Adoption of agentic AI coding tools such as Cursor, Claude Code, Windsurf, and GitHub CoPilot is skyrocketing. While security teams have been scrambling to meet the challenges this creates in code vulnerabilities, there's a much bigger and more urgent risk at play - the new attack surface created by the complex and unsupervised interaction between AI agents, skills, MCPs and other components.
We've already seen how tools such as OpenClaw can break trust boundaries and create backdoors into developer endpoints, but even commercial tools with better security controls and permissions can be manipulated once extened via external tools using MCPs, hooks and skills.
Attend this webinar to learn:
-
What does the new ecosystem look like, and why it represents a huge blind spot for security teams
-
The threat surface it creates and what unvetted and unsupervised agentic AI dev tools and MCPs can do, from prompt injection to privilege escalation and RCE
-
Why the defaults and internal security controls provided by vendors such as Cursor are not sufficient
-
Which best practices and free tools can be used for discovery, visiblity and hardening of this new stack
Key Takeaways:
1. How vibe coding and agentic AI software development tools are creating a huge blind spot of "shadow AI"
2. The threats of prompt injection, data exfiltration and privilege escalation on developer workstations
3. What should security teams pay attention to when vetting and monitoring IDEs, Agent Skills and MCPs
4. Why endpoint security and MCP gateways are insufficient to protect the new stack
