● Covered by 1 source · 1 reportMedium impact1 neutral
Uber Develops MCP Gateway to Standardize AI Agent Access to 800+ Servers and 5,000+ Tools
Key points
- Uber built an MCP Gateway for AI agent service discovery and invocation.
- Platform supports over 800 servers and 5,000 internal tools.
- AutoCrawler automates tool discovery from IDL registries.
- Gateway enforces authorization and sensitive data redaction.
Standardizing AI Agent Interactions
Uber has developed an MCP Gateway to standardize how its AI agents interact with internal services. This initiative addresses challenges such as fragmented integrations, duplicated infrastructure, and inconsistent security practices that arose from the expanding adoption of AI agents within the company. The gateway provides a common layer for discovery, protocol translation, authorization, and execution for over 800 MCP servers and 5,000 tools.
Microservice Architecture for API Exposure
The MCP Gateway operates as a microservice architecture, exposing existing APIs as MCP tools without requiring modifications to the underlying services. Its design separates the MCP Registry, which functions as the control plane, from the Proxy Gateway, responsible for runtime execution. The registry stores tool definitions, schemas, ownership, and enablement settings, while the proxy translates MCP requests into HTTP, gRPC, or TChannel calls and converts responses back to an MCP-compatible format.
Automated Tool Discovery with AutoCrawler
Uber's AutoCrawler automates the discovery of tools within the system. It scans internal interface definition language (IDL) registries, extracting API definitions from Protocol Buffers and Thrift schemas. The AutoCrawler then generates MCP-compatible schemas and uses a large language model to create tool descriptions. Newly discovered tools remain disabled until their owning teams review and explicitly enable them, and configuration changes propagate without requiring service restarts.
Security and Data Governance
The gateway implements authorization at both the server and tool levels, leveraging Uber's internal access control system. It also supports the redaction of personally identifiable information (PII) and other sensitive data from tool responses. For integrations with third-party services like Jira and Google, a separate service manages the exchange of internal user tokens.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Reporting from
Uber has created an MCP Gateway to standardize how its internal AI agents discover and invoke services, addressing fragmentation and inconsistent security across its growing AI agent ecosystem. This platform manages over 800 servers and 5,000 tools, providing a unified layer for discovery, protocol translation, authorization, and execution. The system uses an AutoCrawler for automated tool discovery and integrates with Uber's access control for security.