Security update for perl-DBI

This update for perl-DBI fixes the following issues: CVE-2026-14380: unvalidated string eval interpolation of the Profile package name can lead to arbitrary Perl code execution bsc1271018. CVE-2026-14740: one-byte out-of-bounds read when deleting an initial SQL comment line can lead to a process...

The OpenAI and Anthropic AI Hacking Sprees Are a Messy New Legal Frontier

Both major AI labs’ models broke containment, escaped onto the internet, and hacked other companies. If a human had done that, the law would likely be against them. But a bot?...

CVE-2026-56671 ComfyUI: Path traversal in /experiment/models/preview allows arbitrary image file read

ComfyUI is a modular diffusion model GUI, api and backend with a graph/nodes interface. Prior to 0.28.0, getmodelpreview in app/modelmanager.py joins an unrestricted filename route capture to a selected model directory without a containment check, allowing an unauthenticated remote attacker to us...

Bending the Curve: Operational Cyber Epidemiology for Ransomware

Ransomware is often treated as a detection problem, but the most disruptive incidents behave more like outbreaks. A single foothold can spread through identities, administrative tools, and shared services while responders make time-critical decisions with incomplete visibility. This paper present...

Cyber-Capable AI Agents: Vulnerabilities, Evaluation Containment, and Defensive Response

Cyber-capable AI agents combine language models with tools, memory, and execution en- vironments to perform multi-step offensive-security tasks. Existing work separately measures cyber capability and catalogs attacks against agent components, but provides less guidance on containing a capable age...

Lessons from the OpenAI and Hugging Face Incident: When Safety Filters Disarm the Defender

In July 2026, an OpenAI model escaped its evaluation sandbox and broke into Hugging Face's production infrastructure. It is the first documented end-to-end intrusion carried out by an autonomous AI agent. The most repeated takeaway, "the AI went rogue," is also the least useful one. The real...

GHSA-9XWG-3R6F-JCX2 PyMdown Extensions: Path traversal in the b64 extension lets <img src> read files outside base_path

Summary The b64 extension inlines images referenced by as base64 data URIs. When resolving the src path it joins it onto the configured basepath with os.path.normpath and opens the result directly, with no check that the resolved path stays inside basepath. A src containing ../ sequences, or an...

AIDRS-Hybrid-Intrusion-Detection

AIDRS — AI-Powered Intrusion Detection & Response System A...

PraisonAI Code agent tools fail open without a workspace boundary

PraisonAI Code agent tools fail open without a workspace boundary SummaryPraisonAI Code's agent-compatible CODETOOLS wrappers keep a global workspace root initialized to None. If an application uses CODETOOLS, codereadfile, codesearchreplace, or codeapplydiff before calling setworkspace, the...

Inside the OpenAI – Hugging Face Incident: The AI Breach With No Human Attacker Behind It

OpenAI’s own models broke out of a test sandbox and into Hugging Face’s servers to solve an evaluation, with no human attacker involved. The incident showed how keeping agentic AI safe now depends on how it’s contained, not just on how it’s trained...

Why Modern SOCs Need Multi-Layered Detections

The cycle is over. For years, cybersecurity followed a familiar pattern: defenses improved, attackers adapted, and the back-and-forth continued. Today, AI-equipped attackers are simply outpacing defenses. Most intrusions now bypass endpoint and malware-based detection entirely. The CrowdStrike...

Remote Code Execution

PraisonAI is vulnerable to Remote Code Execution RCE. The vulnerability is due to missing path containment validation in SkillTools.runskillscript, where attacker-supplied absolute file paths are executed without restriction, allowing attackers to execute arbitrary scripts from any location on th...

PT-2026-67359

Impact Authenticated n8n users with workflow create/execute rights could use the Git node's fetch, pull, or push-tags operations to bypass the repository-path containment checks that already protected clone and push. By pointing an allowlisted remote configuration value at a local path outside th...

OpenAI Models Escaped Containment and Hacked Hugging Face

The cybersecurity-focused models, including GPT-5.6 Sol, broke out of a testing sandbox, exploited a zero-day, and gained access to the open internet to pull off the attack...

CVE-2026-58484

Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.12.2, EnvironmentManager.listBackups reads each backup's manifest.json and trusts the manifest's path field. EnvironmentManager.pruneBackups later passes that trusted entry.path directly to rmSyncentry.path, recursive...

CVE-2026-58484

Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.12.2, EnvironmentManager.listBackups reads each backup's manifest.json and trusts the manifest's path field. EnvironmentManager.pruneBackups later passes that trusted entry.path directly to rmSyncentry.path, recursive...

EUVD-2026-46034

Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.12.2, EnvironmentManager.listBackups reads each backup's manifest.json and trusts the manifest's path field. EnvironmentManager.pruneBackups later passes that trusted entry.path directly to rmSyncentry.path, recursive...

CVE-2026-58484

Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to v5.12.2, EnvironmentManager.listBackups() reads each backup’s _manifest.json and trusts the manifest’s path field; pruneBackups() then passes that path to rmSync(), enabling an attacker who can modify a manifest under data//.ba...

EUVD-2026-46018

Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.12.2, AgentRuntime promises scoped file access under a configured sandbox basePath, but its path containment checks use raw string prefix tests. A sandbox base such as /tmp/network-ai-sandbox also matches a sibling pa...

CVE-2026-58481

Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.12.2, AgentRuntime promises scoped file access under a configured sandbox basePath, but its path containment checks use raw string prefix tests. A sandbox base such as /tmp/network-ai-sandbox also matches a sibling pa...