Yes, MockJet has a remote MCP server · OAuth sign-in · no API keys

MockJet MCP Server: Mock APIs from Your AI Assistant

MockJet runs a remote MCP server — connect any AI assistant with one URL, describe the mock APIs you need, and they appear live in seconds. Works with Claude Code, Claude, Codex, Cursor, VS Code, and any MCP client.

AI assistant · MockJet connected
› Create POST /login that returns a token, and a 401 when the password is wrong create_endpoint POST /login add_response 200 {"token": "{{uuid}}"} add_response 401 when body.password = "wrong" ✓ Live at mockjet.io/m/7c1f…/login

Connect MockJet's MCP Server in Under a Minute

Works with any AI app that supports remote MCP servers. One URL, then sign in to MockJet and click Allow.

MockJet MCP server URL
https://mockjet.io/mcp
1

Add MockJet to your AI app

Paste the link above into your AI app, or run one command.

2

Sign in and allow

Log in to MockJet in your browser and click Allow. No API keys needed.

3

Ask for what you need

For example, "Create a GET /users endpoint". Your AI builds it in MockJet.

Add MockJet to Claude Code

  1. Run this in your terminal:
    claude mcp add --transport http mockjet https://mockjet.io/mcp
  2. Start claude, run /mcp, pick mockjet and choose Authenticate.
  3. Sign in to MockJet in the browser window that opens and click Allow.

Add MockJet to Claude desktop or web

  1. Open Settings → Connectors and click Add custom connector. Name it MockJet and paste the URL:
    https://mockjet.io/mcp
  2. Click Connect, sign in to MockJet and click Allow.
  3. Make sure MockJet is switched on in the tools menu of your chat. On Team or Enterprise plans an owner may need to add the connector first.

Add MockJet to Codex

  1. Add the server:
    codex mcp add mockjet --url https://mockjet.io/mcp
  2. Sign in, then click Allow in the browser window that opens:
    codex mcp login mockjet
  3. Using the Codex IDE extension? Open the gear menu → MCP servers → Streamable HTTP, paste the URL, save, then choose Authenticate.

Add MockJet to Cursor

  1. Open Settings → MCP and add a new server, or put this in ~/.cursor/mcp.json:
    { "mcpServers": { "mockjet": { "url": "https://mockjet.io/mcp" } } }
  2. Click Connect (or Needs login) next to MockJet, sign in and click Allow.

Add MockJet to VS Code

  1. Run MCP: Add Server from the Command Palette and choose HTTP, or put this in .vscode/mcp.json:
    { "servers": { "mockjet": { "type": "http", "url": "https://mockjet.io/mcp" } } }
  2. Start the server, sign in to MockJet when prompted and click Allow.

Add MockJet to any MCP client

  1. Add a Streamable HTTP (remote) MCP server with this URL:
    https://mockjet.io/mcp
  2. The client discovers MockJet's OAuth sign-in automatically. Log in and click Allow. No keys or headers needed.

What Your AI Can Do with MockJet MCP

The MockJet MCP server gives your AI app the same building blocks as the dashboard, scoped to the projects you can access.

📁

Projects

list_projects and create_project: find the project to work in, or start a new one with its own mock base URL.

⚡

Endpoints

list_endpoints, get_endpoint, create_endpoint and update_endpoint: any method, path parameters like /users/:id, and fixed, random, or sequential response modes.

🎯

Responses & conditions

add_response, update_response and activate_response: status codes, headers, delays, timeouts, template variables, and rules that match on body, headers, or query params.

📋

Request logs

get_request_logs: your assistant reads what your app actually sent and which response it got, so it can debug the integration with you.

🔐

Secure by default

OAuth sign-in with no API keys to leak. The app acts only on projects you can access, and every tool call is rate limited and recorded.

🔌

Disconnect anytime

See every AI app connected to your account on the Connect AI page, when it was last active, and revoke its access in one click.

Things you can ask

Once MockJet is connected, try asking your AI app one of these.

MockJet MCP Server — Frequently Asked Questions

Yes. MockJet has a free remote MCP server. Paste the URL into Claude Code, Claude desktop or web, Codex, Cursor, VS Code, or any client that supports remote MCP servers. Sign in to MockJet once with OAuth — no API keys — and your AI assistant can create and manage mock APIs in your account just by asking.
Run claude mcp add --transport http mockjet https://mockjet.io/mcp in your terminal. Then start claude, run /mcp, pick mockjet and choose Authenticate. A browser opens — sign in to MockJet and click Allow. The tools are ready immediately.
MCP (Model Context Protocol) is an open standard that lets AI assistants call tools in other services. The MockJet MCP server exposes MockJet's mock API tools, so an assistant like Claude or Cursor can create projects, endpoints, and responses for you instead of you clicking through the dashboard.
Any app that supports remote MCP servers over Streamable HTTP with OAuth, including Claude Code, Claude desktop and web, Codex CLI and IDE extension, Cursor, and VS Code.
No. Your AI app discovers MockJet's OAuth sign-in automatically. You log in to MockJet in your browser and click Allow. The app only gets access to projects you can already access, and you can disconnect it at any time from the Connect AI page in your dashboard.
Yes. The MCP server is available on the free plan. Mocks your assistant creates count toward the same project, endpoint, and response limits as mocks you create in the dashboard.
Yes. Ask for the error cases you need, such as a 401 when the Authorization header is missing, a 429 rate limit, or a 504 timeout. The assistant adds responses with match conditions on the request body, headers, or query parameters, and can read the request logs to check what your app actually sent.

Let your assistant build the mocks.

Free to start. One URL, no API keys.