Skip to main content
Glama
@mnicole-dev/updown-mcp-server

@mnicole-dev/updown-mcp-server

by mnicole-dev

@mnicole-dev/updown-mcp-server

A Model Context Protocol (MCP) server for the Updown.io website monitoring API. Manage checks, view downtimes and metrics, configure alert recipients, and control status pages from any MCP-compatible client.

Features

15 tools covering the full Updown.io API:

Checks (5 tools)

Tool

Description

list-checks

List all monitoring checks with status, uptime, and apdex

get-check

Get details for a single check, optionally including metrics

create-check

Create a new URL monitoring check

update-check

Update check settings (URL, period, threshold, headers, etc.)

delete-check

Delete a monitoring check

Downtimes & Metrics (2 tools)

Tool

Description

get-check-downtimes

Get downtime history for a check (paginated)

get-check-metrics

Get performance metrics (apdex, uptime, timings) for a time range

Nodes (1 tool)

Tool

Description

list-nodes

List all global probe locations used by Updown.io

Recipients (3 tools)

Tool

Description

list-recipients

List all alert recipients (email, Slack, webhooks, SMS)

create-recipient

Create a new alert recipient

delete-recipient

Delete a recipient

Status Pages (4 tools)

Tool

Description

list-status-pages

List all status pages

create-status-page

Create a new public or private status page

update-status-page

Update a status page

delete-status-page

Delete a status page

Related MCP server: uptrack-mcp

Requirements

  • Node.js 18+

  • An Updown.io account with an API key (get it here)

Installation

npm install -g @mnicole-dev/updown-mcp-server

Or run directly with npx:

npx @mnicole-dev/updown-mcp-server

Configuration

Set the following environment variable:

export UPDOWN_API_KEY=your-api-key

Claude Code

Add to your ~/.claude.json:

{
  "mcpServers": {
    "updown": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@mnicole-dev/updown-mcp-server"],
      "env": {
        "UPDOWN_API_KEY": "your-api-key"
      }
    }
  }
}

Claude Desktop

Add to your config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "updown": {
      "command": "npx",
      "args": ["-y", "@mnicole-dev/updown-mcp-server"],
      "env": {
        "UPDOWN_API_KEY": "your-api-key"
      }
    }
  }
}

Examples

Check status of all your monitors

> Show me all my monitoring checks and their current status

Investigate downtime

> Show me the downtime history for check abc123

Get performance metrics

> Get metrics for check abc123 between 2024-01-01 and 2024-01-31, grouped by time

Create a new monitor

> Create a check for https://myapp.com every 30 seconds with alias "My App" and apdex threshold 0.5

Manage recipients

> List all my alert recipients and create a new email recipient for alerts@myteam.com

Create a status page

> Create a public status page with checks abc123 and def456, named "My Services Status"

How it works

  1. The MCP client sends a tool call to the server via stdio

  2. The server authenticates with Updown.io using the X-API-KEY header

  3. Requests are sent to https://api.updown.io

  4. Responses are formatted as human-readable markdown text and returned

Development

git clone https://github.com/mnicole-dev/updown-mcp-server.git
cd updown-mcp-server
pnpm install
pnpm dev          # Run with tsx (requires UPDOWN_API_KEY)
pnpm build        # Build to dist/

License

MIT

Available Tools

15 tools
create-checkC

Create a new monitoring check.

Inputs: url (required), type, alias, period, +9 more.

create-recipientB

Create a new alert recipient.

Inputs: type (required), value (required), name, selected.

create-status-pageC

Create a new public status page.

Inputs: checks (required), name, access_key, visibility, +1 more.

delete-checkA

Delete a monitoring check.

Inputs: token (required).

delete-recipientC

Delete an alert recipient.

Inputs: id (required).

delete-status-pageB

Delete a status page.

Inputs: token (required).

get-checkA

Get details for a single monitoring check, optionally including metrics.

Inputs: token (required), metrics.

get-check-downtimesB

Get the downtime history for a monitoring check.

Inputs: token (required), page.

get-check-metricsA

Get performance metrics for a monitoring check over a time range.

Inputs: token (required), to, from, group.

list-checksA

List all monitoring checks in your Updown.io account.

Inputs: no named parameters.

list-nodesA

List all Updown.io monitoring nodes (global probe locations).

Inputs: no named parameters.

list-recipientsA

List all alert recipients configured in your Updown.io account.

Inputs: no named parameters.

list-status-pagesA

List all public status pages in your Updown.io account.

Inputs: no named parameters.

update-checkC

Update an existing monitoring check.

Inputs: token (required), url, type, alias, +10 more.

update-status-pageC

Update an existing status page.

Inputs: token (required), name, checks, access_key, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 15 tools

    Disambiguation5/5

    Each tool clearly targets a distinct resource and action: checks, nodes, recipients, and status pages each have their own list/get/create/update/delete where applicable, with no overlapping purposes. The specialized check metrics and downtimes tools are unambiguous subsets of check data.

    Naming Consistency5/5

    All tool names follow a consistent kebab-case verb-noun pattern (e.g., list-checks, create-status-page). The pattern is predictable across all resources, with compound names like get-check-downtimes extending the same convention.

    Tool Count5/5

    15 tools is within the well-scoped 3-15 range for a domain-specific server. Each tool maps to a real operation on the Updown.io API, and the count feels right for covering the main resources without unnecessary bloat.

    Completeness4/5

    Checks and status pages have full CRUD coverage, plus additional check-specific operations. Recipients lack an update operation, but since recipients are likely simple contact addresses, create/delete/list may be sufficient. Overall the surface covers the core domain well with only minor gaps.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      D
      maintenance
      An MCP server that allows users to check if a website is experiencing downtime by querying isitdownrightnow.com, providing status information and details about recent downtime events.
      1
      2
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      MCP server for Uptrack uptime monitoring. Manage monitors and incidents from AI agents like Claude, ChatGPT, and Cursor.
      18 npm
      MIT
    • A
      license
      A
      quality
      A
      maintenance
      MCP server for Uptime Kuma that enables monitoring and management of uptime monitors, heartbeats, notifications, tags, and maintenance windows via natural language.
      67
      2,856 npm
      71
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      MCP server for StillOnline uptime monitoring, enabling management of projects, checks, incidents, and public status pages through natural language.
      10
      31 npm
      1
      MIT