Acronis Threat Research Unit (TRU) uncovered a multinational campaign in which a Chinese-speaking threat actor, tracked as Red Heron, rapidly weaponized CVE-2026-60004 to compromise internet-facing instances of Gitea, a self-hosted source-code management platform. The activity progressed from source-code theft to persistent access, credential collection, and lateral movement, including root-level access to a three-node Proxmox cluster. TRU traced a Linux implant to Red Heron’s exposed staging server, revealing the actor’s exploitation tools, reconnaissance databases, command history, stolen repositories, and malware. This provided rare visibility into the operation, from target selection and vulnerability weaponization to post-exploitation activity. Red Heron scanned 1,386 Gitea instances across seven countries and maintained a separate dataset of 477 Taiwan-based systems. Targets were classified using Simplified Chinese labels covering defense, elections, energy, aerospace, telecommunications, government, and research. Confirmed compromises affected organizations in Canada, Argentina, Taiwan, the United States, and Sri Lanka. Within days of the vulnerability’s July 2026 disclosure, Red Heron transformed public proof-of-concept code into an automated framework capable of registering accounts, exploiting vulnerable servers, stealing repositories, and removing selected traces. The campaign demonstrates how quickly n-day vulnerabilities in self-hosted development platforms can expose source code, credentials, secrets, and connected infrastructure. The staging server also contained JITTERLY, a C++ Linux implant supporting more than 30 post-exploitation commands, including shell execution, file transfer, network tunneling, interactive terminal access, and internal pivoting. Embedded inside it was SIXZUT, a previously undocumented LD\PRELOAD rootkit capable of hiding files, processes, and network connections, preventing the implant from being terminated, and relaunching it if the process is stopped while the binary remains present. TRU assesses with moderate confidence that Red Heron operates within a PRC-linked context, based on Simplified Chinese operational material, its classification of Taiwan as part of China, and targeting aligned with apparent strategic collection priorities. We have not identified sufficient evidence linking Red Heron to a previously tracked threat group.

More details - https://www.acronis.com/en/tru/posts/red-heron-exploits-gitea-n-day-flaw-in-multinational-campaign-exposing-new-linux-rootkit/

Acronis Threat Research Unit (TRU) uncovered a multinational campaign in which a Chinese-speaking threat actor, tracked as Red Heron, rapidly weaponized CVE-2026-60004 to compromise internet-facing instances of Gitea, a self-hosted source-code management platform. The activity progressed from source-code theft to persistent access, credential collection, and lateral movement, including root-level access to a three-node Proxmox cluster. TRU traced a Linux implant to Red Heron’s exposed staging server, revealing the actor’s exploitation tools, reconnaissance databases, command history, stolen repositories, and malware. This provided rare visibility into the operation, from target selection and vulnerability weaponization to post-exploitation activity. Red Heron scanned 1,386 Gitea instances across seven countries and maintained a separate dataset of 477 Taiwan-based systems. Targets were classified using Simplified Chinese labels covering defense, elections, energy, aerospace, telecommunications, government, and research. Confirmed compromises affected organizations in Canada, Argentina, Taiwan, the United States, and Sri Lanka. Within days of the vulnerability’s July 2026 disclosure, Red Heron transformed public proof-of-concept code into an automated framework capable of registering accounts, exploiting vulnerable servers, stealing repositories, and removing selected traces. The campaign demonstrates how quickly n-day vulnerabilities in self-hosted development platforms can expose source code, credentials, secrets, and connected infrastructure. The staging server also contained JITTERLY, a C++ Linux implant supporting more than 30 post-exploitation commands, including shell execution, file transfer, network tunneling, interactive terminal access, and internal pivoting. Embedded inside it was SIXZUT, a previously undocumented LD\PRELOAD rootkit capable of hiding files, processes, and network connections, preventing the implant from being terminated, and relaunching it if the process is stopped while the binary remains present. TRU assesses with moderate confidence that Red Heron operates within a PRC-linked context, based on Simplified Chinese operational material, its classification of Taiwan as part of China, and targeting aligned with apparent strategic collection priorities. We have not identified sufficient evidence linking Red Heron to a previously tracked threat group.

More details - https://www.acronis.com/en/tru/posts/red-heron-exploits-gitea-n-day-flaw-in-multinational-campaign-exposing-new-linux-rootkit/

Gitea does not properly validate repository ownership when linking attachments to releases. An attachment uploaded to a private repository could potentially be linked to a release in a different public repository, making it accessible to unauthorized users.

Gitea before 1.23.0 allows attackers to add attachments with forbidden file extensions by editing an attachment name via an attachment API.

Gitea before 1.22.3 mishandles access to a private resource upon receiving an API token with scope limited to public resources.

Gitea before 1.22.2 allows XSS because the search input box (for creating tags and branches) is v-html instead of v-text.

Gitea before 1.22.2 sometimes mishandles the propagation of token scope for access control within one of its own package registries.

In Gitea before 1.21.2, an anonymous user can visit a private user's project.

In Jenkins Gitea Plugin 1.4.4 and earlier, the implementation of Gitea personal access tokens did not support credentials masking, potentially exposing them through the build log.

In Gitea through 1.17.1, repo cloning can occur in the migration function.

Gitea version prior to version 1.5.1 contains a CWE-200 vulnerability that can result in Exposure of users private email addresses. This attack appear to be exploitable via Watch a repository to receive email notifications. Emails received contain the other recipients even if they have the email set as private. This vulnerability appears to have been fixed in 1.5.1.

In Gitea before 1.16.9, it was possible for users to add existing issues to projects. Due to improper access controls, an attacker could assign any issue to any project in Gitea (there was no permission check for fetching the issue). As a result, the attacker would get access to private issue titles.

Gitea before 1.5.4 allows remote code execution because it does not properly validate session IDs. This is related to session ID handling in the go-macaron/session code for Macaron.

Gitea 1.7.0 and earlier is affected by: Cross Site Scripting (XSS). The impact is: Attacker is able to have victim execute arbitrary JS in browser. The component is: go-get URL generation - PR to fix: https://github.com/go-gitea/gitea/pull/5905. The attack vector is: victim must open a specifically crafted URL. The fixed version is: 1.7.1 and later.

Gitea 1.7.2, 1.7.3 is affected by: Cross Site Scripting (XSS). The impact is: execute JavaScript in victim's browser, when the vulnerable repo page is loaded. The component is: repository's description. The attack vector is: victim must navigate to public and affected repo page.

Gitea version 1.6.2 and earlier contains a Incorrect Access Control vulnerability in Delete/Edit file functionallity that can result in the attacker deleting files outside the repository he/she has access to. This attack appears to be exploitable via the attacker must get write access to "any" repository including self-created ones. This vulnerability appears to have been fixed in 1.6.3, 1.7.0-rc2.

Gitea before 1.8.0 allows 1FA for user accounts that have completed 2FA enrollment. If a user's credentials are known, then an attacker could send them to the API without requiring the 2FA one-time password.

models/repomirror.go in Gitea before 1.7.6 and 1.8.x before 1.8-RC3 mishandles mirror repo URL settings, leading to remote code execution.

repo/setting.go in Gitea before 1.7.6 and 1.8.x before 1.8-RC3 does not validate the form.MirrorAddress before calling SaveAddress.

Specific Go Packages Affected github.com/go-gitea/gitea/models

Jenkins Gitea Plugin 1.1.1 and earlier did not implement trusted revisions, allowing attackers without commit access to the Git repo to change Jenkinsfiles even if Jenkins is configured to consider them to be untrusted.

Gitea 0.9.99 through 1.12.x before 1.12.6 does not prevent a git protocol path that specifies a TCP port number and also contains newlines (with URL encoding) in ParseRemoteAddr in modules/auth/repoform.go.

DISPUTED The git hook feature in Gitea 1.1.0 through 1.12.5 might allow for authenticated remote code execution in customer environments where the documentation was not understood (e.g., one viewpoint is that the dangerousness of this feature should be documented immediately above the ENABLEGITHOOKS line in the config file). NOTE: The vendor has indicated this is not a vulnerability and states "This is a functionality of the software that is limited to a very limited subset of accounts. If you give someone the privilege to execute arbitrary code on your server, they can execute arbitrary code on your server. We provide very clear warnings to users around this functionality and what it provides."

An issue was discovered in Gitea through 1.11.5. An attacker can trigger a deadlock by initiating a transfer of a repository's ownership from one organization to another.

Cross Site Request Forgery (CSRF) vulnerability exists in Gitea before 1.5.2 via API routes.This can be dangerous especially with state altering POST requests.

An Authentication Bypass vulnerability exists in Gitea before 1.5.0, which could let a malicious user gain privileges. If captured, the TOTP code for the 2FA can be submitted correctly more than once.

An issue exsits in Gitea through 1.15.7, which could let a malicious user gain privileges due to client side cookies not being deleted and the session remains valid on the server side for reuse.

Server Side Request Forgery (SSRF) vulneraility exists in Gitea before 1.7.0 using the OpenID URL.

Cross Site Scripting (XSS) vulnerability exists in Gitea before 1.5.1 via the repository settings inside the external wiki/issue tracker URL field.

Gitea before 1.4.3 is affected by URL Redirection to Untrusted Site ('Open Redirect') via internal URLs.