At Natoma, we’ve been digging into how organizations are adopting the Model Context Protocol (MCP)— and the early findings are eye-opening. 

Across the companies we’re studying, we’re seeing an average of 225 MCP servers per organization already in use. That’s remarkable, considering we’re still in the very early days of MCP adoption. Today, MCPs are mostly being accessed through IDEs, meaning the early adopters concentrated in technical teams. But this is only the beginning. As tools and workflows across the enterprise start to plug into MCP servers, their footprint — and importance — will grow exponentially. 

Why This Matters 

Each MCP server represents a new integration point, a new surface area for productivity, and a new surface area for risk. While these servers enable developers to move faster, they’re often spun up and managed ad hoc — without centralized visibility or consistent security controls. That raises critical questions: 

• How many MCP servers are running inside your organization right now? 

• Who’s managing them, and how secure are they? 

• Could your teams move even faster if these servers were hosted and managed through a secure, unified hub? 

This is where our partnership with CrowdStrike’s AI-powered security platform comes in. 

A Partnership for Secure MCP Discovery and Management 

Natoma’s platform is now integrated with the AI-powered CrowdStrike Falcon® platform to bring MCP server discovery and security insights into one unified view. By combining Natoma’s AI infrastructure intelligence with the Falcon platform’s comprehensive threat detection, organizations gain: 

• Automated MCP server discovery across their environments 

• Risk visibility and threat context directly from the CrowdStrike Falcon platform 

• Centralized governance and hosting options to manage MCPs securely 

Together, Natoma and CrowdStrike are helping enterprises get ahead of the curve — bringing security, visibility, and scale to one of the fastest-emerging layers of AI infrastructure. 

Voices from the Field 

"We were stunned to see how many MCP servers had organically sprung up across our engineering teams. The Natoma + CrowdStrike integration gave us immediate visibility — not just into the number of servers, but into the real security posture behind them. With emerging threats like tool hijacking, where a malicious MCP could inject instructions an LLM might act on, this visibility is essential." — Head of IAM, Global Financial Services Company 

"For us, AI agents are already embedded in how we deliver personalized fitness experiences. But as our teams experimented with MCPs, the sprawl was becoming a blind spot. Connecting Natoma to CrowdStrike allowed us to regain control — protecting our AI workflows without slowing innovation." — VP of Engineering, Leading Fitness Technology Company 

The Road Ahead 

The growth curve of MCP usage will only accelerate. As AI agents, developer tools, and internal apps increasingly depend on model-aware context, the number of MCP endpoints will multiply across every function of the enterprise. Our mission at Natoma is to make this transition safe, observable, and productive — ensuring that as your organization embraces MCPs, you do so with the same rigor and confidence you apply to every other piece of core infrastructure. With CrowdStrike, we’re setting a new standard for secure AI connectivity — one that blends innovation with resilience. 

Stay tuned: We’ll be sharing more about this integration in the coming weeks, including early insights from organizations leading the way in secure MCP adoption.