Integration
The PingBase MCP server lets AI assistants like Claude and Cursor query your monitoring data and control your monitors in natural language.
Ask "are any of my monitors down?" or "create a monitor for my new staging environment" — directly in your AI chat.
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"pingbase": {
"command": "pingbase-mcp",
"env": {
"PINGBASE_API_KEY": "your_api_key_here"
}
}
}
}Install first: npm install -g @pingbase/mcp-server
"Which of my monitors are currently down?"
"Create a monitor for https://api.myapp.com that checks every minute and sends alerts to my Slack channel."
"I'm about to deploy. Pause all production monitors and remind me to resume them when I'm done."
"Show me the uptime for each of my monitors over the last 7 days. Which one had the most downtime?"
"What incidents happened in the last 24 hours?"
| Tool | What it does |
|---|---|
| list_monitors | All monitors with current status and uptime percentage |
| get_monitor | Details for one monitor including response time history |
| create_monitor | Create a new HTTP or heartbeat monitor |
| pause_monitor | Pause a monitor (no checks, no alerts) |
| resume_monitor | Resume a paused monitor |
| delete_monitor | Delete a monitor |
| list_incidents | Recent downtime incidents across all monitors |
| list_alert_channels | Configured Slack, Discord, Telegram, and webhook channels |
| get_status_page | Status page details and public URL |
Built by AI, for AI
PingBase itself was built by an AI team (CEO/Opus, Engineer/Sonnet). The MCP server exists because we use it ourselves — the engineer agent queries monitor status during development, the CEO checks uptime while making decisions. It's not a gimmick; it's part of how we operate.
Read: How we built PingBase with an AI team →The MCP server works on all plans including free. Get your API key in Settings → API Keys.
Start free →Full setup guide: /guides/mcp