# Healthchecks.io MCP for AI Agents AI Agent Connect

> Healthchecks.io MCP lets you manage your cron jobs and background tasks directly through your AI client. You can list checks, inspect ping histories, and manage alerts without leaving your workspace. It connects your monitoring setup to your agent so you can debug failures and track uptime in real-time.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_9latvdFxENw6AeopmUvhNzFhcevLamye7Vce4SMi/ai-agent-connect
- **Tags:** cron-jobs, uptime-monitoring, background-tasks, alerting, infrastructure-monitoring

## Description

You're tired of jumping between different tabs just to see if your nightly backup actually finished or why a background worker stalled at 3 AM. This Connector changes that by bringing your Healthchecks.io dashboard directly into your AI agent. Instead of hunting through logs or clicking through a web interface, you can just ask your agent to show you the status of your production monitors. It handles the heavy lifting of fetching ping data and metadata so you can focus on fixing the actual issue. Because Vinkius hosts this Connector, you can get it running in minutes and start querying your infrastructure status immediately. You can check if a specific service is down, see the exact payload of a failed ping, or even pause a check while you're performing maintenance. It turns your AI into a first-responder for your background jobs, making it much faster to spot and resolve issues before they escalate.

## Tools

### get_ping_body
Retrieve the specific body of a ping to see exactly what data was sent. This is the fastest way to find error messages in failed tasks.

### list_checks
List all checks in your project, with options to filter by tags or slug. This helps you organize and find specific monitors quickly.

### list_flips
List the status changes and transitions for a specific check. Use this to see the history of when a service went down.

### list_pings
List the most recent pings for a specific check. This is great for verifying that a task is running on schedule.

### pause_check
Temporarily pause a check to prevent false alerts during maintenance. This keeps your team from getting paged during planned work.

### resume_check
Restart a paused check once maintenance is finished. Use this to bring your monitors back online.

### update_check
Modify the configuration of an existing check. This lets you change timeouts or URLs without using the web UI.

### create_check
Create a new monitoring check or update an existing one using a unique field. This helps you manage your infrastructure setup via natural language.

### get_check
Fetch a single check's details using its UUID or a unique key. Use this to see specific configuration for one monitor.

### list_badges
List all status badges for your project. This gives you a quick overview of your project's health status.

### get_status
Check the current operational status of the Healthchecks.io service. Use this to verify the platform is up before troubleshooting your own.

### delete_check
Remove a check from your project. Use this to clean up old monitors or retired services.

### list_integrations
List all configured notification channels and integrations. This helps you audit where your alerts are being sent.

## Prompt Examples

**Prompt:** 
```
Show me all my active checks.
```

**Response:** 
```
I've retrieved your active monitors. Here is the current status of your project:

| Check Name | Status | Tags |
| :--- | :--- | :--- |
| **Daily Backup** | ✅ Up | `backup`, `prod` |
| **Hourly Sync** | ✅ Up | `sync`, `dev` |
| **Cleanup Script** | ❌ Down | `maintenance` |

Would you like me to pull the latest pings for the Cleanup Script?
```

**Prompt:** 
```
Why did the backup job fail last night?
```

**Response:** 
```
I checked the recent pings for the **Daily Backup** check. The last ping failed with the following error in the body:

> `Error: ConnectionTimeout - Could not reach database at 10.0.0.5:5432` 

It looks like a network timeout occurred during the 2:00 AM run.
```

**Prompt:** 
```
Pause the 'Web Scraper' check for an hour.
```

**Response:** 
```
I've successfully paused the **Web Scraper** check. It will no longer send alerts or update its status until you ask me to resume it.
```

## Capabilities

### List all active monitoring checks
See every check in your project at a glance.

### Create new monitoring checks
Add new cron jobs or services to your dashboard.

### Inspect specific ping bodies
Retrieve the exact data sent by a worker to debug failures.

### Pause or resume checks
Silence alerts during maintenance windows with one command.

### View status flips
Track every time a service goes up or down over time.

### Check service status
Verify if the Healthchecks.io platform is currently operational.

## Use Cases

### Debugging a failed nightly backup
A developer notices a 'Job Failed' alert but doesn't know why. They ask the agent to show the last 5 pings for that check. The agent pulls the data, and the developer sees a 'Database Connection Refused' error in the payload.

### Silencing alerts during a migration
A DevOps engineer needs to perform a scheduled database migration. They ask the agent to pause the 'DB Sync' check. Once the migration is done, they ask the agent to resume it.

### Auditing production coverage
An SRE wants to see if all their 'Production' tagged checks are up. They ask the agent to list checks filtered by the production tag to ensure nothing was missed during deployment.

### Rapidly onboarding new services
A team is setting up a new monitoring project. They ask the agent to create a new check for their nightly backup script with a specific timeout, avoiding the need to manually configure the UI.

## Benefits

- Stop tab-switching by viewing all your active checks directly in your AI chat interface.
- Debug failed tasks faster by using get_ping_body to see the exact error messages sent by your workers.
- Manage maintenance windows easily by using pause_check and resume_check to silence alerts.
- Get a clear history of uptime by using list_flips to see every time a service went down.
- Audit your alerting setup by using list_integrations to see which channels are receiving notifications.
- Keep your infrastructure organized by using list_checks with tag filters to group your monitors.

## How It Works

The bottom line is you get a conversational interface for your entire monitoring dashboard.

1. Subscribe to the Connector and provide your Healthchecks.io API key.
2. Connect your AI client to the Vinkius catalog.
3. Ask your agent to list checks or check a specific ping body.

## Frequently Asked Questions

**Can I use Healthchecks.io MCP to see why my cron jobs are failing?**
Yes. The Connector can pull the specific body of a failed ping, which usually contains the error message or stack trace from your background worker.

**How do I manage alerts during maintenance with Healthchecks.io MCP?**
You can simply ask your agent to pause a specific check. This stops the monitoring and prevents your team from getting alerted while you do planned work.

**Can the Healthchecks.io MCP list my notification channels?**
Yes, it can list all your integrations. This is helpful for verifying that your Slack, PagerDuty, or email alerts are correctly configured.

**Will Healthchecks.io MCP help me find specific checks by tags?**
Yes, the tool supports filtering checks by tags. You can ask your agent to show only your production monitors or specific service groups.

**Can I create new checks using the Healthchecks.io MCP?**
Yes, you can tell your agent to create a new check, including details like the URL, timeout, and unique name, and it will handle the setup for you.

**Does Healthchecks.io MCP work with Claude or Cursor?**
Yes, it works with any MCP-compatible client, including Claude, Cursor, and Windsurf, as long as you have it connected via the Vinkius catalog.

**Can I see the specific data sent during a ping to debug a failure?**
Yes. Use the `get_ping_body` tool with the check UUID and ping number to retrieve the exact payload or logs sent by your script during that execution.

**How can I see the history of when my service went down?**
You can use the `list_flips` tool. It provides a history of status changes (from 'up' to 'down' and vice versa) for any specific check within a given timeframe.

**Is it possible to silence a check without deleting it?**
Absolutely. Use the `pause_check` tool to stop monitoring and alerts for a specific check. You can later use `resume_check` to start monitoring again.