Inspiration
After a year or more of daily work, almost every home or freelancer NAS turns into a maze of scattered Downloads, half-sorted "Master" folders, duplicated invoices, and forgotten scripts. Traditional file search only lists matches—it never asks which copy you actually work on, never protects you from accidental deletion, and never remembers your personal cleanup habits.
What it does
Smart NAS Organizer (smart-nas-organizer) is an interactive UGOS AI Skill that turns chaotic NAS folders into a clean, searchable workspace tailored to user decisions:
Interactive Duplicate Resolution: Detects exact duplicates by hash (scan-dedupe / sha256sum), asks via waiting_user which copy is the active working file, and offers safe options for the rest—including moving them to _Archive/Duplicates with a restore manifest. Multi-Parameter Document & Script Views: Lets users browse files by source folder, name, date, format, or content category (Invoices, Books, Technical_Docs, Scripts_Code, Contracts_Legal, Notes_Reports) without moving anything prematurely. Guided Small-Batch Segregation: Collects scattered files into shallow 2–3 level folders inside Documents (or a custom target) after a dry-run preview and waiting_approval checkpoint, plus optional OCR scanning of photographed receipts via ug-photo. Personalized Shortcuts & Memory: Saves user preferences, custom categories, and one-phrase shortcuts in a Markdown profile on the NAS (ug-knowledge) and uses ug-filemgr tags for virtual views without duplicating files.
How we built it
We designed smart-nas-organizer as a pure Markdown orchestration Skill compliant with uliya-skill-kit (catalog v2.2.0). The parent SKILL.md delegates self-contained tasks exclusively to registered UGOS experts:
file-agent (ug-filesystem, ug-filemgr, file-workflow) document-agent (ug-document) photo-agent (ug-photo) knowledge-agent (ug-knowledge)
Detailed protocols for duplicate quarantine, shallow folder hierarchies, personalized shortcuts, Round 1/2 validation, and AI/sandbox disclosures live in modular references/*.md files.
Challenges we ran into
Designing a cleanup workflow that avoids common NAS pitfalls: preventing user burnout from "all-at-once" sorting, avoiding deep 6–10 level folder trees that break path limits, ensuring all file moves go through native UGOS experts so the File Manager index stays in sync, and adhering strictly to uliya-skill-kit static validation rules.
Accomplishments that we're proud of
Achieved static-pass with 0 errors and 0 warnings in uliya-skill-kit/scripts/check-skill.mjs. Built an automated local verification suite (verify_submission.py) validating catalog mappings, ZIP limits, and proposal requirements. Designed a safety-first UX where no file is ever moved or deleted without a dry-run preview and explicit platform approval.
What we learned
Effective NAS automation is not about aggressive auto-deletion—it is about human-in-the-loop safety, shallow hierarchies, tag-based virtual views, and remembering the user's own organizational decisions over time.
What's next for Smart NAS Organizer
In Round 2 on real UGREEN AI NAS (iDX6011 Pro) hardware, we will run end-to-end validation across multi-folder chaos scenarios, tune OCR-assisted receipt discovery, and record a 3-minute YouTube walkthrough demonstrating interactive duplicate archiving and saved shortcuts.
Built With
- ai-nas
- markdown
- ugos
- uliya-skill-kit
