{
  "id": 17803718,
  "title": "Uber Builds MCP Gateway to Govern Access to 800+ Servers and 5,000+ AI Agent Tools",
  "url": "https://urgent.news/2026/11/02/uber-builds-mcp-gateway-to-govern-access-to-800-servers-and-5-000-ai",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-02T13:53:00.000Z",
  "source": {
    "name": "InfoQ",
    "slug": "infoq",
    "url": "https://www.infoq.com/news/2026/11/uber-mcp-gateway-ai-agents/"
  },
  "original_language": "en",
  "account": "Uber has developed an MCP Gateway to streamline the process of AI agents accessing internal services within the company. This platform manages over 800 MCP servers and 5,000 tools, providing a unified layer for discovery, protocol translation, authorization, and execution. The system separates the MCP Registry, which functions as the control plane, from the Proxy Gateway, which handles runtime execution. The registry maintains details about tool definitions, schemas, ownership, and enablement settings, while the Proxy Gateway translates MCP requests into HTTP, gRPC, or TChannel calls and converts the responses back into MCP-compatible results. Analysts have noted that Uber's MCP Gateway resembles established API gateway patterns, offering a structured approach to navigating the complexity often associated with large language models (LLMs). Uber's AutoCrawler automates tool discovery through Cadence by scanning internal interface definition language (IDL) registries and extracting API definitions from Protocol Buffers and Thrift schemas. New tools are initially disabled until their respective teams review and explicitly enable them. Configuration updates are automatically applied to the data plane without necessitating service restarts or redeployments. The gateway ensures security and governance through authorization at both the server and tool levels, as well as redaction of sensitive information like personally identifiable data in responses. For third-party integrations, such as Jira and Google, Uber employs a separate service that converts internal user tokens into compatible third-party authentication tokens. To manage the growing catalog of tools, Uber introduced Omni MCP, which enables incremental discovery, allowing agents to search for servers, discover relevant tools, retrieve individual schemas, and invoke specific tools as needed. Additionally, the platform supports response projection, enabling agents to request specific response fields rather than receiving full payloads. Uber also provides an aifx command-line interface that allows agents to discover and invoke tools, storing the results in files for selective inspection.",
  "summary": "Uber has built an MCP Gateway to standardize how AI agents discover and invoke internal services across more than 800 MCP servers and 5,000 tools. The gateway separates tool registration from runtime execution and includes automated API discovery, access controls, sensitive data redaction, incremental tool discovery through Omni MCP, and response filtering. By Leela Kumili",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}