Semgrep eventsJoin us for a fireside chat with Mudita Khurana—author of SoK: Measuring What Matters for Closed-Loop Security Agents and AppSec leader at Airbnb—as we explore how the next generation of security tools will think, learn, and improve on their own.
As companies move from manual security checks to smarter, automated systems, it’s becoming crucial to measure not just what these agents do, but how they make decisions and their effectiveness. This session breaks down why that shift matters and what it means for the future of AppSec.
Key Topics:
How the industry is moving toward automated “closed-loop” systems that can find, fix, and verify issues end-to-end.
What the research says about making effective agentic code security systems (e.g., planning, learning from mistakes, coordinating tools).
Explaining the Closed-Loop Capability (CLC) score - a benchmark designed to help the entire industry build better AppSec agents.
What the research shows about making effective agentic code security systems (e.g., planning, learning from mistakes, coordinating tools).
How this direction aligns with broader trends in AI-powered security and Semgrep’s own roadmap.