Duplicate Advisory: OpenClaw: MCP loopback owner context is derived from server-issued bearer tokens

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-r6xh-pqhr-v4xh. This link is maintained to preserve external references. Original Description OpenClaw before 2026.4.22 derives loopback MCP owner context from spoofable server-issued bearer tokens in request...

Pelican Web UI Affected by a Privilege Escalation Attack

Background On April 2nd, 2026, a Claude coding agent alerted Pelican PI Brian Bockelman to a privilege escalation vulnerability affecting Pelican's Web User Interface WebUI for various versions between v7.21 and v7.24. Upon further investigation, the Pelican team discovered this attack allows any...

OpenClaw: Agent gateway config mutations could change protected operator settings

Affected Packages / Versions - Package: openclaw npm - Affected versions: 2026.4.20 - Patched version: 2026.4.20 Impact The agent-facing gateway config.patch / config.apply guard did not cover several operator-trusted settings, including sandbox policy, plugin enablement, gateway auth/TLS, hook...

python-dotenv: Symlink following in set_key allows arbitrary file overwrite via cross-device rename fallback

Summary setkey and unsetkey in python-dotenv follow symbolic links when rewriting .env files, allowing a local attacker to overwrite arbitrary files via a crafted symlink when a cross-device rename fallback is triggered. Details The rewrite context manager in dotenv/main.py is used by both setkey...

DOMPurify ADD_ATTR predicate skips URI validation

Summary DOMPurify allows ADDATTR to be provided as a predicate function via EXTRAELEMENTHANDLING.attributeCheck. When the predicate returns true, isValidAttribute short-circuits the attribute check before URI-safe validation runs. An attacker who supplies a predicate that accepts specific...

Traefik: HTTP/2 frames can cause a running server to panic

Summary More Details: - https://nvd.nist.gov/vuln/detail/CVE-2026-27141 - https://pkg.go.dev/golang.org/x/net/http2?tab=versions Patches - https://github.com/traefik/traefik/releases/tag/v3.6.10 - https://github.com/traefik/traefik/releases/tag/v2.11.40 For more information If you have any...

Keystone has an unintended `isFilterable` bypass that can be used as an oracle to match hidden fields

Summary field.isFilterable access control can be bypassed in update and delete mutations by adding additional unique filters. These filters can be used as an oracle to probe the existence or value of otherwise unreadable fields. Specifically, when a mutation includes a where clause with multiple...

vLLM uses Python 3.12 built-in hash() which leads to predictable hash collisions in prefix cache

Summary Maliciously constructed prompts can lead to hash collisions, resulting in prefix cache reuse, which can interfere with subsequent responses and cause unintended behavior. Details vLLM's prefix caching makes use of Python's built-in hash function. As of Python 3.12, the behavior of hashNon...

Querydsl vulnerable to HQL injection through orderBy

Summary The order by method enables injecting HQL queries. This may cause blind HQL injection, which could lead to leakage of sensitive information, and potentially also Denial Of Service. This vulnerability is present since the original querydsl repositoryhttps://github.com/querydsl/querydsl whe...

SAML authentication bypass via Incorrect XPath selector

Ruby-SAML in = 12.2 and 1.13.0 = 1.16.0 does not properly verify the signature of the SAML Response. An unauthenticated attacker with access to any signed saml document by the IdP can thus forge a SAML Response/Assertion with arbitrary contents. This would allow the attacker to log in as arbitrar...

Drupal Full Path Disclosure

core/authorize.php in Drupal 11.x-dev allows Full Path Disclosure even when error logging is None if the value of hashsalt is filegetcontents of a file that does not exist...

Undertow vulnerable to Race Condition

A vulnerability was found in Undertow where the ProxyProtocolReadListener reuses the same StringBuilder instance across multiple requests. This issue occurs when the parseProxyProtocolV1 method processes multiple requests on the same HTTP connection. As a result, different requests may share the...

Django SQL injection vulnerability

An issue was discovered in Django 5.0 before 5.0.8 and 4.2 before 4.2.15. QuerySet.values and valueslist methods on models with a JSONField are subject to SQL injection in column aliases via a crafted JSON object key as a passed arg...

GraphQL Java does not properly consider ExecutableNormalizedFields (ENFs) as part of preventing denial of service

GraphQL Java aka graphql-java before 21.5 does not properly consider ExecutableNormalizedFields ENFs as part of preventing denial of service via introspection queries. 20.9 and 19.11 are also fixed versions...

aimeos/ai-admin-graphql improper access control vulnerability allows an editor to modify admin account

aimeos/ai-admin-graphql is the Aimeos GraphQL API admin interface. Starting in version 2022.04.01 and prior to versions 2022.10.10, 2023.10.6, and 2024.04.6, an improper access control vulnerability allows an editor to modify and take over an admin account in the back end. Versions 2022.10.10,...

MLFlow unsafe deserialization

Deserialization of untrusted data can occur in versions of the MLflow platform running version 0.5.0 or newer, enabling a maliciously uploaded PyTorch model to run arbitrary code on an end user’s system when interacted with...

Read private customer data reclaiming carts in Klaviyo Magento

A researcher identified an endpoint in a thirth party module Klaviyo Magento 2 which allows to read private customer data from stores. It works by reclaiming any guest-cart as your own and reading the private data for the orders in the Magento API...

Withdrawn: JJWT improperly generates signing keys

Withdrawn Advisory This advisory has been withdrawn because it has been found to be disputed. Please see the issue here for more information. Original Description JJWT aka Java JWT through 0.12.5 ignores certain characters and thus a user might falsely conclude that they have a strong key. The...

Server Side Template Injection (SSTI) via Twig escape handler

Summary Due to the unrestricted access to twig extension class from grav context, an attacker can redefine the escape function and execute arbitrary commands. Details https://github.com/twigphp/Twig/blob/3.x/src/Extension/EscaperExtension.phpL99 php / Defines a new escaper to be used via the esca...

Apache Superset: Improper error handling on alerts

An authenticated user with privileges to create Alerts on Alerts & Reports has the capability to generate a specially crafted SQL statement that triggers an error on the database. This error is not properly handled by Apache Superset and may inadvertently surface in the error log of the Alert...

Apache Axis Improper Input Validation vulnerability

UNSUPPORTED WHEN ASSIGNED Improper Input Validation vulnerability in Apache Axis allowed users with access to the admin service to perform possible SSRF. This issue affects Apache Axis through 1.3. As Axis 1 has been EOL, we recommend you migrate to a different SOAP engine, such as Apache Axis...

xkeys seal encryption used fixed key for all encryption

Background NATS.io is a high performance open source pub-sub distributed communication technology, built for the cloud, on-premise, IoT, and edge computing. The cryptographic key handling library, nkeys, recently gained support for encryption, not just for signing/authentication. This is used in...

Ingress-nginx code injection via nginx.ingress.kubernetes.io/permanent-redirect annotation

A security issue was identified in ingress-nginx where the nginx.ingress.kubernetes.io/permanent-redirect annotation on an Ingress object in the networking.k8s.io or extensions API group can be used to inject arbitrary commands, and obtain the credentials of the ingress-nginx controller. In the...

Werkzeug DoS: High resource usage when parsing multipart/form-data containing a large part with CR/LF character at the beginning

Werkzeug multipart data parser needs to find a boundary that may be between consecutive chunks. That's why parsing is based on looking for newline characters. Unfortunately, code looking for partial boundary in the buffer is written inefficiently, so if we upload a file that starts with CR or LF...

OpenSearch uncontrolled resource consumption

Impact An issue has been identified with how OpenSearch handled incoming requests on the HTTP layer. An unauthenticated user could force an OpenSearch node to exit with an OutOfMemory error by sending a moderate number of malformed HTTP requests. The issue was identified by Elastic Engineering an...

JWT token compromise can allow malicious actions including Remote Code Execution (RCE)

Impact A user can reverse engineer the JWT token JSON Web Token used in authentication for Manager and API access, forging a valid NeuVector Token to perform malicious activity in NeuVector. This can lead to an RCE. Patches Upgrade to NeuVector version 5.2.2 or later and latest Helm chart 2.6.3+....

Presto JDBC Server-Side Request Forgery by redirect

Summary Presto JDBC is vulnerable to Server-Side Request Forgery SSRF when connecting a remote Presto server. An attacker can construct a redirect response that Presto JDBC client will follow and view sensitive information from highly sensitive internal servers or perform a local port scan. Detai...

Microsoft Security Advisory CVE-2023-36794: .NET Remote Code Execution Vulnerability

Microsoft Security Advisory CVE-2023-36794: .NET Remote Code Execution Vulnerability Executive summary Microsoft is releasing this security advisory to provide information about a vulnerability in .NET 7.0 and .NET 6.0. This advisory also provides guidance on what developers can do to update thei...

Electron's Content-Secrity-Policy disabling eval not applied consistently in renderers with sandbox disabled

Impact A Content-Security-Policy that disables eval, specifically setting a script-src directive and not providing unsafe-eval in that directive, is not respected in renderers that have sandbox and contextIsolation disabled. i.e. sandbox: false and contextIsolation: false in the webPreferences...

Cargo not respecting umask when extracting crate archives

The Rust Security Response WG was notified that Cargo did not respect the umask when extracting crate archives on UNIX-like systems. If the user downloaded a crate containing files writeable by any local user, another local user could exploit this to change the source code compiled and executed b...

Decidim vulnerable to sensitive data disclosure

Note: added the actual report as a comment. Summary Decidim, a platform for digital citizen participation, uses a third-party library named Ransack for filtering certain database collections e.g., public meetings. By default, this library allows filtering on all data attributes and associations...

XWiki Platform vulnerable to reflected cross-site scripting via xredirect parameter in delete template

Impact Users are able to forge an URL with a payload allowing to inject Javascript in the page XSS. It's possible to exploit the delete template to perform a XSS, e.g. by using URL such as: xwiki/bin/get/FlamingoThemes/Cerulean?xpage=xpart&vm=delete.vm&xredirect=javascript:alertdocument.domain Th...

php-imap vulnerable to RCE through a directory traversal vulnerability

Summary An unsanitized attachment filename allows any unauthenticated user to leverage a directory traversal vulnerability which results in a remote code execution vulnerability. Details An attacker can send an email with a malicious attachment to the inbox, which gets crawled with webklex/php-im...

OutOfMemoryError for large multipart without filename in Eclipse Jetty

Impact Servlets with multipart support e.g. annotated with @MultipartConfig that call HttpServletRequest.getParameter or HttpServletRequest.getParts may cause OutOfMemoryError when the client sends a multipart request with a part that has a name but no filename and a very large content. This...

Shopware Has Improper Control of Generation of Code in Twig rendered views

Impact We fixed with CVE-2023-22731 Twig filters to only be executed with allowed functions. It is possible to pass PHP Closures as string or an array and array crafted PHP Closures was not checked against allow list Patches The problem has been fixed with 6.4.20.1 with an improved override...

Traefik HTTP header parsing could cause a denial of service

Impact There is a vulnerability in Go when parsing the HTTP headers, which impacts Traefik. HTTP header parsing could allocate substantially more memory than required to hold the parsed headers. This behavior could be exploited to cause a denial of service. References - CVE-2023-24534 Patches -...

Undertow client not checking server identity presented by server certificate in https connections

The undertow client is not checking the server identity presented by the server certificate in https connections. This should be performed by default in https and in http/2...

KubeOperator allows unauthorized access to system API

Summary Unauthorized access refers to the ability to bypass the system's preset permission settings to access some API interfaces. The attack exploits a flaw in how online applications handle routing permissions. Affected Version = v3.16.3 Patches The vulnerability has been fixed in v3.16.3...

CRI-O vulnerable to /etc/passwd tampering resulting in Privilege Escalation

Impact It is possible to craft an environment variable with newlines to add entries to a container's /etc/passwd. It is possible to circumvent admission validation of username/UID by adding such an entry. Note: because the pod author is in control of the container's /etc/passwd, this is not...

Goa vulnerable to path traversal

Improper path santiziation in github.com/goadesign/goa before v3.0.9, v2.0.10, or v1.4.3 allow remote attackers to read files outside of the intended directory...

AWS S3 Crypto SDK sends an unencrypted hash of the plaintext alongside the ciphertext as a metadata field

The AWS S3 Crypto SDK sends an unencrypted hash of the plaintext alongside the ciphertext as a metadata field. This hash can be used to brute force the plaintext, if the hash is readable to the attacker. AWS now blocks this metadata field, but older SDK versions still send it...

jsonwebtoken's insecure implementation of key retrieval function could lead to Forgeable Public/Private Tokens from RSA to HMAC

Overview Versions =8.5.1 of jsonwebtoken library can be misconfigured so that passing a poorly implemented key retrieval function referring to the secretOrPublicKey argument from the readme link will result in incorrect verification of tokens. There is a possibility of using a different algorithm...

Jenkins Gitea Plugin vulnerable to Cleartext Transmission of Sensitive Information

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. Gitea Plugin 1.4.5 adds support for masking of Gitea personal access tokens. Administrators unable to update are...

GitPython vulnerable to Remote Code Execution due to improper user input validation

All versions of package gitpython are vulnerable to Remote Code Execution RCE due to improper user input validation, which makes it possible to inject a maliciously crafted remote URL into the clone command. Exploiting this vulnerability is possible because the library makes external calls to git...

crewjam/saml vulnerable to signature bypass via multiple Assertion elements due to improper authentication

Impact The crewjam/saml go library is vulnerable to an authentication bypass when processing SAML responses containing multiple Assertion elements. Patches This issue has been corrected in version 0.4.9. Credit This issue was reported by Felix Wilhelm from Google Project Zero...

OS Command Injection in Apache Airflow

Improper Neutralization of Special Elements used in an OS Command 'OS Command Injection' vulnerability in Apache Airflow Pinot Provider, Apache Airflow allows an attacker to control commands executed in the task execution context, without write access to DAG files. This issue affects Apache Airfl...

Untrusted code execution in Apache XML Graphics Batik

A vulnerability in Batik of Apache XML Graphics allows an attacker to run Java code from untrusted SVG via JavaScript. This issue affects Apache XML Graphics prior to 1.16. Users are recommended to upgrade to version 1.16...

SIF's Digital Signature Hash Algorithms Not Validated

Impact The github.com/sylabs/sif/v2/pkg/integrity package does not verify that the hash algorithms used are cryptographically secure when verifying digital signatures. Patches A patch is available in version = v2.8.1 of the module. Users are encouraged to upgrade. The patch is commit...

TensorFlow vulnerable to `CHECK` fail in `FakeQuantWithMinMaxVarsPerChannel`

Impact If FakeQuantWithMinMaxVarsPerChannel is given min or max tensors of a rank other than one, it results in a CHECK fail that can be used to trigger a denial of service attack. python import tensorflow as tf numbits = 8 narrowrange = False inputs = tf.constant0, shape=4, dtype=tf.float32 min ...

graphql-java vulnerable to Denial of Service via GraphQL query that consumes CPU resources

graphql-java before 19.0, 18.3, and 17.4 is vulnerable to Denial of Service. An attacker send a malicious GraphQL query that consumes CPU resources. The fixed versions are 19.0, 18.3, and 17.4...