Weaponizing MCP: From Chat Tool to Cloud Breach
How MCP Works (and Why It's a Big Attack Surface) MCP (Model Context Protocol) is a standard created by Anthropic for connecting AI models to external tools and data. Think of it as a universal plug s
Habib0x
Latest articles
Context Drift: How I Talked AI Agents Into Giving Up Their Secrets
I've been thinking a lot about how we talk to AI agents and what happens when the conversation goes long enough. Not in a theoretical sense -- I spent about 10 hours in a single session with Pulumi's Neo agent, and somewhere around hour three, someth...
Hacking Neo Pulumi's AI Agent.
Pulumi AI Agent Security Analysis
Building a Spec-Driven Development Plugin for Claude Code
I've been using Claude Code extensively, and one thing kept bothering me: jumping straight into implementation without proper planning. We've all been there—you start coding a feature, realize halfway through that you missed a requirement, then refac...
From 'Safe' AI Sandbox to Multi-Tenant Cloud Breach
A few weeks ago, I posted on LinkedIn about tricking a "secured" sandboxed agent into running arbitrary code with just a prompt. I framed it as a high-stakes game, and the system took the bait. No exploits, no payloads -- just some creative conversat...