AFT

Atlassian Free Trello.

What?

Kanban style task organisation.

Why?

Trello was great, then Atlassian bought it.

How?

• Clone the repo to a machine running docker
• Edit .env to have more secure passwords
• docker compose up -d
• Navigate to http(s)://{docker-host-ip}

Backup Storage

Automatic backups are stored on the host filesystem at ./backups/ (relative to the docker-compose.yml location). This directory is automatically created by Docker and persists across container restarts. You can include this directory in your host backup solution for additional data protection.

Permission Requirements: The container runs as a non-root user (UID 1000). If the ./backups/ directory is not writable, backups will fail with a permission error displayed in the UI. To fix this, run:

sudo chown -R 1000:1000 ./backups
sudo chmod -R 755 ./backups

When?

In one evening for version 1. That's right this is entirely copilot generated with my general guidance. I know what it all does, I have no idea what code was written to achieve it. Use at your own risk.

Features

📋 Board Management

• Create Multiple Boards - Organize different projects with separate Kanban boards
• Update Board Details - Rename boards and modify their properties
• Delete Boards - Remove boards when projects are complete
• Default Board Setting - Set a default board to load on startup
• Board Statistics - View counts of boards, columns, and cards

📊 Column Management

• Flexible Columns - Create custom columns for your workflow (e.g., To Do, In Progress, Done)
• Reorder Columns - Rearrange columns to match your process
• Column Operations - Add, edit, or delete columns as your workflow evolves
• Column Menu - Access column actions via three-dots menu:
  • Move All Cards - Batch move all cards from one column to another (top or bottom position)
  • Archive All Cards - Archive all active cards in a column at once
  • Unarchive All Cards - Unarchive all archived cards in a column (visible in archive view)
  • Delete All Cards - Remove all cards from a column
  • Delete Column - Remove the entire column

🎴 Card Management

• Create Cards - Add task cards with titles and descriptions
• Move Cards - Drag cards between columns to track progress
• Update Cards - Edit card details, titles, and descriptions
• Delete Cards - Remove completed or cancelled tasks
• Archive Cards - Archive completed cards to declutter your board while preserving history
• Unarchive Cards - Restore archived cards back to active view when needed
• View Switching - Switch between Task, Scheduled, and Archived views using the header dropdown
• Batch Operations - Archive or unarchive multiple cards at once via column menu

✅ Checklist Items

• Task Breakdown - Add checklist items to cards for subtasks
• Track Progress - Check off items as you complete them
• Update Checklists - Modify checklist item text and completion status
• Remove Items - Delete checklist items when no longer needed

💬 Comments

• Card Discussion - Add comments to cards for collaboration
• Comment History - View all comments on a card with timestamps
• Delete Comments - Remove outdated or incorrect comments

🔄 Scheduled Cards (Recurring Tasks)

• Template Cards - Convert any card into a recurring task template
• Flexible Scheduling - Create cards automatically on a schedule:
  • Frequency Options: Every N minutes, hours, days, weeks, months, or years
  • Start Date/Time: Configure when the schedule begins
  • End Date/Time: Optional schedule expiration (auto-disables when reached)
• Schedule Management:
  • Enable/Disable: Turn schedules on or off without deleting them
  • Edit Schedules: Update frequency, dates, or settings at any time
  • Delete Schedules: Remove recurring schedules when no longer needed
• Duplicate Control - Prevent duplicate cards in the same column or allow multiple instances
• Live Preview - See the next 4 scheduled run times before saving
• Scheduled View - Dedicated view for managing template cards (separate from task view)
• Automatic Creation - Cards created every minute based on enabled schedules
• Schedule Tracking - Auto-generated comments track which schedule created each card
• Checklist Preservation - Template checklist items are copied to new cards

⚙️ Settings & Configuration

• Customizable Settings - Configure application preferences including default board

Automatic Database Backups - Schedule recurring backups to protect your data

• Configurable frequency (minutes, hours, or days)
• Flexible start time alignment for backup scheduling
• Automatic retention management (keep 1-100 most recent backups)
• Backup health monitoring with status indicators
• Overdue backup detection
• Backups saved to host filesystem via bind mount (./backups/)

🔔 Notifications

• System Notifications - Receive alerts about important events
• Backup Notifications - Alerts for overdue backups or backup failures
• Schedule Notifications - Alerts for schedule errors or when schedules end
• Version Notifications - Automatic alerts when new AFT releases are available
• Action Buttons - Optional action buttons on notifications with secure URL validation
• Notification Center - View and manage all notifications from the header
• Mark as Read - Individual or bulk mark notifications as read and delete
• Notification Badge - Unread count indicator in header

🔧 System Info

• Version Tracking - Monitor application and database schema versions
• DB Stats - Statistics on task entities in the database
• Service Monitoring - Status of running system services and enable/disable toggle

🤖 Background Services

• Backup Scheduler - Automatic database backup service running every minute
  • Monitors backup schedule and creates backups at configured intervals
  • Enforces retention policies automatically
  • Checks disk space before creating backups
  • Creates notifications for failures or overdue backups
• Card Scheduler - Automatic card creation service running every minute
  • Processes enabled schedules and creates cards at scheduled times
  • Enforces duplicate prevention rules
  • Auto-disables schedules past their end date
  • Creates notifications on errors
  • Adds tracking comments to created cards
• Housekeeping Scheduler - Automatic maintenance service running every hour
  • Checks GitHub for new AFT releases
  • Creates notifications when updates are available
  • Prevents duplicate notifications for the same version
  • Can be enabled/disabled via System Info page

🔌 API Documentation

• Interactive API Docs - Built-in Swagger UI at /api/docs
• RESTful API - Full API access for integrations and automation
• Health Checks - Database connectivity and version endpoints

Architecture Decisions

This section documents key architectural choices made in the AFT application.

Custom Date Math vs. dateutil

Decision: Implement custom _add_months() and _add_years() functions instead of using python-dateutil.

Rationale:

• Scheduled cards feature only needs simple calendar math (add N months/years)
• python-dateutil adds 300KB+ dependency for functionality we don't need
• Our custom implementation handles edge cases correctly:
  • Month overflow (Jan 31 + 1 month = Feb 28/29)
  • Leap years (Feb 29 + 1 year = Feb 28)
  • Year boundaries (Dec 31 + 1 month = Jan 31 next year)
• No timezone/DST complexity needed (app uses local time)
• Custom functions are ~50 lines vs. large library

When to Reconsider: If we later need timezone handling, complex recurrence rules (RRULE), or relative deltas with multiple units, consider adding python-dateutil.

Implementation: See server/schedule_utils.py for the custom date math functions.

Background Services: Daemon Threads vs. Celery

Decision: Use simple Python scripts with daemon threads instead of Celery task queue.

Rationale:

• Application is single-user/small-scale
• Simple periodic tasks (backups every N minutes, card creation on schedule)
• Don't need distributed task execution, complex workflows, or advanced retry logic
• Simpler deployment (no Redis/RabbitMQ required)
• Easier to debug and maintain
• Threads run within Flask app container, sharing database connections and context

When to Reconsider: If application scales to multi-user with complex workflows, many background tasks, or need for task prioritization/distribution, consider Celery.

Implementation: See server/backup_scheduler.py and server/card_scheduler.py for examples of the daemon thread pattern.

Single Container vs. Microservices

Decision: Run all background services as daemon threads within the Flask app container.

Rationale:

• Simpler deployment and orchestration
• Shared database connections and SQLAlchemy sessions
• Direct access to Flask app models and utilities
• Lower resource overhead (no inter-service communication)
• Appropriate for single-user application scale

Trade-offs:

• All services restart if Flask app restarts
• Can't scale services independently
• Acceptable for current use case

Implementation: Background services initialized in server/app.py via init_backup_scheduler() and init_card_scheduler() functions.