dashboard-build（仪表板构建）

CRITICAL Guidelines for Dashboard Building

• Ideally, do NOT skip Phase 1 (dashboard-design skill) as that ensures a much smoother build process. If the user just wants to build it, explain the value of a design phase first. Check if there is any existing /spec directory from Phase 1. If yes, use that to guide your build.
• Use your native tools to understand the data well, especially if you build custom charts or when you use specific selectors.
• If the user asks for an example, simply copy the example app and run it. Do not include your own data or change the example.
• When executing any script mentioned below for the first time, it may take a while to install dependencies. Plan accordingly before taking any rash actions.
• When iterating on the dashboard after completing all steps, do not forget key points from below, especially regarding spec compliance and updating and terminal handling: always keep all specs up to date, and always check if terminal output is clean after each iteration.
• Execute all scripts from this skill, and the app.py you will create, with uv run <script_name>.py or uv run app.py - this will ensure you use the correct dependencies and versions.
• ABSOLUTELY NEVER type ANY commands (including sleep, echo, or anything else) in the terminal where the dashboard app is running, even if you started it with isBackground=true. This WILL kill the dashboard process. The dashboard startup takes time - be patient and let it run undisturbed.

Spec Files: Documenting Decisions

IMPORTANT: Each step produces a spec file in the spec/ directory to document reasoning, enable collaboration, and allow resumption in future sessions. Create the spec/ directory if it is not already present at the root of the project.

Step 1: Build dashboard

1. You MUST ALWAYS copy the example app over, and modify it - this ensures less errors!
2. Investigate about the Vizro model by executing the schema fetching script. ALWAYS DO this for all models that you need - do NOT assume you know it. Execute the script like so: uv run./scripts/get_model_json_schema.py <model_name> <model_name2>... where <model_name> is the name of the model you want to get the schema for. You can get an overview of what is available by calling the overview script like so: uv run./scripts/get_overview_vizro_models.py. This will print out all available models and their brief descriptions.
3. Build the dashboard config by changing the copied example app. Important: Very often normal plotly express charts will not suffice as they are too simple. In that case, refer to the custom charts guide to create more complex charts. These MUST be added to the correct section in the python app. Call the custom chart function from the Graph model in your dashboard app.
4. Run your dashboard app with uv run <your_dashboard_app>.py CRITICAL: After running this command, DO NOT run ANY other commands in that terminal. The dashboard takes time to start up (sometimes 10-30 seconds)
5. You MUST read the terminal to check for any errors, but do not put commands like sleep in it. Fix any warnings and even more important errors you encounter. ONLY once you see the dashboard running, inform the user. NEVER run any commands in that terminal after starting the dashboard.
6. When you iterate, no need to kill the dashboard, as we are using debug mode. Just save the file and it will reload automatically. Check the terminal occasionally for any failures. Once failed, you need to restart the dashboard.

Optimizations and common errors

• Colors: Always respect color decisions from the design phase (spec/3_visual_design.yaml). Use Vizro palettes (from vizro.themes import palettes) instead of generic plotly colorscales. This is especially easy to get wrong in custom charts.
• Data loading: For dashboards needing data refresh (databases, APIs) or performance optimization, see the data management guide for static vs dynamic data, caching, and best practices.
• KPI cards: Use kpi_card() for simple metrics, kpi_card_reference() for comparisons. Use reverse_color=True when lower is better (costs, errors). NEVER put kpi_card or kpi_card_reference as a custom chart or re-build KPI cards as custom charts, use the built-in kpi_card and kpi_card_reference (imported from vizro.figures) in Figure model instead. Only accept exceptions for when the KPI card is strictly not possible, for example when dynamically showing text as a KPI card.

REQUIRED OUTPUT: spec/4_implementation.yaml

Save this file BEFORE proceeding to Step 2:

# spec/4_implementation.yaml
implementation:
  app_file: <name>.py
  data_files:
    - [list of data files used]
  data_type: static/dynamic  # static for DataFrames, dynamic for data_manager functions
  data_sources:
    - name: [data source name]
      type: csv/database/api/function
      caching: true/false
      refresh_strategy: [if dynamic: cache timeout or refresh trigger]

spec_compliance:
  followed_specs: true/false
  deviations:
    - spec_item: [What was specified]
      actual: [What was implemented]
      reason: [Why the deviation was necessary]

custom_charts:
  - name: [Function name]
    purpose: [What it does]

Validation Checklist

Before proceeding to Step 2, verify against spec files:

• All specs from spec/1_information_architecture.yaml, spec/2_interaction_ux.yaml and spec/3_visual_design.yaml are implemented if specs exist
• You have read the terminal output of the dashboard app for errors and warnings, you have not put any commands in the terminal after starting the app
• Any deviations are documented in spec/4_implementation.yaml

Step 2: Testing (optional)

This requires the Playwright MCP server to be running. If not available, inform the user and skip this step. Look for mcp__*playwright__* tools.

When conducting the below tests, feel free to go back to Step 1 to fix any issues you find, then come back here.

Basic Testing Flow

1. Navigate to dashboard URL
2. Click through all pages
3. Check console for errors

Use: playwright:browser_navigate(url="http://localhost:8050")
Use: playwright:browser_click(element="Page Name", ref="...")
Use: playwright:browser_console_messages()

Advanced Testing flow

1. Take a screenshot of each page, compare to specs and especially wireframes
2. Document any discrepancies

Important things to check:

• Line charts are readable, and not a mess due to lack of aggregation
• Graphs are legible and not squashed due to Layout

REQUIRED OUTPUT: spec/5_test_report.yaml

Save this file to complete the project:

# spec/5_test_report.yaml
testing:
  launch:
    successful: true/false
    url: http://localhost:8050
    errors: []

  navigation:
    all_pages_work: true/false
    issues: []

  console:
    no_errors: true/false
    errors_found: []

screenshot_tests:
    performed: true/false
    pages_tested: []
    discrepancies:
        - page: [Page name]
            issue: [Description of visual issue]
            fixed: true/false
            notes: [Fix details or reason not fixed]

requirements_met: true/false
dashboard_ready: true/false

Done When

• Dashboard launches without errors, no console errors
• User confirms requirements are met
• All spec files from this Phase 2 saved in spec/ directory

Reference Files