What is ntfy-mcp?

An MCP server for ntfy, the pub-sub notification service that sends push messages to your phone with an HTTP request and nothing else.

It lets MCP clients like Claude Code, Claude Desktop or Codex send you notifications, read back what was sent, revise a notification in place while a job runs, and — with an admin account — create accounts and grant or revoke their access to topics, which otherwise means the ntfy command line on the server.

Why

A long-running task that finishes while nobody is watching is the case ntfy exists for, and it is exactly the case an assistant runs into. The interesting part is not the first message but the second: publish_message returns an id that is also the notification's sequence id, so update_message can revise the same notification rather than adding another one. A build that reports "running", "tests passed" and "deployed" stays one entry on the phone.

The other half is administration. ntfy's authorization model — which account may read or write which topic — is otherwise reachable only through the ntfy command line on the server, and list_users answers "who can read topic X" from a client.

What it is not

• A mail or telephone gateway. ntfy can turn a published message into an email or a phone call. Neither is exposed here: a tool that mails an arbitrary address on model output is a spam relay driven by injectable content, and call places a real, billable call.
• A way to obtain credentials. No tool creates, reads or exchanges an ntfy access token, because every such endpoint hands back a live credential that would then sit in the conversation transcript.
• A subscriber. /sse, /ws and /raw are streams; a tool call is request/response under a timeout. list_messages returns the same data, bounded.
• An upload endpoint. Attachments are attached by URL. The alternative is base64 through the model's context or a local filesystem this server has no business touching.