Skip to content
Get packs

Build a Dashboard

Create a source-backed, reusable monitoring dashboard with defined metrics, filters and QA checks

Source Download skill
Repository
XiaomiMiMo/MiMo-Code
Stars
13,743
Licence
MIT
Files
5 files

What it does

When installed, the agent follows a structured process for building a monitoring or exploration dashboard instead of a one-off report. It first pins down who the dashboard is for and which decisions it supports, defines each metric precisely (formula, time window, filters, owner, freshness), checks the underlying data for joins, duplicates and gaps, and then lays out the page from overview to drill-down. It can build the dashboard as a Streamlit app, a self-contained HTML/JavaScript page, or a connection to an existing BI platform, and it reconciles totals and tests edge cases before handing it over.

Great for

  • Analysts who need a shared, repeatable KPI view rather than a single report
  • Developers building a small internal monitoring page for product or operations metrics
  • Teams that want metric definitions and data freshness shown alongside the charts

Skip it if

  • One-time ad hoc questions, where a simple analysis or report is enough and a full dashboard adds overhead
  • Projects that must use a specific framework such as React, since the skill explicitly avoids requiring one and may choose plain HTML or Streamlit instead; the user should state their preference
  • Sensitive data displays without review: the skill asks the agent to avoid exposing secrets and personal data, but a human must still check what the published dashboard reveals

Try asking

  • “Build a weekly monitoring dashboard for signups, activation and churn from this CSV, with metric definitions on the page”
  • “Create a single-file HTML dashboard I can open locally to track our support ticket backlog and response times”
  • “Make a Streamlit app that filters sales revenue by region and date range, with a data freshness note”
  • “Set up an executive overview page for our marketing KPIs with drill-down tables for each channel”
Step-by-step guide: how to use Build a DashboardInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 5 original files, the MIT licence and an attribution note.

    build-dashboard.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip build-dashboard.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip build-dashboard.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • State your preferred surface (Streamlit, plain HTML, or your BI tool) and your audience up front so the agent does not have to guess
  • Provide metric definitions or point to existing KPI documents; the skill asks the agent to read them first when they are not settled
  • Ask for a QA pass with empty and error states and check that totals reconcile against a source query before sharing the dashboard

What's inside

  • agents/openai.yaml
  • SKILL.md
  • specifications/bi-platform-dashboard.md
  • specifications/html-dashboard.md
  • specifications/streamlit-dashboard.md

Redistributed unmodified from XiaomiMiMo/MiMo-Code under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.