# Cronitor MCP for AI Agents AI Agent Connect > Cronitor lets you manage your entire infrastructure's health through your AI agent. You can track cron job statuses, send heartbeats, and monitor website uptime without leaving your editor. It handles the heavy lifting of checking if your nightly backups actually finished or if your production site is down, then feeds that data directly to your agent for instant analysis and response. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_EHNlqUmdOzFgXZGMmL53K1EbEQVA8FefmcOzppi4/ai-agent-connect - **Tags:** cron-jobs, uptime-monitoring, heartbeats, incident-management, telemetry ## Description Manage your infrastructure's health through natural conversation by connecting your monitoring account to your AI agent. You can track cron job statuses, send heartbeats, and monitor website uptime without ever leaving your editor. It handles the heavy lifting of checking if your nightly backups actually finished or if your production site is down, then feeds that data directly to your agent for instant analysis and response. Instead of hunting for logs or toggling between browser tabs, you stay in your workflow. It's one of the many ways Vinkius helps you connect your existing infrastructure tools to your AI workspace. You get a 24/7 SRE assistant that knows your uptime history and can execute commands like creating maintenance windows or updating notification lists on the fly. You can ask your agent to find out why a specific heartbeat stopped reporting or to query performance metrics to understand latency and success rates. Whether you are trying to organize your monitors into groups or need to create a new status page for your customers, your agent can do it in seconds. This means you spend less time clicking through nested menus and more time actually fixing problems. You can even manage your API keys and audit your notification settings through the same chat interface. It turns a complex monitoring stack into a simple, conversational experience where your agent acts as your primary interface for infrastructure health. You can quickly see which environments are failing and get the exact metrics you need to debug issues without manual searching. This setup removes the friction of infrastructure management by bringing all your monitoring data into a single, intelligent chat window. ## Tools ### create_api_key Generate a new API key for your Cronitor account. This is useful for setting up secure access for your AI agent. ### create_status_page_component Add a new component to an existing status page. Use this to build out specific sections of your public status updates. ### create_environment Set up a new environment for your monitoring. This helps you organize monitors by production, staging, or dev stacks. ### create_group Create a new monitor group. Use this to batch monitors by service or team for easier management. ### create_issue Log a new issue in your system. This lets you track specific incidents as they occur. ### create_maintenance_window Schedule a maintenance window. This tells the system to silence alerts during planned work periods. ### create_notification Create a notification list. This defines who gets alerted when your monitors go down. ### create_site Create a RUM site. Use this to start tracking real user monitoring data for your web applications. ### create_status_page Create a status page. This builds a public-facing page to communicate uptime to your customers. ### delete_api_key Delete an API key. Use this to remove old or compromised keys from your account. ### delete_group Delete a monitor group. Use this to clean up old monitor groupings in your account. ### delete_monitor Delete a monitor. Use this to stop tracking a specific cron job or heartbeat. ### delete_site Delete RUM sites. Use this to remove specific site analytics from your dashboard. ### delete_status_page Delete status pages. Use this to remove public status pages from your account. ### get_aggregates Retrieve rolled-up aggregates. Use this to see summarized performance data over time. ### get_metrics Retrieve performance metrics. Use this to see specific data points like latency and success rates. ### get_monitor Retrieve a specific monitor. Use this to see the full details of a single health check. ### get_site_errors Get site errors. Use this to identify specific issues happening on your RUM sites. ### list_api_keys List API keys. Use this to audit all active keys in your Cronitor account. ### list_status_page_components List status page components. Use this to see all parts of your public status pages. ### list_environments List environments. Use this to see all the different stacks you are monitoring. ### list_groups List monitor groups. Use this to see how your monitors are organized. ### list_issues List issues and incidents. Use this to see a history of problems in your infrastructure. ### list_monitors List or search monitors. Use this to find specific health checks across your whole stack. ### list_notifications List notification lists. Use this to see who is currently receiving alerts. ### list_sites List RUM sites. Use this to see all the websites you are tracking for analytics. ### list_status_pages List status pages. Use this to see all your public status pages. ### pause_group Bulk pause or resume a group of monitors. Use this to silence alerts for an entire stack at once. ### update_environment Update an environment. Use this to change the configuration of a specific monitoring stack. ### update_issue Update an issue state. Use this to move an incident from 'Open' to 'Resolved' or 'In Progress'. ### update_monitors Create or update monitors in bulk. Use this to apply changes to many monitors at once. ### update_notification Update a notification list. Use this to change who gets alerted for your monitors. ### update_status_page Update status pages. Use this to change the content of your public status pages. ### create_monitor Create a new monitor. Use this to start tracking a new cron job, heartbeat, or website. ### delete_status_page_component Delete status page components. Use this to remove specific sections from your public status pages. ### pause_monitor Pause or resume a monitor. Use this to silence alerts for a single specific heartbeat or job. ### query_site_analytics Query site analytics. Use this to pull performance data from your RUM sites. ### send_telemetry Send telemetry events. Use this to send pings and metrics from your application to Cronitor. ### update_api_key Update an API key. Use this to modify the settings of an existing key. ### update_status_page_component Update status page components. Use this to modify specific parts of your public status pages. ### update_group Update a monitor group. Use this to change how your monitors are organized. ### update_site Update RUM sites. Use this to modify the settings for your site analytics. ### bulk_issues Perform bulk actions on issues. Use this to update multiple incidents in a single command. ### clone_monitor Clone an existing monitor. Use this to quickly replicate a health check for a new service. ## Prompt Examples **Prompt:** ``` Check if the nightly backup job finished successfully. ``` **Response:** ``` I've checked your monitors. The 'nightly-db-backup' job completed successfully at 3:15 AM with a 120ms latency. All other heartbeats are healthy. ``` **Prompt:** ``` Silence alerts for the production site for the next 2 hours. ``` **Response:** ``` I've scheduled a maintenance window for your 'production-site' group. Alerts will be silenced until 4:00 PM today. You can see the details in your Cronitor dashboard. ``` **Prompt:** ``` Show me the average latency for our main API over the last hour. ``` **Response:** ``` The P50 latency for the 'main-api' monitor over the last hour was 142ms. Here is the breakdown: - Success Rate: 99.8% - Error Rate: 0.2% - Avg Duration: 145ms ``` ## Capabilities ### Check monitor statuses See the current health of your cron jobs and heartbeats in real time. ### Send heartbeats and telemetry Push pings and execution states directly from your code to your monitoring dashboard. ### Manage status pages Create and update public status pages to keep your users informed during incidents. ### Create maintenance windows Schedule planned downtime to silence alerts and prevent unnecessary notifications. ### Query performance metrics Retrieve latency, success rates, and other key metrics for your infrastructure. ## Use Cases ### Outage Management An SRE notices a spike in errors. They ask the agent to list_issues, see the severity, and use update_issue to move it to 'In Progress' while keeping the team updated. ### Deployment Window A developer is about to run a script. They ask the agent to create_maintenance_window for the next hour so they don't get paged during the rollout. ### Performance Audit A product manager wants to know site speed. They ask the agent to query_site_analytics for the last 24 hours to see the impact of a new feature. ### Bulk Setup A new service launches with many components. The user asks the agent to use update_monitors to create 10 different health checks at once. ## Benefits - Silence alerts instantly by using create_maintenance_window to tell your agent about planned work. This prevents your team from getting paged during known maintenance windows. - Get real-time health checks by using send_telemetry to push pings directly from your backend code. It's the fastest way to verify that your background workers are still running. - Query your site's health using get_metrics to see latency and success rates without opening a browser. You can spot performance regressions before your users do. - Manage your public status pages via create_status_page to keep your customers informed during an outage. Your agent can update the status with a single command. - Organize your infrastructure better by using create_group to batch monitors by environment or service. This makes it much easier to manage large scale deployments. - Audit your security by using list_api_keys to see who has access to your monitoring data. Keep your infrastructure secure by managing keys through your AI client. ## How It Works The bottom line is that it turns your monitoring dashboard into a conversational interface for your AI. 1. Connect your Cronitor API key to the Connector. 2. Tell your agent to list your monitors or check a specific heartbeat. 3. Get real-time status updates and performance data instantly. ## Frequently Asked Questions **Can I use Cronitor MCP to silence alerts?** Yes, you can ask your agent to create_maintenance_window for specific monitors or groups. This stops notifications during planned work so you don't get paged unnecessarily. **Does Cronitor MCP support heartbeats?** Yes, your agent can send_telemetry to report heartbeats. This is great for checking if a background worker is still alive. **Can my agent update our public status page?** Absolutely. You can use the Connector to update_status_page or create_status_page_component to keep your users informed during an incident. **How do I check my cron job performance?** Just ask your agent to get_metrics for a specific monitor. It will pull the latest latency and success data directly from Cronitor. **Can I manage multiple environments with Cronitor MCP?** Yes, you can list_environments and use the Connector to create or update monitors across different stacks in one conversation. **Is it possible to bulk update my monitors?** Yes, the update_monitors tool allows your agent to modify multiple monitors at once, which is much faster than doing it one by one. **How can I manually signal that a job has finished successfully?** You can use the `send_telemetry` tool. Provide the `monitor_key` and set the `state` to 'complete'. This will immediately update Cronitor with the job's success status. **Is it possible to silence alerts for a specific monitor during maintenance?** Yes! Use the `pause_monitor` tool with the `monitor_key` and the number of `hours` you want to pause it. Alternatively, use `create_maintenance_window` for broader scheduling. **How do I check for recent errors on my monitored websites?** Use the `get_site_errors` tool. It retrieves the most recent error logs and failure details for a specific site monitor, helping you debug downtime quickly.