Inspiration
HomeCompass is built to bring privacy-first, local-first intelligence to home organization and emergency readiness using UGOS AI NAS.
Proposal & Requirements (10 Required Questions)
- Project Purpose: Provide a secure, offline household chief-of-staff and emergency document vault.
- Target Audience: Families and individuals requiring local-first data privacy and offline accessibility.
- Core Features: Offline OCR document parsing, local AI search, disaster kit checklist generator, and emergency contact vault.
- AI NAS Integration: Operates natively on UGOS AI NAS leveraging local LLMs and vector indexing.
- Skill Standard Compliance: Adheres strictly to the Uliya Skill Development Standard / UGREEN Skill Developer Package guidelines.
- Technical Architecture: Built natively on UGOS leveraging local embedding models and vector indexing.
- Data Privacy & Security: 100% air-gapped processing; zero external cloud dependencies or phone-home analytics.
- Implementation Roadmap: Prototype created during hackathon; full skill deployment planned.
- Challenges & Learnings: Navigating local AI constraints and optimizing prompt workflows.
- Future Expansion: Integration with smart home sensors and automated emergency alert sync.
How we built it
HomeCompass is built using Google Antigravity and Gemini AI to architect the agent workflow, prompt logic, and declarative SKILL.md specification for UGREEN's Uliya AI engine and UGOS CLI tools without writing manual code.
Log in or sign up for Devpost to join the conversation.