Chris Lentricchia

How Sweet Secures MCP Servers: Live Discovery, Cloud Context and Runtime Enforcement

Sweet discovers MCP servers from live traffic, maps their cloud context and blocks unsafe AI agent actions in runtime.

“Do Something!!” Why Observability Isn’t Security

Observability shows what’s happening. Sweet uses runtime intelligence to attack, fix and defend cloud and AI environments in real time.

When the Test Becomes the Breach: Why AI Security Now Depends on Runtime Enforcement

AI agents can cross intended boundaries. Runtime enforcement helps stop unsafe actions as they happen.

The Rise of Policy Debt And How Sweet Changes the Conversation

How Long Does It Take You to Respond to an Agentic Breach?

AI-powered attacks move at machine speed. See why real-time runtime enforcement is critical to stopping agentic breaches as they happen.

Prompt Injection Doesn't End at the Prompt

Prompt injection begins with an instruction, but the damage rarely ends there. Most defenses focus on preventing malicious instructions from

Sweet Security Enables Organizations to Securely Run Workloads on GKE Autopilot

Sweet Security is now whitelisted for GKE Autopilot, bringing runtime security and workload protection to Google’s managed Kubernetes platfo

Sweet Enables Organizations to Stay Ahead of AI-Speed Attacks with Sweet Attack

Sweet Attack continuously validates exploitable cloud and AI attack paths so teams can stop breaches before attackers do.