Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion community/catalog/compromises/2003/gentoo-rsync.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,4 +16,4 @@ responsible for serving package sources to users.

## References

- https://archives.gentoo.org/gentoo-announce/message/7b0581416ddd91522c14513cb789f17a
- [Gentoo Linux server compromised](https://www.zdnet.com/article/gentoo-linux-server-compromised/)
36 changes: 36 additions & 0 deletions community/catalog/compromises/2025/changed-files.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,36 @@
<!-- cSpell:ignore exfiltrated GHSA mrrh -->

# tj-actions/changed-files GitHub Action Compromise

In March 2025, attackers compromised the popular GitHub Action
`tj-actions/changed-files`, used by over 20,000 repositories to detect file
changes in pull requests.

The threat actor compromised a maintainer's credentials to manipulate Git tags,
redirecting trusted version references to a malicious commit that executed code
during CI/CD workflows.

The injected code captured environment variables and exfiltrated secrets such as
GitHub tokens and API credentials to an external server. This compromise
propagated silently through automated pipelines, as many users relied on mutable
version tags (e.g., v35, v36) instead of immutable commit SHAs, meaning their
workflows automatically pulled and executed the malicious code.

## Impact

This compromise had multiple implications across the GitHub Actions ecosystem
as thousands of repositories were possibly exposed through automate workflows,
any CI/CD runner secrets, repository tokens or organization credentials were
potentially at risk, and overall demonstrated how a single third-party action
could become a high-impact attack vector within trusted build automation
pipelines.

## Type of Compromise

This is a _Publishing Infrastructure_ type of attack as the attacker targeted a
GitHub action which is part of the CI/CD and build automation layer.

## References

- [GitHub Advisory Database - GHSA-mrrh-fwg8-r2c3](https://github.com/advisories/GHSA-mrrh-fwg8-r2c3/)
- [Wiz.io – GitHub Action tj-actions/changed-files Supply Chain Attack (CVE-2025-30066)](https://www.wiz.io/blog/github-action-tj-actions-changed-files-supply-chain-attack-cve-2025-30066)
30 changes: 30 additions & 0 deletions community/catalog/compromises/2025/ghost-action.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,30 @@
# The GhostAction Github Workflow Injection

In September 2025, GitGuardian discovered GhostAction campaign, a large-scale
supply chain campaign in which attackers compromised 327 GitHub user accounts
and injected malicious workflows into 817 repositories, stealing a total of 3,325
secrets. The malicious workflows, often titled "Github Actions Security",
were engineered to enumerate known secret names from legitimate workflow files
(e.g. PyPI tokens, npm tokens, DockerHub, AWS keys) and exfiltrate them via HTTP
POST to attacker-controlled endpoints.

## Impact

The GhostAction campaign compromised the trust and integrity of GitHub's
publishing pipelines, exposing sensitive credentials from hundreds of
open-source projects. The stolen secrets, including registry tokens and cloud
provider keys, could enable attackers to publish malicious packages, access
private infrastructure, or escalate to broader supply chain compromises across
ecosystems like npm, PyPI, and DockerHub. This incident highlights how
manipulating CI/CD workflows can undermine the integrity of the entire
open-source distribution chain.

## Type of Compromise

This compromise falls under the _Malicious Maintainer_ category, as the attackers
gained access to legitimate GitHub maintainer accounts and leveraged their
privileges to inject malicious workflow code.

## References

- [The GhostAction Campaign: 3,325 Secrets Stolen Through Compromised GitHub Workflows](https://blog.gitguardian.com/ghostaction-campaign-3-325-secrets-stolen)
36 changes: 36 additions & 0 deletions community/catalog/compromises/2025/npm-ecosystem.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,36 @@
<!-- cSpell:ignore Shai Hulud Shai hulud Shai-Hulud -->

# Widespread npm Ecosystem Compromise

The Widespread npm Ecosystem Compromise, which began around September 8, 2025,
was a multi-phased incident. The initial phase involved a phishing campaign that
compromised maintainer accounts, leading to the injection of a
cryptocurrency-stealing payload into dozens of popular packages (like chalk and
debug). This was quickly followed by the discovery of the "Shai-Hulud" worm
campaign, which used a self-propagating credential-stealing malware to
compromise over 500 npm packages.

## Impact

The compromise resulted in a widespread infection across the npm ecosystem,
affecting hundreds of packages and potentially thousands of downstream
applications that automatically pulled malicious versions. The injected payloads
enabled credential theft, unauthorized command execution, and persistent access
within both developer and CI/CD environments.

## Type of Compromise

The npm ecosystem is a _Malicious Maintainer_ type of attack as the attackers
managed to gain control of npm maintainer accounts and used their privileges to
push malicious versions of legitimate packages.

## References

- [Breakdown: Widespread npm Supply Chain Attack Puts Billions of Weekly Downloads at Risk](https://www.paloaltonetworks.com/blog/cloud-security/npm-supply-chain-attack/)
- [Ongoing Supply Chain Attack Involving npm Packages](https://www.csa.gov.sg/alerts-and-advisories/alerts/al-2025-093)
- [Shai-hulud supply chain attack spreads token-stealing malware on npm](https://www.reversinglabs.com/blog/shai-hulud-worm-npm)
- [npm Chalk and Debug Packages Hit in Software Supply Chain Attack](https://www.sonatype.com/blog/npm-chalk-and-debug-packages-hit-in-software-supply-chain-attack)
- [Another npm Supply Chain Attack: The 'is' Package Compromise](https://www.stepsecurity.io/blog/another-npm-supply-chain-attack-the-is-package-compromise)
- ["Shai-Hulud" Worm Compromises npm Ecosystem in Supply Chain Attack (Updated September 23)](https://unit42.paloaltonetworks.com/npm-supply-chain-attack)
- ["Massive npm infection: the Shai-Hulud worm and patient zero"](https://securelist.com/shai-hulud-worm-infects-500-npm-packages-in-a-supply-chain-attack/117547)
- [What We Know About the NPM Supply Chain Attack](https://www.trendmicro.com/en_us/research/25/i/npm-supply-chain-attack.html)
37 changes: 37 additions & 0 deletions community/catalog/compromises/2025/nx-platform.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,37 @@
<!-- cSpell:ignore ngularity exfiltrated -->

# The Nx s1ngularity Attack Leading to Credentials Leak

On August 26, 2025, attackers released malicious versions of the nx and @nx/*
npm packages (versions 20.9.0 through 21.8.0). The trojanized packages contained
credential-harvesting malware that scanned each developer system for sensitive
artifacts: GitHub tokens, npm keys, SSH private keys, environment variables,
cryptocurrency wallet files, and AI tool configurations. The malware exfiltrated
stolen credentials via double Base64 encoding and published them to over 1,400
public GitHub repositories, each named in a "s1ngularity-repository-*" pattern
with a single `results.b64` file containing encoded data.

## Impact

The Nx s1ngularity attack had an extensive impact across the open-source and
enterprise ecosystem. In total, over 20,000 files were exfiltrated, affecting
more than 1,700 users worldwide. The attackers leveraged stolen credentials to
make at least 6,700 private GitHub repositories public, exposing sensitive
source code, proprietary configurations, and credentials — some belonging to
major organizations and high-profile projects. This extensive exposure
underscored the cascading risk of software supply chain compromises, where a
single poisoned package can rapidly undermine trust and security across
thousands of interconnected development environments.

## Type of Compromise

This is an _Attack Chaining_ type of compromise with elements of _Dev Tooling_
and _Malicious Maintainer_, as the attackers initially leveraged compromised CI
workflows, published infected Nx packages, and chained the attack to expose
thousands of private repositories across the ecosystem.

## References

- [The Nx "s1ngularity" Attack: Inside the Credential Leak](https://blog.gitguardian.com/the-nx-s1ngularity-attack-inside-the-credential-leak/)
- [s1ngularity Nx Supply Chain Attack: AI-Driven Credential Theft & Mass Exposure](https://hivepro.com/threat-advisory/s1ngularity-nx-supply-chain-attack-ai-driven-credential-theft-mass-exposure/)
- [s1ngularity: Popular Nx Build System Package Compromised with Data-Stealing Malware](https://www.stepsecurity.io/blog/supply-chain-security-alert-popular-nx-build-system-package-compromised-with-data-stealing-malware)
43 changes: 43 additions & 0 deletions community/catalog/compromises/2025/oracle-cloud.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,43 @@
<!-- cSpell:ignore Exfiltrated exfiltrated -->

# Oracle Cloud SSO and Identity Infrastructure Compromise

The Oracle Cloud data breach, publicly disclosed around March 21, 2025, involved
a large-scale compromise of authentication and identity management systems. A
threat actor operating under the alias "rose87168" announced on the black-hat
forum BreachForums that they had exfiltrated a significant number of records
from Oracle Cloud's federated Single Sign-On (SSO) login servers and Lightweight
Directory Access Protocol (LDAP) systems.

The attacker claimed the initial infiltration occurred around mid-February 2025,
possibly exploiting a vulnerability in an older, unpatched component of the
infrastructure, such as Oracle Fusion Middleware 11G or a critical flaw in
Oracle Access Manager (potentially related to CVE-2021-35587). The compromise is
generally believed to have affected legacy Gen 1 servers and not the primary
Oracle Cloud Infrastructure (OCI) Gen 2 environment.

## Impact

The impact was focused on the mass compromise of critical authentication data,
significantly increasing security risks for numerous organizations. The 6
million records stolen included sensitive credentials such as encrypted
SSO/LDAP passwords, key files, and authentication tokens. This exposure created
a high risk of unauthorized account takeover, corporate espionage, and lateral
movement within affected customers' environments, particularly if the encrypted
credentials could be cracked. Furthermore, the threat actor sought to monetize
the breach through extortion, demanding fees from companies to remove their data
from the leak. The incident led to CISA guidance on credential risk mitigation
and resulted in class action lawsuits against Oracle for alleged failure to
implement standard data security practices and timely disclosure.

## Type of Compromise

Even though this was not related to a software package, this is considered to be
a _Publishing Infrastructure_ type of compromise as it originated from
vulnerabilities within Oracle’s identity and authentication infrastructure, a
critical part of its service publishing and access layer.

## References

- [CloudSEK – The Biggest Supply Chain Hack of 2025: 6M Records Exfiltrated from Oracle Cloud](https://www.cloudsek.com/blog/the-biggest-supply-chain-hack-of-2025-6m-records-for-sale-exfiltrated-from-oracle-cloud-affecting-over-140k-tenants)
- [CVE-2021-35587 – Oracle Access Manager Remote Code Execution Vulnerability](https://nvd.nist.gov/vuln/detail/CVE-2021-35587)
31 changes: 31 additions & 0 deletions community/catalog/compromises/2025/review-dog.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
<!-- cSpell:ignore reviewdog -->

# reviewdog/action-setup@v1 GitHub Action Compromise

In March 2025, security researchers discovered that the reviewdog/action-setup@v1
GitHub Action had been compromised. The attacker altered the v1 tag to point to a
malicious commit between March 11 and later reverted it to conceal the compromise.
Encoded payloads were embedded into the install.sh script and any running
workflows using this Action would execute the malicious code. The code, when
executed in CI pipelines, could dump workflow environment variables into logs,
exposing them this way to anyone viewing the CI run.

## Impact

By redirecting the trusted @v1 tag to a malicious commit, the attacker caused
workflows using this Action to execute injected code that printed environment
variables and secrets into build logs. This could lead to the unintentional
disclosure of access tokens, API keys, and credentials, particularly in public
repositories where logs are accessible, undermining the confidentiality of
automated build environments.

## Type of Compromise

This is a _Publishing Infrastructure_ type of compromise, as the attacker
manipulated the Action's distributed version reference (Git tag) rather than its
codebase or maintainer, abusing weaknesses in how automation components are
published and trusted within GitHub's workflow ecosystem.

## References

- [New GitHub Action supply chain attack: reviewdog/action-setup](https://www.wiz.io/blog/new-github-action-supply-chain-attack-reviewdog-action-setup)
34 changes: 34 additions & 0 deletions community/catalog/compromises/2025/rh-gitlab-instance.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,34 @@
<!-- cSpell:ignore exfiltrated -->

# Red Hat Consulting GitLab Instance Breach

In October 2025, Red Hat confirmed a security breach affecting a self-hosted
GitLab instance used internally by its Consulting division. The threat actor
group calling themselves Crimson Collective claimed responsibility, asserting
they had exfiltrated approximately 570 GB of compressed data from 28,000 private
repositories, including around 800 Customer Engagement Reports (CERs). These
CERs often contain sensitive customer architecture diagrams, configuration files,
authentication tokens, and infrastructure details.

## Impact

Attackers claim to have accessed a significant volume of sensitive consulting
and customer data.Stolen CERs may reveal network topologies, access credentials,
and deployment configurations for major enterprise clients. That kind of
information could be leveraged for secondary intrusions or social engineering.
While Red Hat has not confirmed any misuse of the stolen data, the incident
highlights the inherent risk of third-party data exposure within vendor
ecosystems. Red Hat emphasized that no personal data or software supply chain
assets have been confirmed compromised at this stage.

## Type of Compromise

This is a _Publishing Infrastructure_ type of compromise as the compromise
occurred within Red Hat’s internal GitLab environment, which is part of its
development and collaboration infrastructure.
Comment on lines +26 to +28
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Again, I think this is a supply chain attack, but it's not clear that it's a software supply chain attack, despite the compromised infrastructure being a source control repository in this case. (The same attack could be applied against a consulting firm's Sharepoint or Google Workspace accounts with equivalent effect.)

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Agreed, I totally see your point. Do you think that adding a new category such as "3rd Party Vendor" type would make more sense for describing this type of incidents?


## References

- [Security update: Incident related to Red Hat Consulting GitLab instance](https://www.redhat.com/en/blog/security-update-incident-related-red-hat-consulting-gitlab-instance)
- [Red Hat confirms security incident after hackers breach GitLab instance](https://www.bleepingcomputer.com/news/security/red-hat-confirms-security-incident-after-hackers-breach-gitlab-instance)
- [Red Hat GitLab Data Breach: The Crimson Collective's Attack](https://blog.gitguardian.com/red-hat-gitlab-breach-the-crimson-collectives-attack/)
Loading