Towards Cybersecurity SuperIntelligence (CSI): What'S the Best Harness for Cybersecurity?

What is the best harness for cybersecurity AI? Cybersecurity systems are converging on a single execution scaffold per agent, an iterative shell loop driven by a Large Language Model LLM. However, scaffolds are not interchangeable, rarely interoperable, and no single scaffold dominates across all...

The Importance of Out-Of-Band Metadata for Safe Autonomous Agents: The Redpanda Agentic Data Plane

AI agents are increasingly expected to operate as digital employees: accessing enterprise data, making decisions, and taking actions autonomously. But agents are simultaneously less predictable than humans -- prone to hallucination, misinterpretation, and adversarial manipulation -- and more...

RHSA-2026:20588 Red Hat Security Advisory: fence-agents security update

Bulletin has no description...

Important: Red Hat Security Advisory: fence-agents security update

An update for fence-agents is now available for Red Hat Enterprise Linux 8.8 Update Services for SAP Solutions and Red Hat Enterprise Linux 8.8 Telecommunications Update Service. Red Hat Product Security has rated this update as having a security impact of Important. A Common Vulnerability Scorin...

MRMMIA: Membership Inference Attacks on Memory in Chat Agents

Membership inference attacks MIAs test whether a target data record belongs to a system's private data, and have become a standard tool to measure privacy leakage in machine learning systems. Prior work has primarily focused on training corpora or retrieval databases. However, MIAs against agent...

RHEL 8 : fence-agents (RHSA-2026:20588)

The remote Redhat Enterprise Linux 8 host has packages installed that are affected by a vulnerability as referenced in the RHSA-2026:20588 advisory. The fence-agents packages provide a collection of scripts for handling remote power management for cluster devices. They allow failed or unreachable...

Lessons from Penetration Tests on Large-Scale Agent Systems

As AI systems gain increasing autonomy and execution capability, the number of discovered security vulnerabilities continues to rise. However, many of these vulnerabilities are not fundamentally novel, but instead reflect recurring classes of weaknesses long observed in prior computing systems...

azure-ai-generative (>=1.0.0b1 <=1.0.0b3), azure-ai-resources (>=1.0.0b1 <=1.0.0b9) +15 more potentially affected by CVE-2026-2651 via mlflow-skinny (>=3.0.0 <=3.0.1)

mlflow-skinny PYPI version =3.0.0, =1.0.0b1, =1.0.0b1, =0.1.0, =0.1.0, =2.5.0, =0.0.13, =3.0.0, =0.1.0, =0.1.4 and more Source cves: CVE-2026-2651 Source advisory: SNYK:PYTHON-MLFLOWSKINNY-16874026...

databricks-agents (>=0.1.0 <=1.0.0rc1), datamint (>=2.5.0 <=2.5.2) +3 more potentially affected by CVE-2026-2651 via mlflow (>=3.0.0rc2 <=3.0.1)

mlflow PYPI version =3.0.0rc2, =0.1.0, =2.5.0, =0.2.0.dev0, =0.6.7, =0.8.1 Source cves: CVE-2026-2651 Source advisory: SNYK:PYTHON-MLFLOW-16874027...

Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures

The rapid evolution of large language model LLM-driven autonomous agents has given rise to OpenClaw, a new class of open-source agent frameworks that operate as continuously running, skill-augmented systems with persistent memory, multi-channel interaction, and high degrees of autonomy. Such...

Malicious code in get-package-lint (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 383f22ab2e1e8bbb44a44fa3828710f476947837d0b38aa9266eafcbf9959261 Package name typosquats the popular get-package-type and reuses its README/exports verbatim, but adds "postinstall": "node utils.cjs" in package.json...

MAL-2026-4572 Malicious code in get-package-lint (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 383f22ab2e1e8bbb44a44fa3828710f476947837d0b38aa9266eafcbf9959261 Package name typosquats the popular get-package-type and reuses its README/exports verbatim, but adds "postinstall": "node utils.cjs" in package.json...

Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks

We evaluate whether frontier LLMs are ready for cybersecurity through a dual-mode benchmark: white-box function-level vulnerability detection VulnLLM-R, across C/Java/Python and black-box web application security testing five production-style applications with 118 ground-truth vulnerabilities...

CVE-2026-34926

A directory traversal vulnerability in the Apex One on-premise server could allow a pre-authenticated local attacker to modify a key table on the server to inject malicious code to deploy to agents on affected installations. This vulnerability is only exploitable on the on-premise version of Apex...

CVE-2026-34926

A directory traversal vulnerability in the Apex One on-premise server could allow a pre-authenticated local attacker to modify a key table on the server to inject malicious code to deploy to agents on affected installations. This vulnerability is only exploitable on the on-premise version of Apex...

CVE-2026-34926

A directory traversal vulnerability in the Apex One on-premise server could allow a pre-authenticated local attacker to modify a key table on the server to inject malicious code to deploy to agents on affected installations. This vulnerability is only exploitable on the on-premise version of Apex...

Researchers left AI agents alone in a virtual town and watched it all unravel

Tech leaders have spent the past year telling everyone that AI agents are about to run financial systems, file your tax returns, and quietly buy your groceries. Just leave them alone, the rhetoric goes; they'll handle it. But a New York startup left ten of them alone in a virtual town for two...

Beyond Zero: Enterprise Security for the AI Era

The rise of autonomous AI agents and the accelerating velocity of corporate data access are stretching the application-centric model of zero trust security to its breaking point. This paper introduces Beyond Zero, a new security paradigm designed for the AI era. The Beyond Zero architecture...

PT-2026-42465

Name of the Vulnerable Software and Affected Versions Apex One on-premise versions prior to SP1 Build 18012 Apex One new installs versions prior to 17079 Apex One SaaS agent versions prior to 14.0.20731 Description A directory traversal issue in the on-premise management server allows an attacker...

VulnCheck KEV: CVE-2026-34926

A directory traversal vulnerability in the Apex One on-premise server could allow a pre-authenticated local attacker to modify a key table on the server to inject malicious code to deploy to agents on affected installations. This vulnerability is only exploitable on the on-premise version of Apex...