Student Achievement Timeline

A UGREEN AI NAS skill that turns a family's photos, certificates and school documents into a checked activities and honors record for college applications, with one album per activity.

Comment

Inspiration

Since 2022 I've helped high school students and their parents with US college applications, first through videos and one-on-one calls and now through my company, Clarity Prep. Every fall I see the same scramble. The Common App asks for up to 10 activities and 5 honors, each with grade levels, hours and a 150-character description. Families can't remember four years of clubs, competitions, volunteering and awards, and the proof is scattered across years of phone backups. On a family NAS, those photos and scanned certificates are already in one place. They just aren't organized for the moment they matter most.

What it does

Student Achievement Timeline is a Skill for UGREEN AI NAS. A parent says something like "Build Sam's activities list for college apps, he started 9th grade in August 2023, look in Family Documents." The Skill then:

  • confirms the student in the NAS People albums
  • finds the student's photos school year by school year and describes what each event shows
  • finds photos of certificates and trophies with text recognition
  • reads award letters, honor rolls, programs and volunteer logs in the folders the user approved
  • maps every date to a grade, drafts each activity and honor inside the application's character limits, and attaches the evidence ("14 photos, 2 documents")
  • asks only for what the files can't show (hours per week, role), then, after approval, creates one album per activity and saves the final record as a Markdown file on the NAS

It never invents an activity, date or hour count. It leaves out certificates printed with a sibling's name and reports empty school years as empty. A minor's photos and report cards never leave the device.

How we built it

The Skill follows the Uliya Skill Development Standard: one SKILL.md entry point plus four Markdown references (application format rules, privacy and confirmation, test plan, disclosure). The parent Skill only delegates to registered NAS experts: ug-photo to photo-agent, ug-filesystem and file-workflow to file-agent, ug-document to document-agent, and ug-knowledge to knowledge-agent. Every capability was checked against the official capability catalog, and the package passes the uliya-skill-kit 0.7.3 static checker with 0 errors and 0 warnings. I built it with Claude (Anthropic) in Claude Code, the assistant the starter kit recommends. Claude drafted the Skill files and the proposal and ran the checker; AI use is disclosed in the package.

Challenges

  • Writing delegation steps that pass only real results forward (real photo identities, real file paths) instead of guessed paths.
  • Deciding what the Skill should refuse to fill. Hours and roles are the fields families most want an AI to make up, so the Skill asks instead.
  • Handling the edge cases that matter on a real family NAS: siblings' certificates in the same folder, blurry certificate photos, and school years with no photos.

What's next

If selected as a finalist: run the full flow on a UGREEN AI NAS with a fictional sample library, test the sibling-certificate edge case and certificate-only-as-photo case, and record the demo.

Built With

Share this project:

Updates

Submission history