Agentic AI Security The New .env: Measuring Credential Leakage in AI Agent Instruction Files Developers have quietly opened a new place to hard-code their own live secrets: the Markdown files they write to steer their AI coding agents. We measured how often, which files, and why. Zvika Babo & Gabi Nakibly |August 05, 2026
Application Security The AI Is the Phisher: How Prompt Injection Turns Trusted Assistants into Personalized Social Engineers Traditional phishing begins with an unsolicited message. An attacker sends an email or text message and hopes that the recipient will open it, trust it, and take the requested action. Most phishing campaigns are therefore a numbers game: send enough messages, and eventually someone will click. Zvika Babo & Gabi Nakibly |September 03, 2026
Threat Intelligence Inside a GitHub Supply Chain Attack Campaign: Command Injection, Attempts to Exfiltrate CI/CD Data, and Evidence of Execution As part of security research conducted at Radware, we uncovered a coordinated attack campaign targeting public GitHub repositories through Issues, Pull Requests, and comments. Zvika Babo & Gabi Nakibly |October 29, 2026
Agentic AI Security Who Is Prompt-Injecting Coding Agents? AI coding agents now work on GitHub every day: they sort bug reports, review code changes and pick up tasks on their own. To do that, they read text that anyone on the internet can write. Zvika Babo & Gabi Nakibly |November 05, 2026