Guide · Agent skill
How to use the Data Analytics agent skill: install and run it
Updated · written with AI from the project's own documentation and checked against our catalogue data
The Data Analytics skill is a packaged set of instructions that gives an AI coding agent a structured playbook for quantitative product and business analysis. It covers data quality checks, metric diagnostics, KPI design, dashboards, reports, charts, notebooks and market sizing. This guide explains what an agent skill is, how to install this one, how to trigger it with natural requests, and what to check before you rely on its output.
Data Analytics BundleXiaomiMiMo/MiMo-Code · MIT Download the skill Quick start
- 1Download the ZIP. Download the Data Analytics skill ZIP from this directory page.
- 2Unzip into a skills folder. Unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder.
- 3Or upload in the Claude apps. In the Claude apps, open Settings → Capabilities → Skills and upload the ZIP.
- 4Make a natural request. Describe your analysis question, the data you have and the output format you want.
- 5Check the routing. The agent selects a focused workflow and reads its instructions before acting.
- 6Review sources and uncertainty. Read the source notes and limitations in the result before using it.
What is an agent skill and what does this one do?
An agent skill is a folder containing a SKILL.md file and supporting files. When the skill is installed, the agent can read these instructions and follow them for matching tasks. This skill is an entry point: it reads a request, picks the most relevant focused workflow, and then follows that workflow's steps.
The focused workflows include analyzing data quality, diagnosing a metric change, designing KPIs, building a dashboard or report, gathering business context, and sizing a market. The agent keeps source notes and uncertainty visible in its results rather than presenting conclusions without evidence.
- Data quality checks: judge whether data, query results or dashboards are trustworthy enough to use.
- Metric diagnostics: investigate why a metric changed.
- KPI design and reporting: define metrics, guardrails and targets, and report on them.
- Dashboards and reports: build a reusable monitoring view or a one-time analytical report.
- Charts, notebooks and market sizing.
Install the Data Analytics skill
Download the ZIP file from this directory page and unzip it. For a personal install, place the unzipped folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder.
If you use the Claude apps instead, upload the ZIP under Settings → Capabilities → Skills.
- Personal install: ~/.claude/skills/
- Project install: the .claude/skills/ folder inside your project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Trigger the skill with natural requests
You do not need a special command. Describe the analysis you want in plain language, and the agent will match the request to the skill. The skill description says it applies to quantitative product or business analysis, and also whenever Data Analytics is explicitly invoked, so naming it directly is the most reliable option.
Be specific about the question, the data you have and the output you want. The skill routes to the narrowest matching workflow, so a precise request gets a more focused result.
- "Use Data Analytics to check whether this orders export is complete enough to report on."
- "Diagnose why weekly active users dropped last week. Here is the CSV."
- "Design a KPI set for our onboarding funnel with guardrail metrics."
- "Build a self-contained HTML dashboard for our monthly signups."
- "Size the market for a meal-kit subscription in one region, with sources noted."
Choose the output format
Output stays portable. Quick answers come back as concise Markdown with source notes. Durable reports and dashboards can be Markdown, a self-contained HTML file using plain HTML, CSS, SVG and vanilla JavaScript, a notebook, a spreadsheet, or a static image. Ask for the format you need; if you do not, the agent picks one the host supports.
The skill does not produce React or Vite frontend dashboards, and it does not install Node or frontend packages for analytics output.
- Markdown for short answers and notes
- Self-contained HTML file for a shareable report or dashboard
- Notebook for reproducible analysis
- Spreadsheet or static chart image
What the skill needs and what it does not provide
The SKILL.md does not mandate specific software. Some referenced workflows may expect Python, notebooks, or access to data sources that your host environment provides. The skill itself does not connect to warehouses or Google Workspace. Report delivery to Google Docs, Google Slides or PDF is optional and depends on the tools available in your host.
The skill runs scripts and does not use the network, according to its listing. Only use it on data and destinations your host already allows.
- Runs scripts: yes
- Uses network: no
- Data sources: only those the host environment exposes
- Optional: Google Docs, Google Slides or PDF tooling for report delivery
Safety notes before you rely on the output
The skill does not verify your source data. Keep source provenance and uncertainty in view when you read results. The skill is also built to avoid silently replacing a real-data question with sample data, and it does not bypass host policy or perform unapproved external writes.
Do not make legal, medical or financial decisions on the agent's output alone. Check the data and the reasoning yourself first.
- Verify the source data before acting on conclusions.
- Ask the agent to state which data it used and what it could not check.
- Review any script the agent runs, since the skill runs scripts.
Tips for better results
Give the agent the grain, the time window and the metric definitions. The skill's workflows start with intended use and grain, so these details shape the checks it runs.
If a metric is already defined, say so. If it is not, ask for a KPI design or business-context step first. Ask for a QA pass on a finished report, since the skill has a separate validation step.
- Name the question and decision the analysis supports.
- Provide the data file or the host source, and its time range.
- Ask for source notes and uncertainty to be included in the output.
- Request a specific output format if you need one.
What people use it for
- Check a sales or events export for completeness and duplicates before reporting on it.
- Diagnose a week-over-week drop in a product metric and list likely causes with evidence.
- Design a KPI tree with guardrail metrics for a new feature.
- Produce a self-contained HTML monthly KPI report with charts.
- Create a reproducible Python or Matplotlib notebook for a recurring analysis.
Frequently asked questions
Is the Data Analytics agent skill free?
The item is listed under the MIT licence. The source does not state pricing for the skill itself, so check the directory page and the licence file for terms.
How do I install the Data Analytics skill?
Download the ZIP from this page and unzip it 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.
What do I need to run it?
The SKILL.md does not mandate any fixed software. Some workflows may expect Python, notebooks or data access that your host environment provides. Google Docs, Slides or PDF tooling is optional for report delivery.
Can the skill connect to my data warehouse or Google Workspace?
Not out of the box. The skill uses data sources and tools only when the host exposes them, and it does not supply connectors. Its listing says it does not use the network.
Why does the skill not produce React dashboards?
The bundle explicitly excludes React, Vite and frontend build artifacts. Dashboards are delivered as plain HTML, CSS and vanilla JavaScript, Markdown, notebooks, spreadsheets or static images.
Can I trust the numbers it produces?
The skill keeps source provenance and uncertainty visible, but it does not verify your source data. Check the inputs yourself, especially for legal, medical or financial decisions.
How does it compare with a BI tool?
It suits teams producing recurring KPI reports or simple dashboards without a BI tool, and developers wanting reproducible Python or Matplotlib output. If you need interactive frontend dashboards or live warehouse connections, pick a BI platform or a different tool.
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 or business analyst, or a small team, who wants consistent, evidence-backed analysis and portable reports from an AI agent. Pick something else if you need live warehouse connections, interactive React dashboards, or a tool that verifies source data for high-stakes decisions.
Download the skill