Banya LabsBanya Labs/Apps/AcademiaTrack
In progress

App 02  ·  FIG. 002

AcademiaTrack logo

AcademiaTrack

AI-powered student academic performance portal & report card analytics

EdTechAI OCRAnalytics
App not yet live

Customer Target

Parents, Schools & Educators

Status

In progress

Agent Surface B

MCP Ready

Model Context Protocol Enabled

Registered in /.well-known/mcp.json & /llms.txt with deterministic Zod schema validation.

About Product

AcademiaTrack transforms paper and PDF school report cards into actionable academic intelligence for parents and educators.

Key Capabilities: • Automated AI OCR: Instantly scans and extracts subject marks, percentage scores, and qualitative teacher comments from uploaded report cards. • Subject Analytics Hub: Computes subject-by-subject running averages, volatility metrics, net term growth trajectories, and national CAPS level projections. • Historical Subject Timelines: Interactive deep-dive into individual subject performance history, tracking progress across multiple terms and years. • AI Sentiment & Reflections: Synthesizes teacher remarks into parent-friendly behavioral insights and time capsule milestones. • Multi-Student Management: Allows parents to seamlessly toggle between multiple children profiles and compare historic academic growth.

Tech Stack & Infrastructure: • Framework: Next.js 15 (App Router), TypeScript, Tailwind CSS • AI & Vision: Google Gemini API (Multimodal OCR & Sentiment Analysis) • Visualization: Recharts data visualizers • Storage: Firebase Firestore persistent cloud store

Dual-Surface Execution

Human vs. Agent Flow

Every action in AcademiaTrack is mapped bi-directionally across Surface A (Visual UI) and Surface B (Programmatic MCP Engine).

Surface A — Human Ergonomics Flow
  1. 01Upload or scan a student report card (Image or PDF document)
  2. 02Automated AI OCR extracts subjects, percentage marks, and teacher remarks
  3. 03Firestore securely persists multi-term academic history under the student profile
  4. 04Explore the Analytics Hub for subject timelines, CAPS levels, and volatility metrics
  5. 05Review AI Sentiment analysis and growth reflection time capsules
Surface B — Agentic Machine Execution Rails
Protocol: Model Context Protocol (MCP 1.0)Auth: Bearer banya_ag_...

1. Discovery: Scraper/Agent fetches /llms.txt or /.well-known/mcp.json manifest.

2. Tool Invocation: Executes deterministic Zod-validated tool contracts matching academiatrack operations.

3. Replay Protection: Enforces Idempotency-Key: <uuid> to prevent duplicate mutations during LLM retry loops.

4. Token Compression: Returns structured, PII-minimized Markdown/JSON stripping all layout noise.