MCPEasy

Turns any REST API into a hosted remote MCP server. Import an OpenAPI spec or Postman collection, and MCPEasy groups the endpoints into a few clear tools, marks each one read only, write or destructive, and gives you a hosted endpoint with API keys and rate limits for Claude, ChatGPT, Cursor or any MCP client.

Documentation

Works with any client that supports MCP

From spec to live endpoint in four steps

  1. Connect Paste an OpenAPI URL, upload a spec file or drop in a Postman collection. Nothing to install.
  2. Generate MCPEasy groups endpoints into a small set of clear tools and writes descriptions agents understand.
  3. Test Chat with your tools in the playground and see every call, argument and response before going live.
  4. Deploy Get a hosted MCP endpoint with API keys and rate limits. Paste it into any AI client.

Everything an MCP server needs, already built

Hosting, auth, logging and testing come included, so your team only decides what agents may do.

Edit every tool before it ships

Rename tools, rewrite descriptions, hide parameters and mark which calls change data. You stay in control of what agents can do.

POST/bookings

Tool name

create_booking

Description

Book an appointment in an open slot. Always confirm the time and service with the user before calling.

Access level

Test in a real chat

Try your tools with Claude before any customer does.

Where is order ORD-71422?

get_order_status184 ms

It shipped with Royal Mail and should arrive on Nov 13.

Ask something

Every call, logged

See which agent called which tool, how long it took and what came back.

  • 14:02:31get_booking351 ms
  • 13:59:29get_booking440 ms
  • 13:56:59reschedule_booking824 ms

Install anywhere

Copy-ready setup steps for each AI client your users already have.

Add a custom connector with your MCP URL

Keys and rate limits

Issue a key per team or agent, cap requests per minute and revoke access instantly.

  • Production agentsrk_live_8f2a...
  • Support teamrk_live_c31d...
  • Staging testsrk_test_19be...

Limit120 requests / min

One URL, any client

Every server gets its own endpoint. Paste it into the AI tools your customers and team already use.

  • Streamable HTTP endpoint that follows the MCP spec
  • One API key per team, app or agent
  • Config snippets generated for your exact server
{
  "mcpServers": {
    "brightmoor-bookings": {
      "command": "npx",
      "args": [
        "-y", "mcp-remote", "https://brightmoor-bookings.mcp.mcpeasy.net/mcp",
        "--header", "Authorization: Bearer ${MCPEASY_KEY}"
      ],
      "env": { "MCPEASY_KEY": "rk_live_..." }
    }
  }
}

Questions, answered

The short version of how MCPEasy works and what it does with your data.

MCP (Model Context Protocol) is an open standard that lets AI assistants call tools. An MCP server lists the actions your product supports, such as check_availability or create_booking, and runs them against your API when an agent asks.