Runlayer

Runlayer provides enterprise security and governance infrastructure for the Model Context Protocol (MCP), enabling organizations to safely deploy AI agents that interact with internal tools and data. The platform addresses the security gaps in MCP by adding authentication, authorization, audit trails, and governance controls that were absent from the protocol's initial launch.

The company's platform allows enterprises to create, host, and scale MCP servers across their organization with centralized security controls. It integrates with identity providers like Okta and Entra to enforce conditional access and device compliance checks, treating AI agents as enterprise applications subject to the same security policies as other systems.

The platform provides complete observability and audit trails for all AI agent interactions with tools and data.

Runlayer supports all MCP clients including Cursor, VS Code, Claude Code, GitHub Copilot, ChatGPT, Claude Desktop, and Windsurf. Organizations can deploy both local and remote MCP servers through the platform, with security-approved servers available immediately through a catalog system. The platform enables teams to convert internal APIs into MCP servers that appear alongside external ones with identical access controls.

The company was founded by former Zapier executives who built AI products at scale, including the Zapier MCP implementation. The founding team includes Andy Berman (former Director of AI at Zapier), Tal Peretz (who built Zapier MCP), and Vitor Balocco (recognized MCP security expert). Runlayer is backed by Khosla Ventures, Felicis, and SVCI, with advisors including the co-creator of MCP from Anthropic.