README - 🚀 Uptime Agent MCP Server by avimbu

🚀 Uptime Agent MCP Server

Connect your Uptime Agent monitoring system directly to AI assistants like Claude through the Model Context Protocol (MCP).

✨ Features

Real-time Monitoring Access: Allow AI assistants to check your system's uptime status
Incident Management: View and analyze downtime incidents through natural conversation
Monitor Creation: Set up new monitoring endpoints with simple voice or text commands
Secure Integration: Enterprise-grade security for your monitoring infrastructure

🔍 What is Uptime Agent?

Uptime Agent is a powerful monitoring solution that tracks your websites and API endpoints, alerting you when they go down. This MCP server extends Uptime Agent's capabilities by letting you interact with your monitoring system through AI assistants.

🛠️ Installation

Prerequisites

Node.js 18 or higher
An active Uptime Agent account
Your Uptime Agent API key

To obtain your Uptime Agent API key:

Log in to your Uptime Agent Dashboard
Navigate to Account → API Keys
Create a new API key with appropriate permissions
Copy the generated key for use with the MCP server

Option 1: Quick Install via NPM (Recommended)

The fastest way to get started is with our setup command:

npx uptime-agent-mcp setup

This command will:

Install the MCP server
Configure it for use with Claude Desktop
Prompt you for your Uptime Agent API key
Set up all necessary configurations automatically

Option 2: Install via Smithery.ai

To install using Smithery.ai:

Create an account at smithery.ai
Get your personal key from your Smithery account
Run the following command:

npx -y @smithery/cli@latest install @AVIMBU/uptime_agent_mcp --client claude --key <personal_key>

Replace <personal_key> with your actual Smithery personal key.

Option 3: Manual Local Installation

For advanced users who want more control:

# Clone the repository
git clone https://github.com/AVIMBU/uptime_agent_mcp.git
cd uptime_agent_mcp

# Install dependencies
npm install

# Build the project
npm run build

Configure with your API key by creating a .env file:

UPTIME_API_KEY=your-api-key-here
PORT=3000  # Optional, defaults to 3000

Start the server:

npm start
# or directly with
node dist/index.js

🤖 AI Assistant Integration

Setting Up with Claude Desktop

After installing using one of the methods above, your MCP server is automatically configured for Claude Desktop.

If you installed manually, add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "uptime-agent": {
      "command": "npx",
      "args": [
        "-y",
        "uptime-agent-mcp"
      ],
      "env": {
        "UPTIME_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Alternatively, you can use Docker:

{
  "mcpServers": {
    "uptime-agent": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "UPTIME_API_KEY",
        "uptime-agent-mcp"
      ],
      "env": {
        "UPTIME_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Example Conversations

Checking Monitors:

"Claude, show me all my active uptime monitors."

Creating a New Monitor:

"Please create a new monitor for our API endpoint at https://api.mycompany.com/v2/health"

Analyzing Incidents:

"What incidents happened on our production servers last week, and what was the average downtime?"

📊 Available Functions

Monitor Operations

Function	Description	Parameters
`listMonitors`	Get a complete list of all monitoring endpoints	None required
`getMonitor`	Retrieve detailed information about a specific monitor	`id`: Monitor identifier
`createMonitor`	Set up a new endpoint to monitor	`name`: Monitor name `url`: URL to monitor `tracking_type`: Type of monitoring (http, ping, etc.) `check_frequency`: Check interval in seconds

Incident Management

Function	Description	Parameters
`listIncidents`	View all detected downtime incidents	None required
`getIncident`	Get detailed information about a specific incident	`id`: Incident identifier
`listIncidentsByMonitor`	See all incidents for a particular endpoint	`monitor_id`: Monitor identifier

Public Tracking

Function	Description	Parameters
`createAnonymousTracking`	Create public tracking without authentication	`url`: URL to monitor `name`: (Optional) Name for the tracking

Integration with Slack (Coming Soon)

Function	Description	Parameters
`slack_get_users`	List all users in connected Slack workspace	`limit`: Max number of users `cursor`: Pagination cursor
`slack_post_message`	Post notifications to Slack	`channel_id`: Channel to post to `text`: Message content

🐳 Docker Deployment

We provide Docker support for easy deployment:

# Build the Docker image
docker build -t uptime-agent-mcp .

# Run the container
docker run -p 3000:3000 -e UPTIME_API_KEY=your-api-key uptime-agent-mcp

📬 Support

If you have questions or need assistance:

Open an issue on GitHub
Contact us through our website: AVIMBU

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

Developed with ❤️ by AVIMBU