• LATENCY
  • FAILURE
  • INVENTORY
  • HUMAN

How to analyze monitor performance metrics with your AI

Connect Checkly to your agent to turn raw monitoring data into clear troubleshooting insights.

Ask AI about this page

Short answer

How can I use my AI to analyze Checkly performance?

Your agent pulls specific response times and check details directly from Checkly. This lets you find the cause of latency or failures without digging through dashboards. You'll get a breakdown of exactly what happened during a check.

Troubleshooting outcomes.

Where your data lands.

The agent processes your monitoring data into these specific results.

LATENCY

Identify slow responses

The agent pulls response times to show you exactly when latency spiked. You get a clear view of performance trends.

FAILURE

Pinpoint check failures

Your agent looks up specific check details to explain why a monitor failed. It identifies the error or timeout causing the issue.

INVENTORY

Audit all monitors

The agent lists your existing checks to give you a full view of your monitoring coverage. It helps you see what's active.

HUMAN

Manual intervention

The AI identifies the problem, but you decide if you need to roll back a deployment or fix a config. It leaves the final decision to you.

The workflow

What your AI does when a check fails.

The agent moves from broad monitoring to specific data points to find the root cause.

  1. Scan active monitors

    The agent starts by looking at your entire list of configured checks to find the one reporting issues.

    list_checkly_checks

  2. Inspect check status

    Once it finds the failing monitor, the agent pulls the specific configuration and status to understand the context.

    get_check_details

  3. Pull performance data

    The agent digs into the actual response times and timing metrics to see if the failure is a total outage or just slowness.

    get_check_performance_metrics

  4. Summarize findings

    The agent combines the check details and performance numbers into a plain English explanation of what went wrong.

    get_check_detailsget_check_performance_metrics

Try it

Copy these to start.

Use these prompts to begin your investigation.

Starting points

Swap out the bracketed text with your actual Checkly check names or IDs.

Checkly Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_ZKVvNJK1bEgGdJP2vFYcWaZwHDzM0XtD3oz2UTx3/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Checkly capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "checkly-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_ZKVvNJK1bEgGdJP2vFYcWaZwHDzM0XtD3oz2UTx3/mcp"
    }
  }
}
Copy into chat04

  • List all my Checkly checks so I can see what is currently being monitored.

  • Get the details for the 'api-gateway-health' check and tell me why it's failing.

  • Show me the performance metrics for my 'checkout-flow' monitor from the last hour.

  • Compare the response times for my 'auth-service' check against its usual performance.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Checkly once, then ask.

Just link your account once. Your credentials stay encrypted, and you can immediately start asking your agent about your monitor data.

Connect Checkly to your AI

FAQ

How the Checkly Connector behaves

  • 01

    Can my AI change my Checkly monitor settings?

    No. The agent can only read check details and performance metrics. It cannot modify your monitors or change your configurations.

  • 02

    Can the AI fix a failing check automatically?

    No. The agent identifies the cause of a failure by looking at the data, but you must perform the actual fix in Checkly yourself.

  • 03

    What kind of performance data can it see?

    The agent can retrieve specific response times and other performance metrics provided by the Checkly API.

  • 04

    Does the AI see all my monitors?

    Yes, if you ask it to list your checks, it will pull the full list of monitors available in your Checkly account.

  • 05

    Can I ask about a specific check by name?

    Yes. Once the agent has listed your checks, you can ask it to pull details for a specific one by its name or ID.

  • More questions about Checkly? The Connector page answers them. See everything the Checkly Connector can do

Connect Checkly to Claude, Cursor, ChatGPT & more