# Task Decomposition

> Task Decomposition

- Skill: `a5c-ai/task-decomposition` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add a5c-ai/task-decomposition`
- Raw SKILL.md: https://api.skillmd.com/api/skills/a5c-ai/task-decomposition/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: a5c-ai (https://skillmd.com/u/a5c-ai)
- Updated: 2026-08-19
- Page: https://skillmd.com/skills/a5c-ai/task-decomposition

---


# 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