<- Back to home
SecurityAgentsMCP
Agents and safety
Security-focused meetup on MCP networking, agent security, and hardening real-world AI agents.
"Beyond tools in MCP: securing agents in the wild."
Speakers
Roberto Catalano
Beyond Tools in MCP
AWS
Max Mathys
Agent Security and Gandalf
Lakera
Marc Fischer
Securing Real-World AI Agents
Snyk
Overview
Three talks on agent security, prompt injection, and MCP hardening.
Talks
- Beyond Tools in MCP: A Security and Networking Perspective (Roberto Catalano, AWS).
- Agent Security and Gandalf - Insights from the World’s Largest Red Team (Max Mathys, Lakera).
- Securing Real-World AI Agents (Marc Fischer, Snyk).
Highlights
- Lessons from large-scale prompt injection and red teaming.
- Risks of autonomous agents connected to real systems.
- Practical defenses for securing agentic workflows.