectWise…
Anthropic says an internal Claude research model reconstructed how the National Security Agency generated seeds…
Two WordPress Core vulnerabilities can be chained to execute remote code without authentication, according to…
A newly disclosed OpenBSD kernel exploitation chain combines a shared memory accounting flaw with a…
GhostAction attackers have expanded their GitHub credential theft campaign with a second stage that disguises…
Indian authorities have warned that SideCopy hackers are targeting Indian organizations, including academic institutions, with…