Skip to content
Threat Feed

May 2026 (30)

high advisory

Velocity.js Prototype Pollution Vulnerability via #set Directive (CVE-2026-44966)

A prototype pollution vulnerability exists in Velocity.js versions 2.1.5 and earlier, allowing attackers to modify Object.prototype via crafted #set directives in Velocity templates, potentially leading to Denial of Service (DoS) or Remote Code Execution (RCE).

velocityjs <= 2.1.5 prototype-pollution vulnerability velocity.js CVE-2026-44966
2r 1t
high advisory

GitPython config_writer() Newline Injection Bypasses CVE-2026-42215 Patch

An incomplete patch for CVE-2026-42215 in GitPython allows newline injection in the section parameter of the config_writer() function, enabling arbitrary .git/config modification and remote code execution via core.hooksPath.

GitPython newline-injection rce code-injection
2r 3t 1c
critical advisory

Snipe-IT File Upload Vulnerability Leads to Remote Code Execution (CVE-2026-37709)

Snipe-IT versions prior to 8.4.1 are vulnerable to remote code execution due to insecure permissions on file uploads, where an attacker can upload arbitrary files and execute code on the server.

snipe-it remote code execution file upload insecure permissions asset management CVE-2026-37709
2r 1t 1c
medium threat

free5GC SMF Unauthenticated Process-Kill Denial-of-Service via UPI Endpoint

free5GC's SMF is vulnerable to an unauthenticated denial-of-service attack where a crafted POST request to the `/upi/v1/upNodesLinks` endpoint can trigger a `Fatalf` call, terminating the entire SMF process, effectively disrupting network services.

SMF free5GC DoS unauthenticated UPI CVE-2026-44321
2r 1t 1i
high advisory

MailEnable Enterprise Premium Authentication Bypass Vulnerability (CVE-2026-44400)

MailEnable Enterprise Premium 10.55 and earlier is vulnerable to CVE-2026-44400, an improper authorization vulnerability that allows attackers to bypass authentication checks and perform administrative actions by reusing AuthenticationToken cookies.

Enterprise Premium cve authentication-bypass privilege-escalation
2r 1t 1c
high advisory

Babel Plugin Vulnerability Leads to Arbitrary Code Execution via Malicious Input

A maliciously crafted input to Babel's `@babel/plugin-transform-modules-systemjs` or `@babel/preset-env` with `modules: 'systemjs'` can cause the tool to generate arbitrary code execution.

@babel/plugin-transform-modules-systemjs +1 code-generation arbitrary-code-execution babel
2r 1t
high advisory

Open WebUI /responses Endpoint Authentication Bypass Vulnerability

The /responses endpoint in Open WebUI's OpenAI router lacks access control, allowing authenticated users to bypass per-model access controls and interact with any configured model, potentially leading to denial of service, model theft, and access policy bypass.

open-webui authentication-bypass llm owasp
2r
critical advisory

Open WebUI LDAP Empty Password Authentication Bypass

Open WebUI is vulnerable to an LDAP authentication bypass where the LDAP authentication endpoint does not validate that the submitted password is non-empty before performing a Simple Bind against the LDAP server, potentially granting attackers complete account access.

open-webui authentication-bypass ldap
2r 1t
high advisory

fast-uri Host Confusion Vulnerability via Percent-Encoded Authority Delimiters (CVE-2026-6322)

The fast-uri library is vulnerable to host confusion due to improper handling of percent-encoded authority delimiters within the host component, potentially leading to redirection to unintended authorities.

fast-uri host-confusion url-parsing cve-2026-6322
2r 1t 1c
critical advisory

Electerm Arbitrary Code Execution via Crafted URI or CLI Arguments

Electerm versions 3.0.6 through 3.8.14 are vulnerable to arbitrary local code execution via crafted electerm:// URIs or command-line arguments, requiring a user to click a malicious link or open a malicious shortcut file.

Electerm code-execution protocol-handler
2r 1t 1c
high advisory

Electerm Remote Code Execution Vulnerability via Malicious Filenames

A remote code execution vulnerability exists in Electerm versions 3.7.8 and earlier, where a malicious SSH server can inject arbitrary commands into a victim's system by crafting filenames with shell metacharacters that are executed when the user attempts to open or edit the file using the 'open with system editor' or 'edit with custom editor' feature.

electerm rce sftp remote code execution
2r 1t 1c
medium advisory

Fastify accepts-serializer Denial of Service via Unbounded Accept Header Cache Growth

The @fastify/accepts-serializer package is vulnerable to a denial of service (DoS) attack due to unbounded cache growth, where an attacker can send many distinct Accept header variants, causing the cache to grow unbounded, exhausting the Node.js heap, and crashing the process.

@fastify/accepts-serializer dos denial-of-service fastify
2r 1t 1c
high advisory

ZITADEL LDAP Filter Injection Vulnerability in Login Flow

ZITADEL's LDAP identity provider implementation fails to properly escape user-provided usernames before incorporating them into LDAP search filters, allowing unauthenticated attackers to perform LDAP Filter Injection to enumerate usernames and extract sensitive attribute data.

ZITADEL +2 ldap-injection information-disclosure
2r 1i
medium advisory

Dronecode PX4 Autopilot MavlinkLogHandler Stack Buffer Overflow DoS (CVE-2026-32743)

A stack-based buffer overflow vulnerability exists in Dronecode PX4 Autopilot versions up to and including 1.17.0-rc2 that allows an attacker with MAVLink link access to cause a denial of service by creating a deeply nested directory via MAVLink FTP and then requesting the log list, crashing the MAVLink task.

Px4_Drone_Autopilot +1 px4 autopilot drone denial-of-service buffer-overflow
2r 3t 1c
high advisory

ex_webrtc Missing DTLS Fingerprint Validation Allows MITM

The ex_webrtc library is vulnerable to a man-in-the-middle attack due to missing DTLS peer certificate fingerprint validation in the DTLS client role, potentially allowing interception of media and data channels when chained with insecure signaling or a peer with similar validation gaps; upgrade to versions 0.15.1 or 0.16.1 to mitigate this vulnerability.

ex_webrtc +1 webrtc dtls mitm vulnerability
2r 1t
high advisory

n8n-mcp Vulnerable to Path Traversal, SSRF, and Telemetry Exposure

n8n-mcp versions before 2.50.1 are vulnerable to path traversal, redirect-following SSRF, and telemetry payload exposure, potentially leading to sensitive information disclosure and unauthorized access.

n8n-mcp path-traversal ssrf telemetry information-disclosure
2r 1t
high threat

Dirty Frag Linux Kernel Local Privilege Escalation Vulnerability

The Dirty Frag vulnerability (CVE-2026-43284 and CVE-2026-43500) is a Linux kernel local privilege escalation that allows an unprivileged local user to gain root privileges by exploiting flaws in the networking subsystem to overwrite protected file contents in the page cache.

exploited Linux kernel linux privilege-escalation vulnerability dirty_frag
2r 1t
medium advisory

Zebra Node Denial-of-Service Vulnerability via Crafted Orchard Transactions (CVE-2026-41584)

A crafted Orchard transaction with a zero-value rk field can cause a Zebra node to crash due to a panic in the orchard crate, leading to a denial-of-service condition; this vulnerability is identified as CVE-2026-41584 and patched in zebrad version 4.3.1 and zebra-chain version 6.0.2.

zebra-chain +1 denial-of-service zcash cryptography
1r 1t 1c
critical advisory

Compromised intercom-client npm Package Exfiltrates Credentials

A compromised version (7.0.4) of the intercom-client npm package was published using a compromised developer account, containing obfuscated JavaScript that executed during installation to harvest and exfiltrate credentials from the environment, as part of the 'Mini Shai-Hulud' supply chain campaign.

intercom-client +5 supply-chain credential-theft npm
2r 2t
critical threat

Compromised intercom-php Package on GitHub

A malicious commit tagged as version 5.0.2 was pushed to the intercom/intercom-php repository on GitHub, containing a Composer plugin that downloaded the Bun JavaScript runtime and executed an obfuscated credential-harvesting payload, targeting cloud provider credentials, environment variables, SSH keys, and CI/CD secrets.

intercom-php Mini Shai-Hulud supply-chain credential-theft github
2r 1t
medium advisory

Mozilla Firefox Multiple Vulnerabilities

Mozilla released security updates to address vulnerabilities in Firefox and Firefox ESR versions, potentially allowing for exploitation if left unpatched.

Firefox +1 vulnerability browser mozilla
2r 2t
critical advisory

Spring Cloud Config Vulnerabilities Allow Secret Access and Directory Traversal

Multiple vulnerabilities in Spring Cloud Config, including CVE-2026-40981, CVE-2026-40982, and CVE-2026-41002, could allow unauthorized access to secrets and directory traversal attacks, potentially leading to data exposure and system compromise.

Spring Cloud Config spring cloud config vulnerability directory_traversal secret_access
2r 3c
high advisory

TCLBanker Banking Trojan Self-Spreads via WhatsApp and Outlook

TCLBanker is a banking trojan targeting 59 financial platforms, spreading via trojanized Logitech AI Prompt Builder installers and worm modules for WhatsApp and Outlook, enabling remote control and data theft.

AI Prompt Builder +3 banking-trojan malware worm self-spreading brazil logitech
2r 8t
medium threat

Dronecode PX4-Autopilot tattu_can Stack Buffer Overflow (CVE-2026-32707)

A stack-based buffer overflow vulnerability exists in the `tattu_can` driver of Dronecode PX4-Autopilot versions 1.17.0-rc1 and earlier; by injecting specially crafted CAN frames, an attacker can trigger an unbounded memcpy operation, leading to a stack corruption and subsequent crash of the PX4 process, resulting in a denial of service.

PX4-Autopilot Mohammed Idrees Banyamer stack buffer overflow denial of service CVE-2026-32707
2r 1t 1c
critical advisory

Multiple Vulnerabilities in Ivanti Endpoint Manager Mobile

Multiple vulnerabilities in Ivanti Endpoint Manager Mobile allow an attacker to gain administrator privileges, execute arbitrary code with administrator privileges, bypass security measures, manipulate data, and disclose sensitive information.

Endpoint Manager Mobile vulnerability privilege-escalation execution
2r 4t
medium advisory

Ruby Multiple Vulnerabilities Lead to DoS and Information Disclosure

A remote, anonymous attacker can exploit multiple unspecified vulnerabilities in Ruby to perform a denial of service attack or disclose sensitive information.

Ruby dos information_disclosure vulnerability
2r 1t
high advisory

LiteLLM Vulnerability Allows Code Execution and Information Disclosure

A remote, authenticated attacker can exploit a vulnerability in LiteLLM to execute arbitrary program code and disclose sensitive information.

LiteLLM vulnerability code-execution
2r 2t
medium advisory

Microsoft 365 Copilot Business Chat Information Disclosure Vulnerabilities

Multiple vulnerabilities in Microsoft 365 Copilot Business Chat allow an anonymous remote attacker to disclose sensitive information.

Microsoft 365 Copilot Business Chat information-disclosure cloud microsoft365
2r 1t
critical advisory

Red Hat Build of Debezium for Red Hat Application Foundations Vulnerabilities Allow Code Execution

Multiple vulnerabilities in Red Hat Build of Debezium for Red Hat Application Foundations could allow an attacker to execute arbitrary code.

Build of Debezium for Red Hat Application Foundations vulnerability code-execution debezium
2r 1t
high advisory

IBM WebSphere Application Server Liberty Vulnerability Allows Code Execution

An authenticated remote attacker can exploit a vulnerability in IBM WebSphere Application Server Liberty to execute arbitrary program code on the target system.

WebSphere Application Server Liberty websphere rce code_execution vulnerability
2r 1t