Monito

npx add-mcp 'https://www.monito.dev/api/mcp'

The Monito MCP server lets an AI tool run Monito directly: it checks a URL and expectation in a real browser, returns a verdict with evidence, and lets you save the check as a reusable test. The core CLI workflows are available as typed MCP tools; localhost checks and CI setup require the local server.

Two ways to connect

	Hosted	Local
Connect with	https://www.monito.dev/api/mcp	npx -y @monitodev/cli mcp
Sign in	In the browser, the first time you use it	Reuses monito auth login
Install	Nothing	Node.js 20+
Works in	Claude (web and desktop), ChatGPT, Claude Code, Cursor, VS Code	Clients that run local commands: Claude Code, Cursor, VS Code, Codex
Extra tools	None	CI setup for the repository it runs in
Localhost	Not supported	HTTP checks through the built-in tunnel

Use the local server for localhost checks, CI setup tools, or clients that can only run local servers. Both share the core tools and act as your Monito account; the hosted server checks deployed URLs.

Quick start

In Claude Code:

claude mcp add --transport http monito https://www.monito.dev/api/mcp

Open Claude Code, run /mcp and choose monito to sign in. Your browser opens, you approve the connection, and the tools are ready. See Connect a client for Claude, Cursor, VS Code, ChatGPT and the local server.

Then ask:

Check that the checkout on https://staging.example.com still works after my change, and fix anything Monito finds.

What a session looks like

1. The agent calls check with a URL and what should happen. It returns the verdict, cause when available, duration, credits, a report link and the first failing check or bug.
2. If the wait ends early, wait_for_run resumes it without starting another run.
3. On failure, get_run returns evidence for the agent to diagnose and fix the code.
4. Call check after the fix. To keep the check, call save_check with its checkId and a name. Add an automation separately to monitor the saved test.

Checks appear in the project's Runs view, labeled Check, and stay out of saved tests, suites, Discovery coverage and status pages until saved. Monito reuses your project with a matching default-environment origin, or creates one named after the host.

Both MCP servers run checks in the cloud. The local stdio server (monito mcp) can reach HTTP localhost through a built-in tunnel: pass a loopback URL to check, or use localPorts for up to three distinct ports (1024–65535). Keep the server and dev server running. The hosted server at /api/mcp rejects localhost and private-network URLs; use a deployed preview there. The CLI also supports monito check http://localhost:3000/path "…" and --local <port>. See local checks for limits and billing. Local checks cannot be scheduled or added to suites, automations or status pages, even after saving.

Default tools

The default is check,core-read: check, save_check, get_run, wait_for_run, list_projects, list_scenarios, and run_scenario.

Use https://www.monito.dev/api/mcp?toolsets=all or monito mcp --toolsets all for the full catalog. Explicit lists such as core,defects still work. Tools such as get_screenshot, get_credits, and CI setup require their toolset or all.

Agent guides: Claude Code, Cursor, Codex.

The server also offers a verify-change prompt that walks an agent through this loop. In Claude Code it appears as /mcp__monito__verify-change.

Runs, waits and credits

• Every run spends credits, exactly as it does from the dashboard or CLI. get_credits shows the balance.
• Run tools wait up to four minutes for a verdict. If the run takes longer, the tool returns the run ID with done: false; the agent calls wait_for_run to keep waiting. The run is never started twice.
• Pass a requestId (any unique string, such as a UUID) to run_scenario or run_project. If your client retries the call, Monito returns the first launch instead of charging again.
• A failed run means the agent found a bug. An error run is not a verdict about your app; get_run shows the evidence; opt into manage for get_run_logs.

Errors

Tool errors carry the same stable codes as the CLI: auth, not_found, validation, insufficient_credits, plan_required, timeout, network, server and conflict. Credit and plan errors include a link to manage your plan. Tools that delete data require confirm: true.

[

Data retention

How long Monito keeps run evidence, run history and your configuration on each plan.

](https://www.monito.dev/docs/data-retention)[

Connect a Client

Add the Monito MCP server to Claude Code, Claude, Cursor, VS Code, ChatGPT, Codex and other MCP clients.

](https://www.monito.dev/docs/mcp/clients)