Uber Builds MCP Gateway to Govern Access to 800+ Servers and 5,000+ AI Agent Tools

| Source: InfoQ AI/ML

Tags: Uber, Model Context Protocol, MCP Gateway, gRPC, Cadence, AI agents

Uber built an MCP Gateway that lets AI agents discover and call internal services across more than 800 MCP servers and 5,000 tools, separating a registry control plane from a runtime proxy.

Details

Uber has built an MCP Gateway to standardize how AI agents find and invoke internal services, aiming to fix fragmented integrations, duplicated infrastructure and inconsistent security as agent use grows. The platform covers more than 800 MCP servers and 5,000 tools. An MCP Registry acts as the control plane, holding tool definitions, schemas, ownership and enablement settings. A Proxy Gateway handles runtime, translating MCP requests into HTTP, gRPC or TChannel calls and converting the responses back. Existing APIs are exposed as MCP tools without changes to the underlying services, and an AutoCrawler built on Cadence automates tool discovery. Engineers from Microsoft and Pipe17 noted the design closely mirrors established API gateway patterns.