npx mcpblaze@latest skills · no signup

Serverless MCP & API GatewayEverything built-in for human and agent traffic

Secured MCP
Multi-tenant
OAuth sign-in
Authorization
Throttling & quotas
Localhost tunneling
Custom domains

One prompt. No sign-up needed.Chat with your backend safely.

Your app goes behind a proxy and an MCP that bring everything production needs: multi-tenant, authentication (OAuth sign-in) and authorization. In seconds.

Ask your agent to use MCPblaze to ready your backend for launch

~/code/mybackend $ npm run dev

⚡️ My app: A restaurant table booking app, running on localhost:3000

GET /reservations/{id} Get one booking

DELETE /reservations/{id} Cancel a booking

POST /reservations Book a table

~/code/mybackend $ claude

Show me what mcpblaze can do for my backend using npx mcpblaze@latest skills

Done setting up API & MCP Gateway ⚡️ npx mcpblaze createAPIhttps://myapp.abz.run/1.0.0/dev localhost:3000Dev Portalhttps://myapp.portal.abz.run/1.0.0MCPhttps://myapp-acme.mcpblaze.com/1.0.0/dev

Chatting with API (MCP) — npx mcpblaze chatbook me (Ana) a table for 2 tonight✓ Booked: table 1, 2 people at 19:30cancel Ben’s booking✗ Request denied (401). Not yours: only whoever made it can cancel it

~/code/mybackend $ npm run dev

⚡️ My app: A restaurant table booking app, running on localhost:3000

GET /reservations/{id} Get one booking

DELETE /reservations/{id} Cancel a booking

POST /reservations Book a table

~/code/mybackend $

Why it’s safer

Most MCPs don’t know who’s asking and for what. MCPblaze does.

They reach your API with one shared token, so every agent acts as the same user and can do whatever that token can. MCPblaze is an MCP server on top of a full API gateway: it knows who is behind every call, and what they’re allowed to do.

Most MCP servers

Any agent

every user looks the same

no sign-in

MCP server

holds one shared token

the same token, every time

Your backend

sees one caller: the token

Everyone is the same user, and can do whatever the token can.

MCPblaze

Your user's agent

Claude, ChatGPT, Cursor…

signs in as your user (OAuth)

MCP

your API's routes, as tools

API gateway

Who’s asking? the signed-in user
Allowed? only what they may do

with their name on it

Your backend

unchanged

Every call has a person behind it, and rules in front of it.

Authentication

Who’s asking?

An MCP server is there to do things: book a table, cancel an order, issue a refund. Every one of those needs a name on it.

People sign in once, with OAuth, right in their agent. From then on every call reaches your backend with their name on it, signed so it can’t be faked. You write no sign-in code.

Claude Code · connected to Nino Pizza (MCP)

signed in as ana@ninopizza.com

book me a table for 2 tonight

●Booked: table 1, 2 people at 19:30

what your backend receives

POST /reservations

x-abz-user-id user_8f2k… Ana, signed in

x-abz-tenant-id ninopizza her pizzeria

x-abz-identity eyJhbGciOi… signed by MCPblaze: can't be faked

{ "guests": 2, "time": "19:30" }

Authorization

Are they allowed?

Not everyone can do everything. Only whoever made a booking can change or cancel it, admins can act on anything, and everyone else gets a clear no.

The rules come from your OpenAPI spec and run in front of your backend, so no prompt can talk its way past them. Each customer gets its own MCP address too: Nino Pizza’s agents never see Gino Pizza’s bookings.

Claude Code · Nino Pizza (MCP)

Bencancel Ana’s 19:30 booking

✗Not yours: only whoever made it can cancel it

Bencancel my 20:00 booking

✓Cancelled

Ninocancel Ana’s 19:30 bookingadmin

✓Cancelled. Nino is an admin of Nino Pizza

checked by MCPblaze before your backend sees the call

Chat with your backend, safely

Paste one prompt into Claude Code or Codex. It reads your project and puts MCPblaze in front of it. No signup to try.

Read the docs · mcpblaze on npm