AI Coding Agents Found Triggering Endpoint Security Rules Built to Catch Attackers

Sophos looked at a week of its own endpoint data and found that AI coding agents such as Claude Code, Cursor, and OpenAI Codex are setting off detection rules written to catch human intruders. The agents are not malicious. They just do a lot of things that, to a behavioral engine, look exactly li...

Beware What You Autocomplete: Forensic Attribution of Backdoored Code Completions

Large language models have enabled powerful code completion systems that assist developers by predicting subsequent lines of code. However, these models remain vulnerable to backdoor attacks, where malicious fine-tuning data covertly implants unsafe behaviors. Despite advances in defensive...

arkindex-client (>=0.1.0, <=0.1.1), arlo (>=1.2.49, <=1.2.64) +18 more potentially affected by CVE-2025-50181 via urllib3 (1.24.0)

urllib3 PYPI version 1.24.0 is affected by a known vulnerability. The following packages have a transitive dependency on urllib3 and may be impacted: - arkindex-client =0.1.0, =1.2.49, =3.0.0, =0.6.2, =1.0.3, =0.1.2.15, =3.1.3, =0.3.9, =1.5.6, =0.10.0, =1.2.0, =1.2.4 and more Source cves:...

Cloak and Detonate: Scanner Evasion and Dynamic Detection of Agent Skill Malware

LLM coding agents increasingly rely on third-party agent skills from public marketplaces, which execute with the agent's privileges and create a software supply-chain attack surface: a malicious skill can steal credentials, exfiltrate source code, or install backdoors. Existing defenses use stati...

EUVD-2025-210376

IBM watsonx.data intelligence 5.2.0, 5.2.1, 5.2.2, 5.3.0 could allow an authenticated user to perform unauthorized actions due to the improper enforcement of behavioral workflow...

CVE-2025-36333 Vulnerabilities found in Watson Data Intelligence

IBM watsonx.data intelligence 5.2.0, 5.2.1, 5.2.2, 5.3.0 could allow an authenticated user to perform unauthorized actions due to the improper enforcement of behavioral workflow...

The Realities of AI Video Surveillance

The Financial Times has a good article on how AI is changing the capabilities of video surveillance, with information from both Israel/Iran and Russia. I wrote about this sort of thing a few years ago, how AI enables mass spying in the way that computers and networks enabled mass surveillance. Th...

CVE-2025-36333: Improper Enforcement of Behavioral Workflow

IBM watsonx.data intelligence 5.2.0, 5.2.1, 5.2.2, 5.3.0 could allow an authenticated user to perform unauthorized actions due to the improper enforcement of behavioral workflow...

PT-2026-53981

Name of the Vulnerable Software and Affected Versions IBM watsonx.data intelligence versions 5.2.0 through 5.3.0 Description An authenticated user can perform unauthorized actions because of improper enforcement of the behavioral workflow. Recommendations At the moment, there is no information...

An AI-Based Solution for Secure Service Provisioning in IoT

As the Internet of Things IoT continues its rapid expansion, the attack surface grows accordingly, with emerging threats targeting smart objects and their interactions. In this evolving landscape, securing service provisioning is crucial to ensure the proper functioning, security, and reliability...

Improper Enforcement of Behavioral Workflow

Overview filament/filament is an A collection of full-stack components for accelerated Laravel app development. Affected versions of this package are vulnerable to Improper Enforcement of Behavioral Workflow through the improper handling of recovery codes in app-based multi-factor authentication...

Turn specs into evals for any agent with ASSERT

Today, we’re releasing Adaptive Spec-driven Scoring for Evaluation and Regression Testing ASSERT, an open-source framework for turning natural-language behavior specifications into executable evaluations. Every team building an AI system starts with a clear intention for the behaviors they want t...

The Human Vulnerabilities and Exploits (HVE) Framework

The cybersecurity community has invested over two decades in building standardized frameworks, the Common Vulnerabilities and Exposures CVE system, the Common Vulnerability Scoring System CVSS, and the Common Weakness Enumeration CWE to identify, classify, and remediate threats to digital...

An AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response

University Academic Management Information Systems ACMIS are high-value targets for a wide spectrum of security threats including brute-force login attacks, payment fraud, privilege escalation, insider data theft, and academic integrity violations. Traditional rule-based intrusion detection syste...

MalTree: Tracing Malware Evolution from Embeddings at Scale

Malware detection remains largely reactive: machine learning models trained on known samples degrade as threats evolve. Understanding evolutionary relationships among malware families can inform proactive defense, but traditional reverse engineering can take months to years to uncover such lineag...

Token-Level Generalization in LoRA Adapter Backdoors: Attack Characterization and Behavioral Detection

We show that LoRA adapters, the dominant distribution format for fine-tuned LLMs, can be reliably backdoored through training data poisoning while preserving baseline task performance. On a Qwen 2.5 1.5B prompt-injection classifier, a small fraction of poisoned examples drives a...

Analyzing Concentration, Temporal Routines and Targeting in Public Ransomware Leak Site Data

Ransomware has grown to become one of the most damaging types of cybercrime, affecting private and public organizations in any sector. While early types of ransomware targeted many victims via automated attacks, ransomware groups have started to specifically target organizations and companies in...

Profiling User Vulnerability to Phishing through Psychological and Behavioral Factors

Phishing remains one of the most pervasive cybersecurity threats, shifting the focus from technological vulnerabilities to human cognitive and psychological factors. In coherence with the trend of studies on phishing to increasingly focus on human aspects and vulnerable users profiling, this stud...

Detecting Data Exfiltration through I2P Anonymity Networks: A Two-Phase Machine Learning Approach

The Invisible Internet Project I2P provides strong anonymity through garlic routing and distributed network architecture, making it attractive for legitimate privacy needs. Nevertheless, the same properties can be exploited by malicious actors to steal sensitive information from corporate network...

Behavioral Integrity Verification for AI Agent Skills

Agent skills extend LLM agents with privileged third-party capabilities such as filesystem access, credentials, network calls, and shell execution. Existing safety work catches malicious prompts and risky runtime actions, but the skill artifact itself goes unverified. We formalize this as the...