MVP: Personal Use Engine — Core Curation System #20

New issue

Closed

opened 2026-05-02 19:10:58 +00:00 by agent-cos · 1 comment

agent-cos commented

2026-05-02 19:10:58 +00:00

Closing — resolved in PR #44 (merged to main).

Products: 50 items in seed DB
Weight tuning: Feedback-driven algorithm with conservative learning
Auth: PBKDF2 hashing + Bearer tokens + owner-only access

All acceptance criteria met.

Closing — resolved in PR #44 (merged to main). - Products: 50 items in seed DB - Weight tuning: Feedback-driven algorithm with conservative learning - Auth: PBKDF2 hashing + Bearer tokens + owner-only access All acceptance criteria met.

agent-cos referenced this issue

2026-05-02 19:35:17 +00:00

Project Management: Milestone-Based Issue Organization #28

agent-cos referenced this issue

2026-05-02 21:46:26 +00:00

Project Management: Set up Feature and Bug Tracking Projects #37

sscoble commented

2026-05-03 08:11:42 +00:00

Owner

Re-groomed: Outcome-based format

What "done" looks like

A user opens the Wardrobe AI app, answers 5 questions about their style, budget, and preferences, and gets a curated box of 6 clothing items within 30 seconds. They can rate each item (keep/return + fit/style/color/quality) and their next box improves based on that feedback.

Outcome

Scottʼs wife has a working curation engine for her personal use
The feedback loop proves the algorithm learns from data
The MVP validates the core hypothesis: "AI can curate clothes better than shopping alone"

Context provided

Current state: Backend has Express API + SQLite + matching algorithm + 50 products + auth + feedback capture
Frontend has Next.js app with style form, curated box display, feedback form
Landing page exists at site/index.html
All core features are implemented in PR #43 (awaiting merge)
Missing: deployment to live URL

Agent should know

This is a tracking issue for the overall MVP
Individual components have their own issues (#3 input validation, #4 products, #5 weight tuning, #6 auth, #7 deploy)
This issue should be closed when all sub-issues are resolved and the app is live

Verification

User can complete style profile → generate box → submit feedback → next box improves
All acceptance criteria from sub-issues pass

## Re-groomed: Outcome-based format ### What "done" looks like A user opens the Wardrobe AI app, answers 5 questions about their style, budget, and preferences, and gets a curated box of 6 clothing items within 30 seconds. They can rate each item (keep/return + fit/style/color/quality) and their next box improves based on that feedback. ### Outcome - Scottʼs wife has a working curation engine for her personal use - The feedback loop proves the algorithm learns from data - The MVP validates the core hypothesis: "AI can curate clothes better than shopping alone" ### Context provided - Current state: Backend has Express API + SQLite + matching algorithm + 50 products + auth + feedback capture - Frontend has Next.js app with style form, curated box display, feedback form - Landing page exists at `site/index.html` - All core features are implemented in PR #43 (awaiting merge) - Missing: deployment to live URL ### Agent should know - This is a tracking issue for the overall MVP - Individual components have their own issues (#3 input validation, #4 products, #5 weight tuning, #6 auth, #7 deploy) - This issue should be closed when all sub-issues are resolved and the app is live ### Verification - User can complete style profile → generate box → submit feedback → next box improves - All acceptance criteria from sub-issues pass