Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 

Repository files navigation

Hyperping MCP Server

Official Model Context Protocol server for Hyperping. It exposes uptime monitoring, incident history, on-call schedules and reporting to AI agents over HTTP.

This is a remote server. There is nothing to install or self-host: point your client at the endpoint with an API key. This repository holds the registry manifest, the client configuration and the tool reference.

Full documentation: hyperping.com/docs/mcp

Endpoint

URL https://api.hyperping.io/v1/mcp
Transport Streamable HTTP, JSON-RPC 2.0
Auth Bearer token
Scope One project per key
Tools 26 (22 read, 4 write)

Quick start

  1. In Hyperping, go to Project Settings, API Keys and create a read_only or read_write key.
  2. Add the server to your MCP client (see below).
  3. Restart the client, then ask it things like "what is down right now?" or "what was our uptime on the API monitor last month?"

Claude Desktop, Claude Code, Cursor, Windsurf

{
  "mcpServers": {
    "hyperping": {
      "type": "http",
      "url": "https://api.hyperping.io/v1/mcp",
      "headers": {
        "Authorization": "Bearer sk_..."
      }
    }
  }
}

stdio clients

Clients that only speak stdio can bridge through mcp-remote:

{
  "mcpServers": {
    "hyperping": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://api.hyperping.io/v1/mcp",
        "--header",
        "Authorization: Bearer sk_..."
      ]
    }
  }
}

Tools

Monitors

Tool Scope Description
list_monitors read Paginated monitors, filterable by up, down, paused, ssl_expiring.
get_monitor read One monitor by UUID.
search_monitors_by_name read Case-insensitive substring search over names and URLs.
create_monitor write Create an HTTP, ICMP, port or DNS monitor.
update_monitor write Patch a monitor. Unspecified fields are preserved.
pause_monitor write Stop checks and alerts for a monitor.
resume_monitor write Resume a paused monitor.

delete_monitor is deliberately not exposed. Use the dashboard or the REST API.

Status and outages

Tool Scope Description
get_status_summary read Up, down and paused counts, plus what is currently down and since when.
list_outages read Paginated outages, filterable by status, type or search term.
get_outage read One outage, with acknowledgements, description and root cause.
get_outage_timeline read Detection, cross-region verification, alert dispatch, acknowledgement, resolution.
get_monitor_outages read Outages scoped to a single monitor.

Reporting

Tool Scope Description
get_monitor_uptime read Uptime percentage over a window, by hour, day, week or month.
get_monitor_response_time read Latency trend over a window.
get_monitor_mttr read Mean time to resolve, in seconds.
get_monitor_mtta read Mean time to acknowledge, in seconds.

Diagnostics

Tool Scope Description
get_monitor_anomalies read Flapping, latency spikes and other detected anomalies.
get_monitor_http_logs read Recent HTTP probe logs, filterable by level.
list_recent_alerts read Up and down notifications over a date range.

On-call

Tool Scope Description
list_on_call_schedules read All schedules, with rotation config and current on-call.
get_on_call_schedule read One schedule, with full rotation detail and linked policies.
list_escalation_policies read All escalation policies in the project.
get_escalation_policy read One policy, with step sequence, schedules and contact channels.
list_team_members read Project users, to resolve IDs coming from schedules and policies.

To answer "who is on call for monitor X?", chain get_monitor, get_escalation_policy, get_on_call_schedule, list_team_members.

Integrations

Tool Scope Description
list_integrations read Slack, Telegram, Discord, PagerDuty, OpsGenie, Teams, webhook and others.
get_integration read One integration, with its channel-specific config.

Rate limits

Limits apply per project, and both windows are enforced.

Method Per minute Per hour
initialize 5 20
tools/list 30 200
tools/call 60 600

Open the session once and reuse it rather than calling initialize per request. Exceeding a limit returns a JSON-RPC error with a retry_after_seconds hint.

Machine-readable spec

For a dense, agent-friendly spec covering tools, query patterns and response shapes, point your client at hyperping.com/mcp/llms.txt.

Support

Open an issue here, or reach the team through hyperping.com/contact/sales.

License

MIT for the contents of this repository. Use of the hosted service is governed by the Hyperping terms and privacy policy.

About

Official MCP server for Hyperping: uptime, API, cron and server monitoring, outages, reporting and on-call, exposed to AI agents over Streamable HTTP.

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors