Skip to main content
The @devhelm/mcp-server package provides a Model Context Protocol server that lets AI coding assistants interact with your DevHelm account. Use it in Cursor, Claude Desktop, or any MCP-compatible client.

​
Install

npm install -g @devhelm/mcp-server

​
Configure in Cursor

Add to your Cursor MCP settings (.cursor/mcp.json): 
{
  "mcpServers": {
    "devhelm": {
      "command": "npx",
      "args": ["-y", "@devhelm/mcp-server"],
      "env": {
        "DEVHELM_API_TOKEN": "dh_live_xxxxxxxx"
      }
    }
  }
}

​
Configure in Claude Desktop

Add to your Claude Desktop config (claude_desktop_config.json): 
{
  "mcpServers": {
    "devhelm": {
      "command": "npx",
      "args": ["-y", "@devhelm/mcp-server"],
      "env": {
        "DEVHELM_API_TOKEN": "dh_live_xxxxxxxx"
      }
    }
  }
}

​
Available tools

The MCP server exposes tools for managing your monitoring stack:
CategoryTools
MonitorsList, get, create, update, delete, pause, resume, test
IncidentsList, get, create, resolve
Alert channelsList, get, create, update, delete, test
Notification policiesList, get, create, update, delete
ConfigEnvironments, secrets, tags CRUD
DeploymentDeploy config, validate, diff
Status dataList services, get uptime, track dependencies
Resource groupsList, get health
DashboardOverview metrics

​
Example interactions

Once connected, you can ask your AI assistant things like:
  • “List all my monitors and their current status”
  • “Create an HTTP monitor for api.example.com/health checking every 60 seconds from us-east”
  • “Show me active incidents”
  • “Set up a Slack alert channel with this webhook URL”
  • “Deploy my devhelm.yml config”
  • “What’s the uptime for GitHub over the last 30 days?”

​
Transport modes

The MCP server supports:
  • stdio (default) — for Cursor and Claude Desktop
  • Streamable HTTP — for web-based MCP clients

​
Next steps

CLI overview

Full CLI command reference for terminal workflows.

Quickstart

Create your first monitor in under 5 minutes.