Assessment architecture
The item bank uses an original math-readiness blueprint with SAT Math domain proportions scaled to Algebra, Advanced Math, Problem-Solving and Data Analysis, and Geometry and Trigonometry.
Project Brief
Assessment-to-Remediation Pipeline
A public-safe prototype for connecting math diagnostic evidence to feedback, remediation, mastery checks, and decision-support reporting.
Overview
From diagnostic design to instructional action
This project shows an inspectable workflow for moving from a standards-aligned math-readiness diagnostic to reviewable instructional artifacts. The current public MVP focuses on assessment authoring, preview generation, automated review, and offline Canvas New Quizzes payloads.
The workflow is intentionally local, static, and reproducible. It does not claim to be a validated standardized test, a live LMS deployment, or a production Canvas integration.
Current Artifacts
- 36 original math-readiness diagnostic items
- Two 30-minute modules scaled to a one-hour form
- Student-facing HTML preview
- Instructor review HTML preview
- Automated review gate with passing public artifact set
- GPT-5.5 item review marked
POLISH_RECOMMENDED - GPT-5.5 visual review with no presentation blocker
- Offline Canvas New Quizzes payload with 36 included items
- Public-safe project brief and portfolio case-study draft
System Proof
What this project demonstrates
Review-gated release
Generated previews, deterministic checks, GPT-5.5 item review, GPT-5.5 visual review, and an automated review gate create a public-safe release boundary.
Canvas dry-run boundary
The project generates offline Canvas New Quizzes payload artifacts and dry-run validation without touching a live LMS or using private course identifiers.
Analytics-to-action design
The planned next slice connects synthetic attempts to scoring, item/domain analysis, misconception tags, feedback, remediation recommendations, and mastery-check evidence.
Public-safety discipline
The repo forbids real student data, parent data, school-private records, grades, submissions, rosters, emails, API keys, OAuth tokens, and copied commercial assessment items.
Safety Boundary
Prototype, not production
This is a public-safe assessment-authoring and review-gate MVP. It should be described as an internal math-readiness diagnostic prototype, not a validated standardized test, live LMS deployment, or production Canvas integration.