Task Decomposition

Break epics into parallelizable tasks with acceptance criteria, effort estimates, and dependency graphs.

509 stars

Best use case

Task Decomposition is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Break epics into parallelizable tasks with acceptance criteria, effort estimates, and dependency graphs.

Teams using Task Decomposition should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/task-decomposition/SKILL.md --create-dirs "https://raw.githubusercontent.com/a5c-ai/babysitter/main/library/methodologies/ccpm/skills/task-decomposition/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/task-decomposition/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How Task Decomposition Compares

Feature / Agent	Task Decomposition	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

Break epics into parallelizable tasks with acceptance criteria, effort estimates, and dependency graphs.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Task Decomposition

Break epics into parallelizable tasks with acceptance criteria, effort estimates, and dependency graphs.

## Agent
Task Analyst - `task-analyst`

## Workflow
1. Analyze epic and identify 4-5 parallel work streams
2. Decompose each stream into concrete tasks (parallel)
3. Set acceptance criteria in Given/When/Then format
4. Estimate effort (S/M/L/XL) and set parallelization flags
5. Build cross-stream dependency graph
6. Validate coverage against PRD user stories
7. Fill gaps with additional tasks if needed
8. Write task files to .claude/epics/<featureName>/<N>.md

## Inputs
- `projectName` - Project name
- `featureName` - Feature identifier
- `epic` - Epic from Phase 2
- `prd` - PRD from Phase 1
- `parallelAgents` - Max parallel agents (default: 5)

## Outputs
- Task list with acceptance criteria and effort estimates
- Work stream definitions
- Dependency graph with execution waves
- Coverage report with PRD traceability

## Process Files
- `ccpm-task-decomposition.js` - Standalone decomposition
- `ccpm-orchestrator.js` - Phase 3 of full lifecycle

Related Skills

work-unit-decomposition

509

from a5c-ai/babysitter

Decompose implementation plans into discrete work units with enumerated DoD items, file scope declarations, dependency mapping, and human checkpoint flags.

story-decomposition

509

from a5c-ai/babysitter

Break technical specifications into small, implementable stories with dependency ordering

work-decomposition

509

from a5c-ai/babysitter

Decompose goals into MEOWs (Molecular Expressions of Work) - trackable atomic units following Gas Town's bead-based work model.

process-builder

509

from a5c-ai/babysitter

Scaffold new babysitter process definitions following SDK patterns, proper structure, and best practices. Guides the 3-phase workflow from research to implementation.

Workflow & Productivity

babysitter

509

from a5c-ai/babysitter

Orchestrate via @babysitter. Use this skill when asked to babysit a run, orchestrate a process or whenever it is called explicitly. (babysit, babysitter, orchestrate, orchestrate a run, workflow, etc.)