UCL

Documentation

A hosted MCP gateway: aggregate any number of MCP servers and databases behind a single endpoint, add per-user OAuth (bring-your-own-account), governance (RBAC/ABAC, PHI/PII redaction, audit logging) and per-tool config, then connect Claude, Cursor, ChatGPT, or any AI agent.

Get started

Concepts

Guides

Reference

Frequently asked

Do I need to host anything?

No. It is a hosted gateway. You connect MCP servers and get a URL; there is nothing to deploy or run.

Can I connect more than one MCP server?

Yes. The gateway aggregates unlimited MCP servers and databases behind one endpoint and namespaces each tool so calls route to the right server.

What is the Model Context Protocol (MCP)?

MCP is an open protocol that lets AI applications connect to external tools and data sources through a standard interface. An MCP gateway aggregates many such servers behind one endpoint.

How is a gateway different from an iPaaS like Zapier?

An iPaaS runs predefined automations. An MCP gateway exposes live, callable tools to an AI agent at runtime, with per-call identity and governance - the agent decides what to call.

Where are per-user tokens stored?

Encrypted at rest in the gateway vault, keyed to the user's verified identity. Tokens are never returned to the LLM and are refreshed automatically.

Can a user revoke their connection?

Yes. Each user can disconnect any connector from their account, which deletes the stored credential immediately.