Guide · Agent skill
How to use the Cloud Monitoring Metric Finder agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The Cloud Monitoring Metric Finder is an agent skill, a packaged set of instructions that an AI agent loads when your request matches its purpose. This one helps you find the exact Cloud Monitoring metric names for a Google Cloud service such as Cloud Run, BigQuery or Cloud SQL. After this guide you will know what the skill does, how to install it, how to phrase requests so it triggers, and what it needs from your project.
Cloud Monitoring Metric Findergoogle/skills · Apache-2.0 Download the skill Quick start
- 1Download the skill. Download the ZIP of the cloud-monitoring-metric-selection skill from the page where you found this guide.
- 2Install it. Unzip it into ~/.claude/skills/ for personal use, or into your project's .claude/skills/ folder. In the Claude apps, upload the ZIP under Settings, Capabilities, Skills.
- 3Set up the MCP server. Make sure your agent has the Google Cloud Monitoring MCP server at https://monitoring.googleapis.com/mcp configured with Google credentials. If it is missing, the skill can add the entry for you.
- 4Restart the chat if prompted. If the skill added the server entry, start a new chat session so the list_metric_descriptors tool becomes available.
- 5Ask with your project ID. Name the service, the metric concept and your real Google Cloud project ID, for example asking for Cloud Run latency metrics in your project.
- 6Review the metric names. Check the returned metric type strings and confirm them in your project before you use them in dashboards, alerts or monitoring-as-code.
What does the skill make the agent do?
When you ask about monitoring a Google Cloud service, the agent queries Google Cloud Monitoring for the list of metrics that service exposes. It then filters that list by keywords such as CPU, latency or connections to suggest the most relevant ones.
The skill is designed to use live data from your project rather than guessing metric names from memory. This matters when you are building dashboards, alerts or monitoring-as-code and need the exact metric type strings.
- Retrieves metric descriptors for a service or resource from the Cloud Monitoring API
- Filters descriptors locally by keywords in their type, display name and description
- Returns a short list of key metrics for each service it checks
How do you install the skill?
Download the skill ZIP from the page where you found this guide. Then choose one of the two methods below.
For a personal setup, unzip the download into ~/.claude/skills/. To make the skill available only inside one project, unzip it into that project's .claude/skills/ folder instead.
If you use the Claude apps rather than a local setup, open Settings, go to Capabilities, then Skills, and upload the ZIP there.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into your project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
What does the skill need before it works?
The skill needs a Google Cloud project with Cloud Monitoring API access. It also needs the Google Cloud Monitoring MCP server at https://monitoring.googleapis.com/mcp, configured with Google credentials, and an MCP-capable agent runtime.
The gcloud CLI is optional. You only need it if you want the agent to run terminal commands, which are then tagged with an attribution header.
- A Google Cloud project with Cloud Monitoring API access
- The Google Cloud Monitoring MCP server (https://monitoring.googleapis.com/mcp) with Google credentials
- An MCP-capable agent runtime
- gcloud CLI (optional)
How do you trigger the skill with a natural request?
Name the Google Cloud service and the kind of metric you want, and include your project ID. The agent needs a real project ID before it queries anything, so requests without one will prompt a question back to you.
Good requests state the service, the concept and the project. For example, you might ask the agent to list the Cloud Run metrics for your project that relate to latency, or to find the Cloud SQL connection metrics in a named project.
- Ask for metrics by service: 'List Cloud Run metrics in project my-real-project related to latency'
- Ask about a resource: 'Which BigQuery metrics show bytes scanned for my project?'
- Ask for a discovery list: 'Find Cloud SQL metric types about connections'
What happens if the query tool is missing?
If the agent cannot find a tool for listing metric descriptors, the skill checks common MCP configuration file locations and adds a google-cloud-monitoring server entry. It merges this entry into the existing file so your other servers are kept. It then asks you to restart or start a new chat so the tools refresh.
This step edits configuration files on your machine. If you do not want the agent changing files such as claude_desktop_config.json, review the change or set up the MCP server yourself before using the skill.
- Config files it checks include ~/.gemini/config/mcp_config.json, ~/.codeium/windsurf/mcp_config.json, cline_mcp_settings.json and claude_desktop_config.json
- The server entry it adds uses the authProviderType google_credentials and enables the list_metric_descriptors tool
"google-cloud-monitoring": {
"url": "https://monitoring.googleapis.com/mcp",
"authProviderType": "google_credentials",
"enabledTools": [
"list_metric_descriptors"
]
}What should you check in the results?
The skill lists metric descriptors for each service prefix, such as bigquery.googleapis.com/ or compute.googleapis.com/, and follows pagination until all results are retrieved. Use the metric type strings it returns as the source of truth when you write dashboards or alerts.
If live calls fail, the skill is designed to say so. It states the error, the fallback source it used, such as public documentation, and the risk that the data may be stale or missing custom metrics.
- Check that the metric type strings match the service you asked about
- Note any fallback warning and verify the metric in your own project before relying on it
What people use it for
- Finding the exact metric type string for a Cloud Run latency chart before writing a dashboard
- Listing BigQuery metrics such as bytes scanned to decide what to alert on
- Checking which Cloud SQL connection metrics exist before writing monitoring-as-code
- Exploring what telemetry a Compute Engine or Pub/Sub resource exposes during an operations review
Frequently asked questions
Is the Cloud Monitoring Metric Finder skill free?
The skill is published under the Apache-2.0 licence. The source does not state pricing for Google Cloud Monitoring itself, so check Google Cloud's own pricing for your project.
What do I need to install the agent skill?
You need a Google Cloud project with Cloud Monitoring API access, the Google Cloud Monitoring MCP server configured with Google credentials, and an MCP-capable agent runtime. The gcloud CLI is optional.
Where do I put the skill folder?
Unzip the download into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. In the Claude apps you can instead upload the ZIP under Settings, Capabilities, Skills.
Why does the agent ask for my project ID?
The skill requires a real Google Cloud project ID before it calls any API. It will not query placeholder names such as my-project-id or YOUR_PROJECT_ID, so it asks you to provide one.
Does this skill run scripts or use the network?
According to the skill's listing, it does not run scripts but it does use the network, because it queries the Google Cloud Monitoring API through the MCP server. It may also edit MCP configuration files to add the server entry.
Does this skill interpret metric values or set up alerts?
No. The skill finds and filters metric names for a service. It does not interpret metric values or set up alerting policies, so use other tools or documentation for those tasks.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you are building dashboards, alerts or monitoring-as-code on Google Cloud and need exact metric names from your own project. Pick something else if you cannot share a real project ID, do not want the agent editing MCP configuration files, or need help reading metric values or configuring alerting policies.
Download the skill