Most API security programs are built around one assumption: if an API matters, it eventually shows up in traffic that crosses the…

30‑second summary:

Coding agents install dependencies, run unfamiliar code, start development servers, and execute tests. Giving them that freedom directly on…

A technical breakdown of a High-severity (CVSS 7.7) vulnerability in Claude Code’s macOS sandbox, and the parsing ambiguity behind it

What nine days of Claude Code changelog entries taught me about trusting an agent with my shell

A GitHub collaborator with only the triage role could shift Claude Code Action’s authorized trigger boundary and inject post-authorization…

A parent-directory swap timed against a permission check let sandboxed repo code redirect a host-owned file write. CVSS 7.7, High.

A beginner-friendly CTF tutorial: understand web requests, then change URL parameters, form fields, cookies, and HTTP methods in four local…

A Self-Healing WordPress Infection, and How I Cleaned It With Claude Code

As more applications are being developed, I’ve noticed a clear rise in Supabase usage. More and more teams are using it to handle…

TL;DR AIR Security disclosed Plugin4Shell, a zero-click remote code execution flaw in four AI coding agents. The bug defeats SHA pinning, the check meant to lock a plugin to reviewed code. Claude …

I watched this AI-powered tool explode on GitHub Trending this week. Here’s how it’s redefining vulnerability discovery for developers like…

How an agentic AI tool turned a public Langflow CVE into a working proof of concept, and why that should worry defenders

Artificial intelligence is transforming how cybersecurity professionals investigate software, analyze vulnerabilities, and understand…

How we built an internal AI-powered vulnerability research workflow — and what we learned about making AI useful for real-world penetration…

Linus Torvalds recently tasked the artificial intelligence tool Claude Code with authoring modifications for a vital Linux kernel subsystem. Consequently, this experiment yielded a threefold accel…

Most priced AI agent skills cost nothing to run. In Skillselion’s pricing census of October 16, 2026, 70.4% of the 4,902 skills that carry…

Noma Labs found a high-severity Claude Code Desktop flaw that let malicious HTML files steal local credentials and private keys through the preview pane.

A task-fit shortlist for binary reversing, source-led bounty work, and authorized API testing. I read the linked SKILL.md files; I have not…

TL;DR Zscaler ThreatLabz uncovered a global ARTEX AI hacking operation that targeted organizations across 20 countries and multiple industries. The suspected cybercriminal combined a modified ARTE…