25 articles
A wire-level analysis found Grok Build 0.2.93 uploading tracked source code and full Git history independently of what the agent read. Here is what was proven, what changed, and how developers should respond.
Fast takedowns do not protect systems that auto-install malicious packages or extensions in the first minutes after release. Minimum package age turns time into a practical supply chain defense.
ChocoPoC hides a remote access trojan inside trojanized CVE proof-of-concept repositories on GitHub, using a malicious PyPI dependency chain to compromise the researchers who clone them.
A 0DIN proof of concept against Claude Code demonstrates how a clean-looking repository can lead to runtime command execution. The structural risk behind the attack applies to any AI coding agent with shell access.
When DuckDuckGo's AI killed Trump with rabies, the world laughed. When AI coding assistants invent package names, attackers register them. Nobody's laughing then.
Novee's Cordyceps research is a reminder that GitHub Actions workflows are executable attack surface, not harmless YAML. Here is how to audit the trust boundary before an outside pull request borrows maintainer authority.
A practical home-user checklist for auditing MCP servers, AI assistant tools, local permissions, and supply-chain risk before a trusted setup turns into an exposed one.
Mini Shai-Hulud and Miasma show how supply chain malware can move from npm install-time execution into Claude Code hooks, VS Code tasks, and CI/CD persistence.
A fake OpenAI repo hit #1 trending on Hugging Face with 244K downloads in 18 hours. Here's every attack vector targeting AI model repositories — and how to defend against them.
Trend Micro documented QLNX, a Linux RAT that combines credential harvesting, LD_PRELOAD persistence, PAM backdoors, and rootkit behavior. The real risk is not one infected host - it is the supply chain access behind it.
npm packages no longer publish instantly. GitHub's staged publishing forces a 2FA-gated human approval before any version hits the registry — here's what it means and how to enable it.
GitHub says an employee device was compromised through a poisoned third-party VS Code extension and internal repositories were exfiltrated. Here is the fact-checked breakdown for defenders.
TeamPCP has compromised hundreds of open-source packages and stolen half a million credentials. But their OPSEC is leaking — and someone is already hunting them.
TeamPCP's Shai-Hulud is a TypeScript/Bun C2 framework targeting GitHub Actions CI/CD pipelines — it steals GitHub tokens, exfiltrates via a fake git domain, and has now been open-sourced for anyone to deploy.
How attackers turn GitHub Actions' shared build cache into a supply chain weapon — real cases, attack mechanics, detection logic, and mitigations.
Autonomous AI agents are already inside enterprise environments — and most security teams have no idea what they're doing. Here's what attackers exploit and how to defend against it.
840,000 GhostPoster victims, 3.2M+ in GitLab campaign, 4.3M+ in ShadyPanda—browser vendors removed extensions but never told users. Self-regulation failed.
Your CI/CD pipeline stores production credentials, runs code automatically, and trusts pull requests. Here's how attackers exploit that — and the detection logic to catch them.
65% of Forbes AI 50 companies leaked secrets on GitHub with 94-day median remediation time. Blue team guide to detect, prevent, and respond to repository leaks.
Unicode's invisible characters are being weaponized — hiding malicious code in repositories, hijacking AI agents, and bypassing security reviews without leaving a trace visible to human eyes.
MCP servers let AI assistants control your tools — but most users install them without understanding the attack surface. Here's what attackers already know.
On March 31, 2026, a trusted npm package with 400 million monthly downloads was backdoored for three hours. Here's how it worked and why it keeps happening.
OpenClaw went from 0 to 180,000 GitHub stars in weeks — and then came the RCE, 30,000 exposed instances, and a supply chain attack poisoning its entire skill marketplace.
Discover which real-world cyberattacks Zero Trust prevents—and which ones it doesn't. Analyzed through 2025-2026 breach data including ransomware campaigns, insider threats, supply chain compromises, and social engineering attacks.
Millions run shared Colab notebooks without reading them. Here's what that actually costs you — from Google Drive exfiltration to OAuth token theft and supply chain attacks.