SHIELDS: Automating OS Hardening with Iterative Multi-Agent Remediation

Security misconfigurations remain a leading cause of OS-level compromise, and manually keeping systems compliant with standards like Defense Information Systems Agency DISA Security Technical Implementation Guides STIGs is a tedious and expensive process. Existing compliance automation tools can...

CVE-2026-10717

Out of bounds write and reads in openSeaChest’s --showSCSIDefects in Seagate’s openSeaChest v25.05.3 on all supported platforms allows for writing defect information out of bounds for very large defects lists via a very bad drive with lots of defects or a maliciously crafted SCSI device’s defect...

php: signed integer overflow in metaphone()

A flaw was found in PHP. The metaphone function in ext/standard/metaphone.c uses a signed int variable to track the current position within the input string. When an input string is longer than 2,147,483,647 bytes, a signed integer overflow can occur, leading to undefined behavior and an...

CVE-2026-37228

FlexRIC v2.0.0 contains a reachable assertion in e2aprecvsctpmsg src/lib/ep/e2apep.c. The function allocates a fixed 32KB receive buffer and enforces assertrc = 32,768 bytes to crash the near-RT RIC, iApp, E2 Agent, or xApp process via SIGABRT. No valid E2AP PDU is required. All four SCTP endpoin...

brace-expansion: brace-expansion: Denial of Service due to excessive memory allocation when expanding large numeric ranges

A flaw was found in the brace-expansion library. This vulnerability allows an attacker to cause a Denial of Service DoS by providing a large numeric range for expansion. The library allocates excessive memory to generate all intermediate elements before applying the maximum limit, leading to high...

EUVD-2026-33797

In applySimpleFieldMaxSize of DataRowHandler.java, there is a possible way to insert a large contact name due to improper input validation. This could lead to local denial of service with no additional execution privileges needed. User interaction is not needed for exploitation...

From Untrusted Input to Trusted Memory: A Systematic Study of Memory Poisoning Attacks in LLM Agents

Memory is a core component of AI agents, enabling them to accumulate knowledge across interactions and improve performance. However, persistent memory introduces the risk of memory poisoning, where a single adversarial memory write can exert long-term influence over agent behavior. We present a...

Backdoor Unlearning Generalization: A Path toward the Removal of Unknown Triggers in LLMs

Backdoor attacks in Large Language Models LLMs are a growing security concern, where models can generate adversary-chosen content. Existing defenses target backdoors one at a time and typically require knowledge of the trigger, leaving the defender at a structural disadvantage when unknown...

CVE-2026-0085

In applySimpleFieldMaxSize of DataRowHandler.java, there is a possible way to insert a large contact name due to improper input validation. This could lead to local denial of service with no additional execution privileges needed. User interaction is not needed for exploitation...

CVE-2026-0085

CVE-2026-0085 relates to a vulnerability in the DataRowHandler.java function applySimpleFieldMaxSize , where improper input validation could allow inserting an unusually large contact name. The result is a local Denial of Service with no extra privileges and no user interaction required. Document...

-cascade-scan

cascade-scan AI Agent security evaluation framework — autom...

Linux Distros Unpatched Vulnerability : CVE-2026-45149

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - The brace-expansion library generates arbitrary strings containing a common prefix and suffix. From 5.0.0 to before 5.0.6, the max option was being applied too...

CVE-2026-37228

FlexRIC v2.0.0 contains a reachable assertion in e2aprecvsctpmsg src/lib/ep/e2apep.c. The function allocates a fixed 32KB receive buffer and enforces assertrc = 32,768 bytes to crash the near-RT RIC, iApp, E2 Agent, or xApp process via SIGABRT. No valid E2AP PDU is required. All four SCTP endpoin...

PT-2026-45591

In applySimpleFieldMaxSize of DataRowHandler.java, there is a possible way to insert a large contact name due to improper input validation. This could lead to local denial of service with no additional execution privileges needed. User interaction is not needed for exploitation...

Cross-Vendor Sola ISPM Benchmark: Evaluating Agentic AI for Federated Identity Security Reasoning

The rapid proliferation of multi-cloud and SaaS platforms has transformed Identity Security Posture Management ISPM into a fundamentally cross-vendor challenge: critical misconfigurations and privilege escalation paths increasingly span multiple identity providers, infrastructure layers, and...

SUSE CVE-2026-44390

NLnet Labs Unbound up to and including version 1.25.0 has a vulnerability when handling replies with very large RRsets that Unbound needs to perform name compression for. Malicious upstream responses with very large RRsets with records that don't share a suffix above the root can cause Unbound to...

SUSE CVE-2026-48156

pypdf is a free and open-source pure-python PDF library. Prior to 6.12.0, an attacker who uses this vulnerability can craft a PDF which leads to long runtimes. This requires cross-reference streams with /W 0 0 0 values and large /Size values. This vulnerability is fixed in 6.12.0...

SUSE CVE-2026-48735

pypdf is a free and open-source pure-python PDF library. Prior to 6.12.1, an attacker who uses this vulnerability can craft a PDF which leads to large memory usage. This requires parsing large XMP metadata, possibly with lots of unnecessary elements. This vulnerability is fixed in 6.12.1...

NeuroLog: Reasoning You Can Audit -- Neuro-Symbolic Vulnerability Discovery Via LLM Facts, Datalog, and SMT

Vulnerability discovery on C/C++ source asks the analyst to choose between heavyweight static analysers, which need a working build before a single query runs, and free-form LLMs, which read source readily but invent details and lose track of cross-function dataflow on real codebases. We present...

Infinite loop

Overview Affected versions of this package are vulnerable to Infinite loop through the Avro Decoder process. An attacker can exhaust CPU resources by providing a specially crafted payload with a large block-count value, causing the decoder to perform excessive iterations before propagating an...