A supply chain attack has compromised the node-ipc npm package, with malicious versions 9.1.6, 9.2.3, and 12.0.1 containing obfuscated stealer and backdoor functionality. The attack vector involved takeover of a dormant maintainer account through an expired email domain. The malware fingerprints host environments, enumerates and reads local files including SSH keys, cloud credentials, database configurations, and various developer secrets. Collected data is compressed into a gzip archive and exfiltrated via DNS TXT queries to attacker-controlled infrastructure disguised as legitimate Azure domains. The payload targets over 100 file patterns across macOS and Linux systems, focusing on developer credentials from AWS, Azure, GCP, Kubernetes, Docker, npm, GitHub, and numerous other services. The malicious code executes during CommonJS module loading, forking a detached child process to perform credential harvesting while avoiding detection through obfuscation and DNS-based covert channels.
The Nexa Blocks – Gutenberg Blocks, Page Builder for Gutenberg Editor & FSE plugin for WordPress is vulnerable to Server-Side Request Forgery (SSRF) in versions up to and including 1.1.1. This is due to the import_demo() function accepting a user-supplied URL in the demo_json_file POST parameter and passing it directly to wp_remote_get() without any URL validation or restriction against internal or private network destinations. The nexa_blocks_nonce required for the AJAX action is publicly exposed in the HTML source of any frontend page where the plugin is active via wp_localize_script on the enqueue_block_assets hook, effectively making the nonce available to all visitors and bypassing any intended authentication barrier. This makes it possible for unauthenticated attackers to make server-side HTTP requests to arbitrary internal or external destinations, potentially exposing internal services, cloud metadata endpoints such as the AWS instance metadata service, localhost services, and other resources not intended to be publicly accessible. A secondary SSRF vector also exists whereby image URLs extracted from the attacker-controlled JSON response are subsequently fetched via a second wp_remote_get() call, allowing chained exploitation through a crafted JSON payload.
A vulnerability in the Linux kernel's algif_aead module, known as Copy Fail (CVE-2026-31431), improperly handles in-place cryptographic operations. This flaw allows a local attacker to escalate privileges or potentially escape a container. Additional security issues affecting the Cryptographic API, Packet sockets, and TLS protocol were also identified. These vulnerabilities could be exploited to compromise the system. Fixes have been released for multiple Linux kernel versions, including those for various cloud and specialized platforms.
A contractor for the Cybersecurity & Infrastructure Security Agency (CISA) publicly exposed highly privileged AWS GovCloud credentials and numerous internal system passwords in a GitHub repository. The repository contained plaintext passwords, cloud keys, tokens, and internal documentation related to CISA's software development and deployment processes. The exposure lasted from November 2025 until May 2026, when the repository was taken offline after notification. CISA confirmed awareness and ongoing investigation, stating no indication of data compromise so far. The exposed credentials included administrative access to multiple AWS GovCloud accounts and internal repositories, posing a significant security risk.
Other noteworthy stories that might have slipped under the radar: Nvidia cloud gaming data breach, Android 17 security upgrades, FBI warning after ShinyHunters hacks Canvas. The post In Other News: Big Tech vs Canada Encryption Bill, Cisco’s Free AI Security Spec, Audi App Flaws appeared first on SecurityWeek .
Spring Cloud AWS versions from 3.0.0 up to 4.0.1 did not verify the signature of incoming SNS messages on HTTP/HTTPS endpoints using specific notification mapping annotations. This allowed an unauthenticated attacker who knows the endpoint URL to send crafted HTTP POST requests mimicking SNS Notification or SubscriptionConfirmation messages. The vulnerability is fixed in version 4.0.2.
CVE-2026-42141 is a Server-Side Request Forgery (SSRF) vulnerability in the Xibo CMS digital signage platform prior to version 4.4.1. Authenticated users with Library upload permissions can exploit this flaw to make arbitrary HTTP requests from the CMS server to internal or external network resources. This can lead to scanning internal infrastructure, accessing cloud metadata endpoints such as AWS IMDS, interacting with unauthenticated internal services, or exfiltrating data. The vulnerability has a high severity with a CVSS score of 7.7 and is fixed in version 4.4.1. The vendor manages remediation for this cloud-hosted service.
Multiple vulnerabilities were discovered in the Linux kernel, including in the AppArmor Linux Security Module and various kernel subsystems such as GPU drivers, network drivers, Sun RPC protocol, and XFRM subsystem. These flaws could allow an unprivileged local attacker to load, replace, or remove AppArmor profiles, potentially causing denial of service, exposure of kernel memory, local privilege escalation, or container escape. The issues affect various Linux kernel versions used in Ubuntu 16.04 and 18.04 LTS, including specialized kernels for cloud environments. Fixes have been released and are available through Ubuntu security updates.
Ech0 contains a Server-Side Request Forgery (SSRF) vulnerability in the fetchPeerConnectInfo function, which uses an unsafe HTTP request method without URL validation. Authenticated users can exploit this to make the server request arbitrary URLs, including internal and cloud metadata endpoints such as AWS instance metadata and Kubernetes API endpoints. This vulnerability affects versions prior to 1.4.8-0.20260503040602-091d26d2d942 and has a high severity rating.
### Impact
In the SDK embedder path (`N8NDocumentationMCPServer` constructor, `getN8nApiClient()`, and `validateInstanceContext()`), the synchronous URL validator in `SSRFProtection.validateUrlSync()` had no IPv6 checks. IPv4-mapped IPv6 addresses such as `http://[::ffff:169.254.169.254]` bypassed the cloud-metadata, localhost, and private-IP range checks. An attacker able to supply an `n8nApiUrl` value could cause the server to issue HTTP requests to cloud metadata endpoints (AWS IMDS, GCP, Azure, Alibaba, Oracle), RFC1918 private networks, or localhost services. Response bodies are returned to the caller (non-blind SSRF), and the `n8nApiKey` is forwarded in the `x-n8n-api-key` header to the attacker-controlled target.
The first-party HTTP server deployment was not primarily affected — it has a second async validator (`validateWebhookUrl`) that catches IPv6 addresses.
Impact category: **CWE-918** (Server-Side Request Forgery).
### Affected
Deployments embedding n8n-mcp as an SDK using `N8NDocumentationMCPServer` or `N8NMCPEngine` with user-supplied `InstanceContext` on versions **v2.47.4 through v2.47.13**.
### Patched
**v2.47.14** and later.
- npm: `npx n8n-mcp@latest` (or pin to `>= 2.47.14`)
- Docker: `docker pull ghcr.io/czlonkowski/n8n-mcp:latest`
### Workarounds
If developers cannot upgrade immediately:
- **Validate URLs before passing to the SDK** — reject any `n8nApiUrl` whose hostname is an IP literal (bracketed IPv6 or dotted IPv4) before calling `N8NDocumentationMCPServer` / `getN8nApiClient()`. Accept only URLs with DNS-resolvable hostnames.
- **Restrict egress at the network layer** — block outbound traffic from the n8n-mcp process to RFC1918 ranges (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16), link-local `169.254.0.0/16`, and cloud metadata endpoints. Defense-in-depth against this class of issue and recommended even after upgrading.
- **Do not accept user-controlled `n8nApiUrl` values** — if the project's integration derives the URL from internal configuration only, this vulnerability is not reachable.
Upgrading to v2.47.14 is still strongly recommended.
### Credit
Reported by @manthanghasadiya.
A supply chain operation dubbed 'Mini Shai Hulud' compromised SAP-related npm packages by injecting malicious preinstall scripts that execute during installation. The campaign leverages multi-stage payloads to harvest developer and CI/CD secrets from GitHub, npm, and major cloud providers, exfiltrating data via attacker-controlled GitHub repositories. Malicious versions of legitimate SAP ecosystem packages execute obfuscated payloads that collect GitHub tokens, npm credentials, cloud secrets from AWS, Azure and GCP, Kubernetes tokens, and GitHub Actions secrets. The malware includes propagation logic to infect additional repositories and features browser credential theft capabilities. It performs language checks to avoid Russian-speaking systems. Attribution points to TeamPCP based on shared RSA public keys and overlapping techniques from previous operations.
Xinference, an open-source distributed AI model inference framework, suffered a supply chain attack when attackers compromised PyPI release credentials of maintainers and published three malicious versions (2.6.0, 2.6.1, 2.6.2) on April 22, 2026. The malicious code, encoded in Base64 layers within __init__.py, executes automatically upon library installation or import, collecting cloud credentials, SSH keys, API tokens, database passwords, cryptocurrency wallets, and environment variables. The payload specifically targets AWS environments through metadata service exploitation and uploads stolen data to attacker-controlled infrastructure. The attack affects users who downloaded these versions from PyPI, which has over 680,000 total downloads. Attribution remains unclear as TeamPCP's name appears in the code but the group denies involvement, suggesting third-party impersonation.
CVE-2026-6912 is a high-severity vulnerability in AWS Ops Wheel affecting the Cognito User Pool configuration. It allows remote authenticated users to escalate privileges to deployment admin by exploiting improper control over dynamically-determined object attributes via a crafted UpdateUserAttributes API call. This enables attackers to set the custom:deployment_admin attribute and manage Cognito user accounts. The vulnerability is present before pull request #165 and affects version 0 of AWS Ops Wheel. AWS manages remediation for this cloud service and has released a fix. Users should redeploy from the updated repository and patch any forked or derivative code accordingly.
Xibo is an open source digital signage platform with a web content management system and Windows display player software. An authenticated Server-Side Request Forgery (SSRF) vulnerability in versions prior to 4.4.1 allows users with DataSet permissions to make arbitrary HTTP requests from the CMS server to internal or external network resources. This can be exploited to scan internal infrastructure, access local cloud metadata endpoints (e.g., AWS IMDS), interact with internal services that lack authentication, or exfiltrate data. Exploitation of the vulnerability is possible on behalf of an authorized user who has both of the following privileges, which are not granted to non-admins as standard: Include "Add DataSet" button to allow for additional DataSets to be created independently to Layouts. Users should upgrade to version 4.4.1 which fixes this issue. Upgrading to a fixed version is necessary to remediate. Users unable to upgrade should revoke such privileges from users they do not trust.
OpenClaw versions before 2026.3.28 have an environment variable sanitization vulnerability due to an incomplete blocklist that omits GIT_TEMPLATE_DIR and AWS_CONFIG_FILE. This flaw allows attackers to exploit approved execution requests by redirecting git or AWS CLI behavior through malicious configuration files, potentially leading to execution of untrusted code or loading of malicious credentials. The vulnerability has a medium severity score of 5.8 and affects local attack vectors with high attack complexity and limited privileges required. A patch is available, and as OpenClaw is a cloud service, the vendor manages remediation server-side.
Vexa is an open-source, self-hostable meeting bot API and meeting transcription API. Prior to 0.10.0-260419-1910, the Vexa webhook feature allows authenticated users to configure an arbitrary URL that receives HTTP POST requests when meetings complete. The application performs no validation on the webhook URL, enabling Server-Side Request Forgery (SSRF). An authenticated attacker can set their webhook URL to target internal services (Redis, databases, admin panels), cloud metadata endpoints (AWS/GCP credential theft), and/or localhost services. Version 0.10.0-260419-1910 patches the issue.
Multiple security vulnerabilities were identified in the Linux kernel for Azure cloud systems, affecting cryptographic API, Netfilter, and network traffic control subsystems. These flaws could potentially allow an attacker to compromise the system. Fixed versions have been released, and users are advised to update and reboot their systems. Due to an ABI change, recompilation of third-party kernel modules is necessary after updating.
## Impact
An authenticated Server-Side Request Forgery in `n8n-mcp` allows a caller holding a valid `AUTH_TOKEN` to cause the server to issue HTTP requests to arbitrary URLs supplied through multi-tenant HTTP headers. Response bodies are reflected back through JSON-RPC, so an attacker can read the contents of any URL the server can reach — including cloud instance metadata endpoints (AWS IMDS, GCP, Azure, Alibaba, Oracle), internal network services, and any other host the server process has network access to.
The primary at-risk deployments are multi-tenant HTTP installations where more than one operator can present a valid `AUTH_TOKEN`, or where a token is shared with less-trusted clients. Single-tenant stdio deployments and HTTP deployments without multi-tenant headers are not affected.
## Affected versions
`n8n-mcp` ≤ `2.47.3` (all versions up to and including 2.47.3).
## Patched versions
`n8n-mcp` `2.47.4` and later.
## Workarounds
If you cannot immediately upgrade:
1. **Egress filtering at the network layer** — block outbound traffic from the `n8n-mcp` container to RFC1918 ranges (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16), link-local `169.254.0.0/16`, and any other internal ranges. This defends against any future SSRF-class issue and is recommended even after upgrading.
2. **Disable multi-tenant headers** — if your deployment does not require per-request instance switching, unset `ENABLE_MULTI_TENANT` and do not accept `x-n8n-url` / `x-n8n-key` headers at the reverse proxy.
3. **Restrict `AUTH_TOKEN` distribution** — ensure the bearer token is only held by fully trusted operators until you can upgrade.
## Remediation
Upgrade to `n8n-mcp` 2.47.4 or later. No configuration changes are required; the fix adds validation at the URL entry points and normalizes URLs at the API client layer.
## Credits
Reported by the Eresus Security Research Team. @ibrahmsql
Multiple security vulnerabilities were identified in the Linux kernel affecting various subsystems including ARM64 architecture, Cryptographic API, Netfilter, and network traffic control. These flaws could potentially allow an attacker to compromise the system. The issues are addressed in updated Linux kernel versions for multiple distributions and cloud platforms. Users are advised to apply the available patches and reboot their systems to ensure the fixes are applied. Due to ABI changes, recompilation of third-party kernel modules may be necessary.
LinkAce is a self-hosted archive to collect website links. Prior to 2.5.4, LinkRepository::update and CheckLinksCommand::checkLink do not check for private IPs. An authenticated user can read responses from internal services (AWS IMDSv1, cloud metadata, internal APIs) by creating a link with a public URL and then updating it to a private IP. The links:check cron job makes the request server-side without IP filtering. This can expose cloud credentials, internal service data, and network topology. This vulnerability is fixed in 2.5.4.