Moje Shia – Movement Aftercare

Inspiration

I work as a massage therapist. After a session, a client needs a clear reminder of the simple movements discussed with their therapist. I wanted to make that handover quick enough to fit into the last moments of an appointment.

What it does

Moje Shia turns selected treatment areas into a small home movement plan. The therapist selects the neck, trapezius, shoulders, or forearms on a body map or with labelled controls. The app previews the matching movements and creates a client-facing plan with a shareable link and QR code.

The core flow is: select areas, review the matched movements, create a plan, and open the client's view. The client view includes plain-language instructions, repetition guidance, and stop conditions. No client name or medical record is required; the shared URL contains the selected body areas.

Scope

This is a standalone competition proof of concept. It is separate from my working practice-management application and does not use real client records. It demonstrates one focused handover flow rather than a full booking system or patient database.

The demonstrated movement matching should not be confused with AI diagnosis. The app presents general aftercare for therapist review and does not replace medical assessment.

What I learned

A useful proof of concept can be small. The important part is making the whole journey understandable, from the therapist's selection to the client's instructions. Keeping names and medical records out of this flow also makes the demonstration easier to share without exposing clients.

Challenges

The main product challenge was keeping the therapist's task short while making the client view clear on a phone. Another was separating this competition prototype from the application and records used in my real practice.

Next steps

Validate the wording and movement library with appropriate professional review, and gather feedback on the therapist-to-client handover before expanding the scope.

How it was built and source provenance

The separate competition prototype was created in October 2026. On October 23, its visible flow was used as the reference for a newly written HTML, CSS and JavaScript reconstruction because the original source could not be retrieved. This public repository contains the reconstruction, not recovered original source. It carries forward the four-area interaction and movement wording; it uses no production application code or client records.

Codex assisted with planning, implementation and documentation using the official Devpost Learn Skill Pack. The repository includes scope.md, prd.md, spec.md, a verification checklist and source provenance. QRCode.js is bundled under its MIT license. Matching uses a fixed catalogue; no runtime AI API or database is required. The earlier hosted prototype and the new repository are separate builds.

Built With

• web