---

name: task-breakdown description: Activate when breaking down implementation plans into detailed task lists with agent contexts, acceptance criteria, and status tracking license: MIT

Task Breakdown

Guide for creating detailed task lists and specialized agent contexts from implementation plans, enabling efficient parallel execution with clear acceptance criteria.

When to Use This Skill

Activate when:

• Breaking down implementation plans into actionable tasks
• Creating specialized agent contexts for task execution
• Defining acceptance criteria and completion requirements
• Establishing task dependencies and priorities
• Setting up status tracking for project progress
• Managing context window efficiency with sub-tasks

Task Breakdown Process

Phase 1: Plan Analysis

1. Parse the implementation plan structure
2. Extract phase information and objectives
3. Identify all tasks within each phase
4. Analyze task complexity and relationships

Phase 2: Task List Generation

1. Create detailed task descriptions
2. Define acceptance criteria for each task
3. Establish priority levels and dependencies
4. Group related tasks for efficient execution

Phase 3: Context Strategy

• Simple Phases (≤2 tasks): Single phase context
• Complex Phases (>2 tasks): Phase context + individual task contexts
• Cross-Phase Tasks: Shared context with references
• Specialized Tasks: Domain-specific contexts

Task Structure

## Task: [Task Name]
**ID**: [Phase].[Task Number]
**Priority**: High/Medium/Low
**Complexity**: Simple/Medium/Complex

### Description
[Detailed description of what needs to be implemented]

### Acceptance Criteria
- [ ] [Specific, testable requirement 1]
- [ ] [Specific, testable requirement 2]
- [ ] [Specific, testable requirement 3]

### Dependencies
- **Prerequisite Tasks**: [Tasks that must be completed first]
- **External Dependencies**: [Third-party services, approvals]

### Deliverables
- [Specific code files, configurations]
- [Test cases and validation results]

Task Complexity Levels

• Simple: Configuration changes, minor fixes (1-4 hours)
• Medium: New components, API endpoints (1-3 days)
• Complex: Major features, integrations (1-2 weeks)
• Epic: Complete subsystems (2+ weeks) - should be broken into sub-tasks

Context File Organization

project/
├── phase1/
│   ├── claude.md (phase coordination)
│   ├── phase_status.md (phase progress)
│   ├── tasks.md (task list)
│   ├── task1/
│   │   ├── claude.md (task context)
│   │   └── status.md (task progress)
│   └── task2/
│       ├── claude.md (task context)
│       └── status.md (task progress)
├── phase2/
│   └── [similar structure]
└── shared/
    └── utilities/
        └── claude.md (shared context)

Sub-Task Creation Triggers

Create sub-tasks when:

• Context window approaching 80% capacity
• Processing large documentation (>50KB)
• Complex phases with >5 interdependent tasks
• Tasks requiring multiple specialized skills
• Analysis of extensive codebases

Mandatory File Requirements

Every task directory MUST contain:

1. claude.md: Task-specific agent context
2. status.md: Task progress tracking

Every phase directory MUST contain:

1. claude.md: Phase coordination context
2. phase_status.md: Phase progress tracking
3. tasks.md: Task list for the phase

Key Principles

• Actionability: Each task must have clear, executable steps
• Measurability: Every task has testable acceptance criteria
• Dependency Clarity: All prerequisites clearly identified
• Context Efficiency: Match context size to task complexity
• Status Integration: Progress tracked in status files
• No Empty Directories: Always populate with required files