AGENT SKILL

create-dashboard

Builds comprehensive Amplitude dashboards from requirements or goals, organizing charts into logical sections with appropriate layouts. Use when creating a complete dashboard from scratch or assembling existing charts into a cohesive view.

License 许可
MIT
最近更新
2026-11-13
Stars
★ 29
Forks
156

01 前言 Front Matter

name
create-dashboard
description
Builds comprehensive Amplitude dashboards from requirements or goals, organizing charts into logical sections with appropriate layouts. Use when creating a complete dashboard from scratch or assembling existing charts into a cohesive view.

03 SKILL.md 正文

Create Dashboard

Create new team or initiative dashboards, organize scattered charts, build executive reporting, or set up review cadence dashboards.

Instructions

Step 0: Discovery (if unfamiliar with the feature)

Before building, understand what you're tracking:

  • Search for existing dashboards/charts related to the topic
  • Search for relevant events with Amplitude:search_amp_data_taxonomy
  • Use get_properties to understand available properties for segmentation
  • Ask user for clarification on primary goals, key segments, or time horizons

Step 1: Define Dashboard Purpose

Clarify:

  • Who is the audience?
  • What decisions will it inform?
  • How frequently will it be reviewed?
  • What's the narrative structure?

Step 2: Gather or Create Charts

If existing charts found (>5 relevant):

  • Use Amplitude:search_amp_entities to find relevant existing charts
  • Use Amplitude:get_amplitude_charts with include: "definition" to retrieve their definitions
  • Identify gaps that need new charts

If few/no charts exist (<5 relevant):

  • Switch to "greenfield build" mode
  • Use Amplitude:query_amplitude_data to create needed charts
  • Keep the returned chart edit IDs; use_amp_dashboards promotes them when creating the dashboard
  • Consider searching for relevant events first with entityTypes: ["EVENT", "CUSTOM_EVENT"]

Creating new charts:

  • Prototype with query_amplitude_data to verify data
  • Collect the returned chart edit IDs before creating the dashboard

Step 3: Plan the Layout

Organize into logical sections:

  1. Summary Row: Key metrics at a glance (headline view)
  2. Trend Section: How things are changing
  3. Breakdown Section: Segments and dimensions
  4. Detail Section: Supporting analyses

Step 4: Create the Dashboard

Use Amplitude:use_amp_dashboards with action: "create" and:

  • Clear, descriptive name
  • Rows with appropriate heights (375, 500, 625, or 750px)
  • Charts sized appropriately (3-12 columns)
  • Rich text headers for sections
  • Chart display configurations in chartMetas:
    • metric_only: Headline KPIs (single number)
    • series: Trend lines (default view)
    • converted: Funnels (conversion view)
    • table: Data tables

Step 5: Add Context

Include rich text blocks for:

  • Dashboard purpose and audience
  • How to interpret key metrics
  • Links to related resources
  • Last updated or review schedule

Layout Guidelines

Content Type Suggested Width Suggested Height
Headline metric 3-4 columns 375px
Trend chart 6-12 columns 500px
Comparison chart 6 columns 500px
Detailed table 12 columns 625px
Section header 12 columns 375px

Best Practices

  • Put most important metrics above the fold
  • Use consistent chart sizing within rows
  • Group related metrics together
  • Add explanatory text for complex metrics
  • Ask user about focus areas if multiple valid approaches exist
  • Create charts in batches to minimize tool calls

Common Issues

Query errors (500/400):

  • Simplify: remove complex groupBy, reduce date ranges, avoid nested properties
  • Verify events/properties exist using search first
  • Use eventsSegmentation with groupBy instead of dataTableV2 for top N lists

No data returned:

  • Check event names are exact matches (case-sensitive)
  • Verify date range covers when events were tracked
  • Confirm user segments aren't too restrictive

04 同仓库更多 Skills

CognitionAI
CognitionAI/devin-marketplace

business-data

暂无简介

★ 29
CognitionAI
CognitionAI/devin-marketplace

workflow

Creates durable, resumable workflows using Vercel's Workflow SDK. Use when building workflows that need to survive restarts, pause for external events, retry on failure, or coordinate multi-step operations over time. Triggers on mentions of "workflow", "durable functions", "resumable", "workflow sdk", "queue", "event", "push", "subscribe", or step-based orchestration.

★ 29
CognitionAI
CognitionAI/devin-marketplace

workflow

Vercel Workflow SDK expert guidance. Use when building durable workflows, long-running tasks, API routes or agents that need pause/resume, retries, step-based execution, or crash-safe orchestration with Vercel Workflow.

★ 29
CognitionAI
CognitionAI/devin-marketplace

verification

Full-story verification — infers what the user is building, then verifies the complete flow end-to-end: browser → API → data → response. Triggers on dev server start and 'why isn't this working' signals.

★ 29
CognitionAI
CognitionAI/devin-marketplace

vercel-storage

Vercel storage expert guidance — Blob, Global Config (formerly Edge Config), and Marketplace storage (Neon Postgres, Upstash Redis). Use when choosing, configuring, or using data storage with Vercel applications.

★ 29
CognitionAI
CognitionAI/devin-marketplace

vercel-services

Configure and troubleshoot Vercel Services for multiple frontends and backends in one project. Use when composing a polyglot or multi-service application on one Vercel deployment; defining the `services` key, service-targeted rewrites, or service bindings in `vercel.json`; or running all services with `vercel dev`.

★ 29