スキル一覧に戻る
natemoovs

moovs-walkthrough-capture

by natemoovs

0🍴 0📅 2026年1月25日
GitHubで見るManusで実行

SKILL.md


name: moovs-walkthrough-capture description: Captures animated GIF walkthroughs of Moovs application workflows. Use when documenting features, creating training materials, or building help content. Automates browser navigation, screenshots, and GIF generation using Playwright.

Moovs Walkthrough Capture Skill

This skill automates the creation of animated GIF walkthroughs for the Moovs platform. It navigates through user workflows, captures screenshots at key steps, and stitches them into shareable GIFs.

Why This Matters

Documentation that shows rather than tells is more effective. When onboarding operators, training dispatchers, or creating help content, animated walkthroughs:

  • Reduce support tickets by showing exactly how to complete tasks
  • Create consistent training materials
  • Document features for release notes and marketing
  • Capture bug reproduction steps for engineering

When to Use This Skill

Use this skill when:

  • Creating training materials for operators
  • Documenting a new feature for release notes
  • Building help center content
  • Capturing a bug reproduction workflow
  • Creating sales demo materials

Prerequisites

Before running walkthroughs, ensure:

  1. Python environment with required dependencies:

    pip install -r .claude/skills/moovs-walkthrough-capture/requirements.txt
    
  2. Playwright browsers installed:

    playwright install chromium
    
  3. Access credentials for the Moovs environment you're capturing (staging recommended)

Process Overview

  1. Define the workflow - Describe what user journey to capture
  2. Configure capture settings - Environment URL, viewport, timing
  3. Execute capture - Playwright navigates and screenshots each step
  4. Generate GIF - Stitch screenshots with configurable timing
  5. Output - GIF saved to specified location

Workflow Definition

Workflows are defined as a sequence of steps. Each step includes:

workflow:
  name: "Create New Trip"
  environment: staging
  viewport: { width: 1280, height: 800 }

  steps:
    - action: navigate
      url: /dispatch
      screenshot: true
      label: "Dispatch Dashboard"

    - action: click
      selector: "[data-testid='new-trip-btn']"
      wait_for: "[data-testid='trip-form']"
      screenshot: true
      label: "New Trip Form"

    - action: fill
      selector: "[data-testid='passenger-name']"
      value: "John Smith"
      screenshot: true
      label: "Enter Passenger"

    - action: click
      selector: "[data-testid='vehicle-select']"
      screenshot: true
      label: "Select Vehicle"

    - action: click
      selector: "[data-testid='save-trip']"
      wait_for: ".success-toast"
      screenshot: true
      label: "Trip Created"

Available Actions

ActionDescriptionRequired Fields
navigateGo to a URLurl
clickClick an elementselector
fillType into a fieldselector, value
selectChoose dropdown optionselector, value
waitPause for timingduration (ms)
scrollScroll to elementselector
hoverHover over elementselector

Selectors

Use these selector strategies (in order of preference):

  1. data-testid (most reliable): [data-testid='trip-form']
  2. Aria labels: [aria-label='Create Trip']
  3. CSS selectors: .btn-primary, #submit-btn
  4. Text content: text=Create Trip

Refer to references/moovs-routes.md for URL patterns and references/common-workflows.md for pre-built workflow definitions.

GIF Configuration

gif_settings:
  frame_duration: 2000 # ms per screenshot
  transition: fade # none, fade, slide
  loop: true
  max_width: 800
  quality: high # low, medium, high
  annotations: true # add step labels

Running a Capture

Interactive Mode

When you invoke this skill, describe the workflow you want to capture:

"Capture a GIF of creating a new trip in the dispatch dashboard"

The skill will:

  1. Identify the workflow steps needed
  2. Show you the proposed sequence for approval
  3. Execute the capture
  4. Generate the GIF

Script Mode

For repeatable captures, run directly:

python .claude/skills/moovs-walkthrough-capture/scripts/capture_workflow.py \
  --workflow "new-trip" \
  --env staging \
  --output ./walkthroughs/new-trip.gif

Pre-defined Workflows

See references/common-workflows.md for ready-to-use workflows:

  • new-trip - Create a new trip in dispatch
  • edit-reservation - Modify an existing reservation
  • assign-driver - Assign a driver to a trip
  • customer-booking - Customer portal booking flow
  • shuttle-schedule - View and manage shuttle schedules

Output

Generated GIFs are saved to:

  • Default: ./walkthroughs/{workflow-name}-{timestamp}.gif
  • Custom: Specify with --output flag

Each capture also generates:

  • {name}-frames/ - Individual screenshots
  • {name}.json - Workflow metadata and timing

Troubleshooting

Common Issues

IssueSolution
Element not foundWait for element to load, check selector
Screenshot blankAdd wait step after navigation
GIF too largeReduce max_width or quality
Timing feels offAdjust frame_duration

Debug Mode

Run with verbose logging:

python scripts/capture_workflow.py --workflow "new-trip" --debug

Starting the Process

When the user invokes this skill:

  1. Ask: "What workflow would you like to capture? This could be a specific user journey like 'creating a trip' or a feature demo."

  2. Based on their response:

    • If it matches a pre-defined workflow, offer to use it
    • Otherwise, help define the step sequence
  3. Confirm the workflow steps before executing

  4. Run the capture and report results

Important: Always capture on staging first. Never run automated captures against production without explicit approval.

Reference

スコア

総合スコア

50/100

リポジトリの品質指標に基づく評価

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

レビュー

💬

レビュー機能は近日公開予定です