Skip to content

task.md

Latest commit

 

History

History
78 lines (64 loc) · 4.43 KB

task.md

File metadata and controls

78 lines (64 loc) · 4.43 KB
Raw

GovAssist Comprehensive Task & Roadmap Document

This is the single source of truth for all project planning, tasks, and documentation phases for GovAssist.

Phase 1: MVP Core Development (Completed)

  • Scaffold repository, FastAPI backend, React (Vite) frontend.
  • Implement database models (Users, Chats, Docs, Schemes).
  • Implement RAG Pipeline & LLM Connector (Gemini API integration).
  • Integrate FAISS Vector Store and Document Ingestion.
  • Implement Eligibility Rule Engine for Schemes.
  • Implement Authentication & Session Store (JWT).
  • Create Frontend Chat UI and Navigation Layout.
  • Implement OCR & PDF Parsing with Tesseract & PyMuPDF.
  • Write baseline test suite and Dockerize (Docker-compose) stack.

Phase 2: Production-Readiness & Automation (Current Status)

Database & State Management

  • Migrate to PostgreSQL: Replaced local SQLite (sql_app.db) with Supabase PostgreSQL.
  • Implement Alembic Migrations: Replace Base.metadata.create_all() with Alembic migrations to safely manage schema changes.
  • Redis Caching: Implement Redis to cache frequently accessed scheme data.

Security & Secrets Hardening

  • Enforce Environment Secrets: Removed hardcoded "CHANGE_ME_IN_PROD". Server fails if secret key is missing.
  • Sanitize .env Loading: Ensure .env parser strictly strips trailing spaces from API keys.
  • Rate Limiting: Implemented rate limiting (slowapi) on authentication and chat endpoints (60 req/min).
  • Sentry Error Tracking: Integrated Sentry SDK for production exception tracking.

Storage, Automation & Dashboards

  • Supabase Cloud Storage: Automatically backup user-uploaded documents to govassist-documents.
  • Automated Data Scraper: Built automated scraping pipeline using APScheduler.
  • Owner Dashboard Toggles: Added live UI switches to toggle Scraper, Supabase, and LLM features.
  • Fix Frontend Navigation: Corrected "Sign In" and "Sign Up" grouping.

Phase 3: Final Year Evaluation Polish (Pending)

  • Real-time Analytics Dashboard: Hook up metrics in OwnerDashboard.jsx (Active Users, Uptime, etc.) to live database queries.
  • Build Real Web Scraper: Upgrade the mock background scraper to actually extract HTML from Wikipedia's live scheme data.
  • Export to PDF Feature: Allow users to download their personalized scheme "Eligibility Report" as a PDF.
  • Advanced Automated Testing: Write exhaustive Python pytest and JS jest tests to demonstrate robust engineering.

Phase 4: Full Documentation Suite (Pending)

Generate the complete suite of final-year project documentation inside docs/complete-suite:

  • 1_Project_Overview.md
  • 2_Features.md
  • 3_Architecture.md
  • 4_UML_Diagrams.md
  • 5_API_Reference.md
  • 6_Database_Schema.md
  • 7_Developer_Guide.md
  • 8_Admin_Manual.md
  • 9_User_Manual.md
  • 10_Testing_QA.md
  • 11_Performance_Report.md
  • 12_Deployment_Guide.md
  • 13_Security_Compliance.md
  • 14_Changelog.md
  • 15_Executive_Summary.md

Phase 5: Enterprise Production & Security (The "Real World")

These are the enterprise-grade requirements necessary to launch this platform to real citizens on the internet.

  • Real Authentication (Google/OTP): Replace fake JWT login with Firebase/Supabase Auth for 1-click Google OAuth and Mobile OTP verification.
  • Email Integration: Integrate SendGrid or AWS SES for real password reset links and welcome emails.
  • PII Encryption: Implement AES-256 encryption at rest for sensitive user profile data (Income, Caste, Gender) to comply with the DPDP Act.
  • Cloud Vector Database: Migrate from local FAISS to a scalable cloud vector DB like Pinecone, Milvus, or pgvector to prevent RAM exhaustion.
  • Asynchronous Task Queues: Implement Celery + RabbitMQ/Redis to offload heavy PDF document parsing and LLM generation from the main server thread.
  • Official API Integration: Replace the Wikipedia scraper with official API Setu / UMANG API integrations for legally compliant, real-time scheme data.
  • Strict Security Policies: Enforce strict CORS policies, CSRF tokens, and helmet headers.
  • Real-time Telemetry & APM: Replace the remaining mock charts in the Owner Dashboard (Hourly Traffic, Regional Data, Uptime, Latency) with real Application Performance Monitoring data (e.g., Prometheus/Grafana or Datadog).