🏗️ Architecture: What Jarvis-level features to add for groundbreaking AI assi...

Date: 2026-02-27
Time: 00:22:44 UTC
Session: gspace1-149.6
Agent: angular
Confidence: 88%
Status: ✅ Documented

📋 Decision Context

Question:

What Jarvis-level features to add for groundbreaking AI assistant experience?

Task: Session 149: Phase 5 - Jarvis-level features

🎯 Decision Made

Chosen:

Created 4 services: AmbientBarComponent (persistent bottom strip with quick input, context badge, task indicator, synced across tabs via BroadcastChannel), ProactiveMemoryService (recalls relevant memories on route change, debounced 1.5s), SessionContinuityService (cross-device session handoff detection), VittoVoiceService (Web Speech API STT/TTS with wake word). Created voice.plugin.ts for plugin integration.

Options Considered:

• No alternatives documented

💭 Reasoning

Following Build UI Buy Engine: Web Speech API for voice (free, built-in, zero dependencies), BroadcastChannel for presence sync (instant, no network). ProactiveMemoryService is debounced to 1.5s to avoid spamming on rapid navigation. SessionContinuityService detects platform via window globals (TAURI_INTERNALS, Capacitor) — same proven pattern as TauriBridgeService and CapacitorBridgeService. Ambient bar is gograb-ai only initially (user's preference) to validate UX before expanding.

📁 Files Involved

• libs/libs-ai/vitto-chat/src/lib/components/ambient-bar/ambient-bar.component.ts
• libs/libs-ai/vitto-chat/src/lib/services/proactive-memory.service.ts
• libs/libs-ai/vitto-chat/src/lib/services/session-continuity.service.ts
• libs/libs-ai/vitto-chat/src/lib/services/vitto-voice.service.ts
• libs/libs-ai/vitto-chat/src/lib/plugins/voice.plugin.ts

🔗 Related

• Decision ID: dev_1772151764372_rmfmzj7
• Session: 149
• Workspace: gspace1

Auto-generated by MCP Smart Consolidation