TOML Conventions

Understanding the _A, _B, _C phase naming system for sprint management.

Phase Naming System

ÆtherLight uses a suffix system to indicate task status at a glance:

Active

Tasks currently being worked on. The "Start Next Task" button pulls from _A phases first.

Backlog

Planned tasks not yet started. Move to _A when ready to begin.

Completed

Finished tasks. Keep for reference and audit trail.

File Structure

# ACTIVE_SPRINT.toml

[sprint]
name = "User Authentication"
created = "2025-01-15T10:30:00Z"
updated = "2025-01-16T14:22:00Z"
status = "active"

# Active phases (being worked on now)
[phases.1_A]
name = "JWT Implementation"
description = "Add JSON Web Token authentication"
tasks = [
  { id = "jwt-001", title = "Create JWT utility", status = "in_progress", assignee = "you" },
  { id = "jwt-002", title = "Add token validation", status = "pending" }
]

[phases.2_A]
name = "Login Endpoints"
description = "REST API for authentication"
tasks = [
  { id = "login-001", title = "POST /auth/login", status = "pending" },
  { id = "login-002", title = "POST /auth/logout", status = "pending" }
]

# Backlog phases (planned for later)
[phases.3_B]
name = "OAuth Integration"
description = "Social login support"
tasks = [
  { id = "oauth-001", title = "Google OAuth", status = "pending" },
  { id = "oauth-002", title = "GitHub OAuth", status = "pending" }
]

# Completed phases
[phases.0_C]
name = "Project Setup"
description = "Initial configuration"
completed_at = "2025-01-14T18:00:00Z"
tasks = [
  { id = "setup-001", title = "Initialize Next.js", status = "completed" },
  { id = "setup-002", title = "Add TypeScript", status = "completed" }
]

Task Properties

Property	Required	Description
`id`	Yes	Unique identifier for the task
`title`	Yes	Short description of the task
`status`	Yes	pending, in_progress, or completed
`assignee`	No	Who is working on this task
`depends_on`	No	Array of task IDs this depends on
`files`	No	Related file paths for context

Multiple Sprint Files

You can have multiple sprint files for different work streams:

project/
├── ACTIVE_SPRINT.toml      # Main sprint
├── BACKLOG.toml            # Future features
├── BUGFIX_SPRINT.toml      # Bug fixes
└── TECH_DEBT.toml          # Refactoring tasks

Use the sprint dropdown in the extension to switch between them.

Best Practices

1.Keep _A phases focused - Only 1-2 active phases at a time
2.Move tasks, don't delete - Move to _C when done for history
3.Add file paths - Helps AI understand context
4.Use dependencies - Prevents starting tasks before prerequisites
5.Commit TOML files - They're Git-friendly and serve as documentation