How do I install the Planning And Task Breakdown skill?

Run npx skillmds@latest add itsual/planning-and-task-breakdown in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Planning And Task Breakdown safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Planning And Task Breakdown?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Planning And Task Breakdown free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Planning And Task Breakdown?

itsual (@itsual) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Planning And Task Breakdown by itsual

Name: Planning And Task Breakdown
Author: itsual

Planning And Task Breakdown

Break complex work into small, verifiable tasks with clear acceptance criteria. Use whenever starting a non-trivial feature, refactor, or project, or when the user asks for a plan, task list, breakdown, or roadmap of implementation steps. Prefer this skill before writing significant code.

itsual Updated Sep 21, 2026

File contents

Planning and Task Breakdown

Overview

Turn ambiguous or large goals into a sequence of small, independently verifiable tasks. Good planning dramatically increases the success rate of agent-driven work.

When to Use

Starting any non-trivial feature or change
User asks for a plan, breakdown, tasks, or implementation steps
Before beginning coding on a new area
When a previous attempt failed due to scope or unclear requirements

Core Process

1. Clarify the Goal

Restate the objective in one or two sentences. Surface assumptions explicitly:

ASSUMPTIONS:
1. ...
2. ...
→ Correct me if any of these are wrong.

2. Identify Scope Boundaries

What is explicitly in scope? What is out of scope for this iteration?

3. Decompose into Vertical Slices

Prefer thin vertical slices (end-to-end value) over horizontal layers when possible.

Each task should be:

Small enough to complete and verify in one focused session
Independently testable
Ordered by dependency

4. Write Clear Acceptance Criteria

For every task, define how we will know it is done (tests, behavior, observable outcome).

5. Sequence and Estimate Roughly

Order tasks. Note any that can be parallelized.

Output Format

Produce a structured plan:

## Goal
[One sentence]

## Assumptions
- ...

## Out of Scope
- ...

## Tasks

### Task 1: [Short title]
**Description**: ...
**Acceptance criteria**:
- [ ] ...
- [ ] ...
**Dependencies**: none / Task X

### Task 2: ...

Quality Checks

Are tasks small enough?
Does each have verifiable acceptance criteria?
Is the first task a good starting point (low risk, high learning)?
Have major risks or unknowns been called out?

Common Pitfalls

Tasks that are still too large ("Implement the whole auth system")
Missing acceptance criteria
Hidden assumptions not surfaced
Starting implementation before the plan is agreed

Verification

A good plan lets another agent (or human) pick up any task and know exactly what "done" looks like without further clarification.

---name: planning-and-task-breakdowndescription: Break complex work into small, verifiable tasks with clear acceptance criteria. Use whenever starting a non-trivial feature, refactor, or project, or when the user asks for a plan, task list, breakdown, or roadmap of implementation steps. Prefer this skill before writing significant code.---# Planning and Task Breakdown## OverviewTurn ambiguous or large goals into a sequence of small, independently verifiable tasks. Good planning dramatically increases the success rate of agent-driven work.## When to Use- Starting any non-trivial feature or change- User asks for a plan, breakdown, tasks, or implementation steps- Before beginning coding on a new area- When a previous attempt failed due to scope or unclear requirements## Core Process### 1. Clarify the GoalRestate the objective in one or two sentences. Surface assumptions explicitly:```ASSUMPTIONS:1. ...2. ...→ Correct me if any of these are wrong.```### 2. Identify Scope BoundariesWhat is explicitly in scope? What is out of scope for this iteration?### 3. Decompose into Vertical SlicesPrefer thin vertical slices (end-to-end value) over horizontal layers when possible.Each task should be:- Small enough to complete and verify in one focused session- Independently testable- Ordered by dependency### 4. Write Clear Acceptance CriteriaFor every task, define how we will know it is done (tests, behavior, observable outcome).### 5. Sequence and Estimate RoughlyOrder tasks. Note any that can be parallelized.## Output FormatProduce a structured plan:```markdown## Goal[One sentence]## Assumptions- ...## Out of Scope- ...## Tasks### Task 1: [Short title]**Description**: ...**Acceptance criteria**:- [ ] ...- [ ] ...**Dependencies**: none / Task X### Task 2: ...```## Quality Checks- Are tasks small enough?- Does each have verifiable acceptance criteria?- Is the first task a good starting point (low risk, high learning)?- Have major risks or unknowns been called out?## Common Pitfalls- Tasks that are still too large ("Implement the whole auth system")- Missing acceptance criteria- Hidden assumptions not surfaced- Starting implementation before the plan is agreed## VerificationA good plan lets another agent (or human) pick up any task and know exactly what "done" looks like without further clarification.

Frequently asked questions