v4.0.0 — Phase IV: Local Kubernetes Deployment

What's New in v4.0.0

This release marks the completion of Phase IV — local Kubernetes deployment of the DoIt productivity platform using Minikube, Helm Charts, kubectl-ai, and Kagent.

Kubernetes & Infrastructure

• Production-ready Dockerfiles for all three services (frontend, backend, MCP) using consistent multi-stage builds
• Umbrella Helm chart (doit/) with three subcharts (frontend/, backend/, mcp/) — deploy everything with a single helm install
• Ingress with path-based routing (/ → frontend, /api → backend, /mcp → MCP service)
• ConfigMaps & Secrets support via a gitignored values.secret.yaml for sensitive values (DB URLs, API keys, auth secrets)
• Health check endpoints on all services for Kubernetes liveness and readiness probes
• ClusterIP Services for backend and MCP; Ingress handles all external traffic
• Setup documentation for Minikube (recommended: 4 CPUs / 8GB RAM)
• Support for kubectl-ai and kagent for AI-assisted cluster operations

Application Features

• Socket.IO real-time updates
• Team invitations workflow
• Dashboard enhancements
• Archiving for tasks and projects
• Notifications system
• Dark mode improvements across Hero cards and UI

Compatibility

• Same chart works for local Minikube and can be adapted for cloud deployment in Phase V
• Full feature parity with the existing Docker Compose setup

Doit v3.0.0 — Conversational MCP Chatbot

Doit v3.0.0 — Conversational MCP Chatbot

Phase 3 of the hackathon: a conversational AI interface powered by the Model Context Protocol (MCP), enabling users to manage their tasks, notes, projects, and categories through natural language chat.

Architecture

• MCP Service: Python 3.13+ with FastMCP, Google Gemini agent, and FastAPI (SSE streaming)
• Frontend: Next.js chat UI with real-time streaming, markdown rendering, and conversation management
• Backend: New conversation & message models with Alembic migrations on PostgreSQL (Neon)
• Deployment: Docker multi-service compose (backend, mcp, frontend) + GitHub Actions CI/CD to Hugging Face Spaces

Features

Conversational AI Chat

• Full chat page (/chat) and floating ChatWidget accessible from anywhere in the app
• Server-Sent Events (SSE) streaming for real-time AI responses
• Markdown rendering in chat messages with syntax highlighting
• Conversation history with create, select, and delete support
• Optimistic UI updates for instant feedback
• Custom audio player component for media interactions

MCP Tool Suite

• Tasks tools: Create, list, update, and delete tasks across projects via natural language
• Notes tools: Create, list, update, and delete notes with category support
• Projects tools: Create, list, update, and delete projects with task summaries
• Categories tools: Create, list, update, and delete categories
• Dashboard tool: Get a full summary of user stats (projects, tasks, notes, categories)

AI Agent

• Google Gemini-powered conversational agent with tool calling
• Authenticated per-user context — the agent operates on the signed-in user's data
• Resilient error handling and graceful fallbacks

Backend Enhancements

• Conversation and Message database models with full Alembic migration
• Media delete endpoint for notes
• API proxy routes from Next.js frontend to MCP service

DevOps & Infrastructure

• Docker Compose with backend, mcp, and frontend services, named volumes, and env files
• GitHub Actions workflows for deploying backend and MCP service to Hugging Face Spaces
• Standardized mcp_service package with uv scripts for dev/prod

Technical Highlights

• FastMCP mounted as ASGI sub-app within FastAPI for unified routing
• SSE-based chat streaming with proper lifecycle management
• Next.js API proxy routes for MCP service communication
• Suspense-wrapped sidebar with query param-based chat routing
• 82 files changed, ~6,900 lines added across the MCP service, frontend chat UI, and infrastructure

PRs Included

• #16 — Add MCP service, conversation models & migration
• #17 — Add MCP chat UI, API proxies, and tools
• #18 — Standardize mcp_service and add uv scripts
• #19 — Add docker services, UI polish, and MCP updates
• #20 — Improve Gemini agent resilience and chat UI routing

Previous Releases

• v2.0.0 — Full-stack web app (Next.js + FastAPI)
• v1.0.0 — CLI task manager (doit-cli/)

🤖 Generated with Claude Code

v2.0.0

Doit v2.0.0 — Full-Stack Web App Release

The complete full-stack web application for Doit, transforming the CLI task manager into a modern web platform.

Architecture

• Frontend: Next.js 16 with TypeScript, Tailwind CSS, and shadcn/ui
• Backend: FastAPI with SQLAlchemy, Alembic migrations, and PostgreSQL (Neon)
• Auth: Better Auth with Google and GitHub social sign-in
• Deployment: Containerized with Docker

Features

Authentication & Users

• Email/password authentication
• Google and GitHub social sign-in (OAuth)
• Session management and protected routes

Dashboard

• Overview stats with project and note counts
• Category management with color coding
• Quick navigation to projects and notes

Notes

• Rich note editor with markdown formatting toolbar
• Multimedia support with image previews
• Category-based filtering and organization
• Note titles, editing, and confirmation dialogs

Projects & Task Management

• Project creation with Kanban/Grid framework selection
• Kanban Board: Drag-and-drop task management across TODO, In Progress, and Done columns
• Grid View: Alternative grid layout for task overview
• View toggle with backend persistence
• Task dialog for creating/editing tasks with priority levels (Low, Medium, High) and due dates
• Collaboration models for future team features

UI/UX

• Dark mode support
• Responsive sidebar with project navigation
• Confirmation dialogs for destructive actions
• Optimistic UI updates across the app

Technical Highlights

• Next.js API proxy routes for backend communication
• ProjectsContext for global state management
• Alembic database migrations
• Vitest frontend test setup
• Docker containerization for backend

Previous Release

• v1.0.0 — CLI task manager (now in doit-cli/)

🤖 Generated with Claude Code

DoIt v1.0.0 - cli

Release Date: January 8, 2026
Hackathon: Phase I Complete - In-Memory Python Console App
Points: 100/100

Overview
This release marks the successful completion of Phase I of Hackathon II: Spec-Driven Development. DoIt is a modern, beautiful command-line task manager built entirely using Spec-Driven Development with Claude Code and Spec-Kit Plus.

Phase I Achievement ✅
All Phase I requirements fulfilled:

✅ All 5 Basic Level Features: Add, Delete, Update, View, Mark Complete
✅ Spec-Driven Development: Complete specification history in /specs folder
✅ Clean Architecture: Proper separation of concerns (Models, Services, Storage)
✅ Test Coverage: 30/30 unit tests passing
✅ Zero Manual Coding: All code generated from refined specifications
Key Features
Task Management
Add Tasks: Create with title and description (/add "Title" -d "Description")
Multi-Task Mode: Batch creation with /add --multi
View Tasks: Rich formatted tables with filters (all/pending/completed)
Update Tasks: Modify title or description
Delete Tasks: Remove with confirmation prompts
Mark Complete: Interactive mode with arrow-key navigation
User Experience
🎨 Beautiful Terminal UI with gradient logo and Rich formatting
⚡ Smart Autocomplete with command completion and suggestions
🎯 Interactive Mode with arrow-key navigation
💬 Slash Commands for intuitive operations
📊 Real-time Stats in bottom toolbar
Technical Excellence
Architecture: Clean separation (Models/Services/Storage)
Type Safety: Full type hints with mypy validation
Code Quality: Ruff linting with zero warnings
Documentation: Comprehensive specs and docs
Testing: 100% critical path coverage
Installation

Clone the repository
git clone https://github.com/anasahmed07/doit.git
cd doit/doit-cli

Install with uv (recommended)
uv tool install .

Run the application
doit

Quick Start Commands
Command Description
/add "Task" -d "Description" Add a new task
/list [all|pending|completed] View tasks
/complete [id] Toggle completion
/update Update task
/delete Delete task
/help Show all commands
Spec-Driven Development
This release demonstrates the power of AI-assisted development:

Constitution Phase: Defined project principles
Specification Phase: Comprehensive feature specs
Planning Phase: Generated implementation plan
Task Breakdown: Actionable task list
Implementation: Code generated by Claude Code
Validation: 30 passing tests, all criteria met
Zero manual coding - All code generated from specifications!

Platform Support
✅ Windows (Windows Terminal, PowerShell, CMD)
✅ macOS (Terminal.app, iTerm2, Alacritty)
✅ Linux (Any terminal with 256-color support)
What's Next: Phase II
Phase II will transform DoIt into a full-stack web application:

Frontend: Next.js 16+ with App Router
Backend: FastAPI with RESTful API
Database: Neon Serverless PostgreSQL
Authentication: Better Auth
Credits
Development: Spec-Driven Development using Claude Code and Spec-Kit Plus
AI Assistant: Claude Opus 4.5 by Anthropic
Developer: Anas Ahmed
Hackathon: Panaversity Hackathon II

What's Changed

• Update settings, move docs, and add hackathon PDF by @anasahmed07 in #6
• folder restructure for the phase 2 by @anasahmed07 in #7

Full Changelog: v0.1.1...v1.0.0

DoIt v0.1.0 - Initial Release

DoIt v0.1.0 - Initial Release 🎉

First stable release of DoIt - a modern, beautiful command-line task manager built with Python 3.13+.

✨ Features

• 🎨 Beautiful Terminal UI - Rich colors, panels, and formatted output
• ⚡ Smart Autocomplete - Tab completion with command templates and placeholders
• 💬 Slash Commands - Intuitive /command syntax for all operations
• 🚀 Lightning Fast - In-memory storage for instant responses
• 🎯 Interactive Mode - Arrow key navigation and visual task selection
• 🌈 Gradient Logo - Eye-catching ASCII art with color gradients
• 📊 Real-time Stats - Live task counts in the bottom toolbar
• ✅ Full CRUD - Create, read, update, delete, and toggle tasks

🚀 Installation

# Clone the repository
git clone https://github.com/anasahmed07/doit.git
cd doit

# Install with uv (recommended)
uv sync

# Or with pip
pip install -e .

📖 Usage

Start the application:

doit

Type / to see all available commands with autocomplete.

📋 Available Commands

• /add "<title>" -d "<desc>" - Add a new task
• /add --multi - Add multiple tasks
• /list [filter] - List tasks (all/pending/completed)
• /complete [id] - Toggle task completion
• /update <id> - Update a task
• /delete <id> - Delete a task
• /clear - Clear screen
• /help - Show help
• /quit - Exit application

🛠️ Requirements

• Python 3.13 or higher
• Dependencies: rich>=13.7.0, prompt-toolkit>=3.0.43

🌐 Platform Support

• ✅ Windows (Windows Terminal, PowerShell, CMD)
• ✅ macOS (Terminal.app, iTerm2, Alacritty)
• ✅ Linux (Any terminal with 256-color support)

📝 What's Included

• Complete task management functionality
• 30/30 tests passing with full coverage
• Type-safe Python implementation
• Clean, documented codebase
• MIT License

---

Made with ❤️ and Python 3.13+