This document outlines our high-level strategic priorities. It's a living document, not a set of unbreakable promises. For the status of individual tasks, see our GitHub Issue Tracker.
AutoCat's goal is to be:
- Intelligent: Use AI to make app organization effortless.
- Simple: Match the core Pixel Launcher experience while adding power.
- Stable: Provide a rock-solid, reliable foundation based on AutoCat 16.
- Full Branding Migration: Completed the transition from Lawnchair to AutoCat across all internal classes, file structures, and log tags.
- AI Semantic Search: Implemented intent-based searching, allowing users to find apps by purpose (e.g., searching "Finance" for banking apps).
- Dual Folder Sync: Categories now sync perfectly to both the app drawer and the home screen workspace.
- Multi-Language Intelligence: Localized AI prompts and reasoning supporting 10+ global languages.
- Settings UX Overhaul: Reorganized settings into a logical, comfortable structure with a top-level AutoCat entry and integrated Discovery cards.
- Performance & Stability: Implemented non-blocking startup initialization and a font inflation cache to eliminate UI lag.
- LLM Reliability: Robust exponential backoff and "Smart Fallback" (Circuit Breaker) for AI providers.
This is our active development sprint. The goal is to polish the new AI features and gather community feedback on the Development 4 release.
- Optimizing database queries for very large app lists (>500 apps).
- Refining "Auto-Pilot" model selection based on real-world accuracy tracking.
Once the Development 4 foundation is fully stable, our focus will shift to deeper launcher customization.
- Proper icon swipe gestures (AutoCat style).
- Folder "cover" mode.
- Fuzzy search sensitivity slider.
We have successfully rebased onto the latest Android 16 (AOSP) source. AutoCat is now tracking
the 16-dev branch as its primary development target.
Highly-requested features that are blocked by external dependencies or require significant research.
- Widget Stacking: A highly complex feature requiring deep architectural investigation.
- On-Device ML: Exploring local categorization models to reduce API dependency.