Attack a CI/CD Pipeline: Build a Self-Hosted Runner Hijack Lab with a GitHub Actions Walkthrough

Build a hands-on CTF-style lab that shows how poisoned pull requests and persistent self-hosted runners lead to secret exfiltration, then harden with OIDC…

Continue ReadingAttack a CI/CD Pipeline: Build a Self-Hosted Runner Hijack Lab with a GitHub Actions Walkthrough

Tool Hijacking: The 2024 Papers That Predicted Agent Attacks

By November 2024, AI-agent security research had already documented the attack class that production incidents would later make infamous. InjecAgent (March 2024, ACL Findings) benchmarked 1,054 indirect-injection scenarios across 30 agents, finding ReAct-prompted GPT-4 attacked successfully roughly a quarter of the time. Breaking Agents (July 2024) demonstrated malfunction amplification through agentic loops. Together with 2023’s foundational indirect-prompt-injection work, they mapped how tools, descriptions, and fetched content become command channels. This survey walks the papers, the hijack taxonomy, and the controls that predate the incidents.

Continue ReadingTool Hijacking: The 2024 Papers That Predicted Agent Attacks

Sisense Breach: CI Credentials, AWS Keys and a CISA Advisory

On April 24, 2024, CISA and the FBI advised every Sisense customer to rotate credentials after attackers compromised the BI vendor’s development environment — and by week’s end, Sisense-issued AWS keys were circulating publicly. This piece reconstructs the five-day arc from detection to contained, explains why business-intelligence platforms are credential funnels that turn vendor CI/CD breaches into customer incidents, and extracts the third-party-risk doctrine the episode left behind for every embedded-analytics supply chain.

Continue ReadingSisense Breach: CI Credentials, AWS Keys and a CISA Advisory