What does this Ability do?

Voice-activated morning briefing that fetches weather, Google Calendar, and Gmail in parallel, then uses an LLM to synthesize the data into a single ~60-second spoken summary. Weather is personalized per user via IP geolocation. Demonstrates real-world OAuth 2.0 integration with Google APIs and parallel async data fetching.

Suggested Trigger Words

• good morning
• give me my brief
• daily brief
• brief me
• start my day
• what did I miss

Type

• New community Ability
• Improvement to existing Ability
• Bug fix
• Documentation update

External APIs

• No external APIs
• Uses external API(s):
  • Open-Meteo API (weather) - Free, no API key required
  • Google Calendar API - OAuth 2.0 (developer's personal account for demo)
  • Gmail API - OAuth 2.0 (same credentials as Calendar)
  • ip-api.com (geolocation) - Free, no API key required

Testing

• Tested in OpenHome Live Editor
• All exit paths tested (said "stop", "exit", etc.)
• Error scenarios tested (API down, bad input, etc.)

Testing Details:

• ✅ Parallel fetching verified (all APIs called concurrently)
• ✅ Token refresh mechanism tested with expired tokens
• ✅ Graceful degradation tested (ability works with partial data if one API fails)
• ✅ Follow-up commands tested ("repeat", "check my calendar", exit words)
• ✅ Location auto-detection and caching verified
• ✅ LLM synthesis produces natural ~60 second briefings
• ✅ Speech splitting works correctly (1-2 sentences per speak() call)

Checklist

• Files are in community/google-daily-brief/
• main.py follows SDK pattern (extends MatchingCapability, has register_capability + call)
• README.md included with description, suggested triggers, and setup
• resume_normal_flow() called on every exit path
• No print() — using editor_logging_handler
• No hardcoded API keys — using placeholders with comments
• No blocked imports (redis, connection_manager, user_config)
• No asyncio.sleep() or asyncio.create_task() — using session_tasks
• Error handling on all external calls

Anything else?

Key Features:

• Parallel Data Fetching: Uses asyncio.gather() to fetch weather, calendar, and email concurrently, completing in 2-3 seconds
• LLM Synthesis: Processes raw API data into natural, conversational spoken briefing
• Automatic Token Refresh: Handles expired Google OAuth tokens automatically and retries failed requests
• Smart Speech Splitting: _split_for_speech() method breaks long briefings into 1-2 sentence chunks for natural voice delivery
• Graceful Degradation: Individual API failures don't crash the ability - continues with available data
• Location Persistence: Auto-detects and caches user location to avoid repeated API calls
• Follow-up Commands: Supports "repeat" to replay briefing and "check my calendar" for calendar-only view

Technical Highlights:

• Demonstrates proper OAuth 2.0 implementation with refresh token flow
• Shows best practices for parallel async operations in OpenHome
• Implements persistent storage for user preferences
• Uses LLM for data synthesis (not just API readout)
• Voice-first design throughout (short responses, natural language)

Demo Note: This ability uses the developer's personal Google account credentials (hardcoded) as a demonstration. In production, each user would connect their own Google account via OAuth flow.

Performance: First spoken word within 6 seconds of trigger (2-3s data fetch + 2-3s LLM synthesis).

Demo Video: https://www.loom.com/share/cf4374e3211a479d9e334e91fb1d45a0