Automates browser-based interactions with Umbraco.AI demo site using Playwright. Handles login, navigation, and editing of AI entities (connections, profiles, prompts, agents). Use when testing UI workflows, creating or editing AI entities through the backoffice, or demonstrating Umbraco.AI features.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.
Automates browser-based interactions with Umbraco.AI demo site using Playwright. Handles login, navigation, and editing of AI entities (connections, profiles, prompts, agents). Use when testing UI workflows, creating or editing AI entities through the backoffice, or demonstrating Umbraco.AI features.
This skill depends on the demo-site-management skill for infrastructure:
# Typical workflow
/demo-site-management start # Start the demo site
/demo-site-automation login # Login via browser
/demo-site-automation navigate-to-connections # Navigate to section
/demo-site-automation create-connection OpenAI # Create entity
Success Criteria
After login: User sees Umbraco dashboard
After navigation: Correct section visible with list of entities
After create: New entity appears in list with success notification
After edit: Changes saved and reflected in list
After status: Clear report of demo site and browser state
Example Usage
# Start demo site and login
/demo-site-management start
/demo-site-automation login
# Create a connection
/demo-site-automation create-connection OpenAI
# Navigate and edit
/demo-site-automation navigate-to-profiles
/demo-site-automation edit-profile "Default Chat"# Work with add-on packages
/demo-site-automation navigate-to-prompts
/demo-site-automation create-prompt
# Check status anytime
/demo-site-automation status
Tips
Always check if demo site is running before browser interactions
Take snapshots frequently to understand page state
Use browser_snapshot instead of browser_take_screenshot for actions (faster, more accurate)
Wait 2-3 seconds after navigation for Lit components to render
If elements not found, take snapshot and adjust selectors
Use browser_evaluate for complex queries (e.g., counting entities)
Test with different providers (OpenAI, Anthropic, Amazon, Google)