Guide · Agent skill
How to use the metrics-dashboard agent skill for product KPIs
Updated · written with AI from the project's own documentation and checked against our catalogue data
The metrics-dashboard skill is a set of instructions that teaches an AI agent to help you design a product metrics dashboard. It is for product managers setting up KPIs for a product or feature, and for analysts planning product analytics or monitoring. After this guide you will know how to install the skill, ask for it in plain language, and what the markdown specification it produces contains.
Product Metrics Dashboard Designerphuryn/pm-skills · MIT Download the skill Quick start
- 1Download the ZIP. Get the metrics-dashboard ZIP file from this page and unzip it. Confirm the folder contains SKILL.md.
- 2Place the skill folder. Move the folder into ~/.claude/skills/ for all projects, or into .claude/skills/ inside one project.
- 3Or upload it in the apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the skill.
- 4Describe your product. Tell the agent what you want to measure, such as a new feature or an existing product area.
- 5Add context files. Attach or point to existing dashboards, analytics data, OKRs or strategy docs so the agent reads them first.
- 6Review the specification. Check the North Star, input metrics, metric table, layout, review cadence and alert routing in the markdown output.
- 7Validate and build. Test each threshold against real data and build the dashboard with your analytics tool, since the skill does not do this for you.
What the skill makes the agent do
Once the skill is installed, the agent can help you design a product metrics dashboard from scratch. It organizes your measures into four layers: a North Star metric, three to five input metrics that drive it, health metrics that act as guardrails, and business metrics such as revenue, cost and unit economics.
For each metric, the agent fills in a table with the exact calculation, data source, chart type, target and alert threshold. It also proposes a dashboard layout, a review schedule from daily to quarterly, and alert routing that says who is notified and how.
The output is a markdown specification document. It is not a live dashboard. The skill does not connect to analytics tools or data, and it includes no code or data access, so someone or other tooling still has to build the dashboard.
Install the skill from a ZIP file
Download the ZIP file for the skill from this page and unzip it. The skill's folder contains a file named SKILL.md. Place the folder in ~/.claude/skills/ to make it available across your projects, or in a project's .claude/skills/ folder to make it available only in that project.
You can also upload the skill in the Claude apps under Settings → Capabilities → Skills.
- Personal install: ~/.claude/skills/
- Project install: .claude/skills/ inside your project
- Apps: Settings → Capabilities → Skills, then upload
Trigger the skill with a natural request
The skill's description says to use it when creating a metrics dashboard, defining KPIs, setting up product analytics, or building a data monitoring plan. You do not need a special command. Describe what you are trying to measure and the agent should pick up the skill.
Give context where you can. The skill's instructions say that if you provide files such as existing dashboards, analytics data, OKRs or strategy docs, the agent should read them first.
- Design a metrics dashboard for our new onboarding flow.
- Help me define KPIs for the sign-up feature and set alert thresholds.
- Audit our current metrics and tell me which ones are vanity metrics.
- Draft a data monitoring plan for our checkout product, using the OKRs in this file.
What the output looks like
The skill asks the agent to save the dashboard specification as a markdown document. Each metric row covers the definition, data source, visualization, target and alert threshold, so you can review the plan line by line before building anything.
The skill also recommends tools based on your context. Its examples include Amplitude, Mixpanel and PostHog for product analytics, Looker, Metabase and Mode for SQL-based dashboards, and Datadog and Grafana for operational health.
- Metric table with calculation, source, chart type, target and threshold
- Dashboard layout sketch with North Star, inputs, health and business rows
- Review cadence: daily, weekly, monthly, quarterly
- Alert questions: what triggers investigation, who is alerted, expected response time
Requirements and safety notes
The skill lists no requirements. It does not run scripts and does not use the network. It does not analyze your data, so the thresholds it suggests are starting points to check against real numbers, not conclusions.
The skill draws on general frameworks: Lean Analytics, AARRR and HEART. Use it as a planning aid and validate the thresholds before you rely on them for alerts.
- Requires: nothing listed
- Runs scripts: no
- Uses network: no
- Does not connect to analytics tools or read your live data
Tips for better dashboards
Ask the agent to audit your existing metrics against the criteria the skill uses: a good metric is understandable, comparative over time, a ratio or rate, and behavior-changing. Metrics that do not change anyone's behavior are the ones to question.
Check that each input metric actually moves the North Star. Keep the number of input metrics small, since the skill's structure is built around three to five.
- Give the agent real context such as OKRs and existing dashboards.
- Ask for a leading versus lagging classification of your metrics.
- Treat every threshold as a first draft to validate with your own data.
What people use it for
- Setting up KPIs for a new product or feature before launch.
- Planning a product analytics setup with targets and chart types for each metric.
- Designing a monitoring plan with alert thresholds and a named response owner.
- Auditing an existing set of metrics to find vanity metrics that do not drive decisions.
Frequently asked questions
How do I install the metrics-dashboard agent skill?
Download the ZIP from this page and unzip it. Place the folder in ~/.claude/skills/ for personal use or in a project's .claude/skills/ folder. You can also upload it in the Claude apps under Settings → Capabilities → Skills.
Does the metrics-dashboard skill require anything else to run?
The skill lists no requirements. It does not run scripts and does not use the network, according to its listing.
Is the metrics-dashboard skill free?
The skill is published under the MIT licence. This page does not list pricing, so check the licence text in the repository for the terms.
Will the skill create a live dashboard for me?
No. It produces a markdown specification document. It does not connect to analytics tools or data, so the dashboard must be built by a person or other tooling.
Which analytics tools does the skill recommend?
It suggests Amplitude, Mixpanel and PostHog for product analytics, Looker, Metabase and Mode for SQL-based dashboards, and Datadog and Grafana for operational health.
Are the alert thresholds reliable?
The skill does not analyze your data. It draws on general frameworks, so treat its thresholds as starting suggestions and validate them against your own numbers.
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 a product manager or analyst who needs a clear first draft of KPIs, targets, chart types and alert thresholds for a new product or feature. Pick something else if you need a live dashboard built or connected to your data, or if you want the thresholds derived from your own analysis.
Download the skill