create-plan

Decompose design docs into execution-ready subtasks with commands and outputs.

Updated Nov 4, 2022
One-click install
npx skills add https://github.com/CityBear3/dotfiles --skill create-plan-citybear3
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-plan
Source: https://github.com/CityBear3/dotfiles/tree/main/claude/skills/create-plan
Command: npx skills add https://github.com/CityBear3/dotfiles --skill create-plan-citybear3

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents ambiguous engineering work by decomposing a design doc or clear scope into small, independently verifiable tasks that an agent team can execute without extra context.

Core Features & Use Cases

  • Self-contained implementation plans: Produces a plan where each task includes exact file paths, concrete code where needed, exact commands, and expected outputs.
  • Bite-sized, verifiable work units: Breaks work into tasks designed to be independently testable and commit-ready.
  • TDD-aligned behavior changes: Forces behavior-changing tasks to follow red→green→refactor discipline with pinned verification per task.
  • Safe planning storage: Saves plans to docs/plans/YYYY-MM-DD-<feature-name>.md unless overridden by existing conventions.

Quick Start

Use the create-plan skill to generate a self-contained implementation plan for your approved Design Doc by invoking /create-plan during the planning phase.

Frequently Asked Questions about create-plan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I decompose a design doc into execution-ready subtasks for autonomous agents?

Implementation planning decomposes a design doc into small, independently verifiable tasks with exact file paths, concrete commands, and expected outputs. This creates self-contained plans requiring no extra context for agent teams to execute.

What is TDD-aligned task decomposition for engineering workflows?

TDD-aligned task decomposition forces behavior-changing tasks to follow red, green, refactor discipline with pinned verification per task. This ensures each work unit is independently testable and commit-ready before moving forward.

How do I create self-contained implementation plans with exact file paths and commands?

Self-contained implementation plans are generated by specifying explicit affected files, concrete step instructions including exact commands, and expected outputs. Plans are saved to a configured directory like docs/plans for safe storage.

Does implementation planning work for handing off engineering work to agent teams?

Implementation planning applies directly to autonomous engineering work handed off to agent teams. It prevents ambiguous work by producing bite-sized, verifiable units designed to be independently testable and commit-ready.

Where are generated engineering documentation plans saved by default?

Engineering documentation plans are saved to docs/plans/YYYY-MM-DD-feature-name.md unless overridden by existing conventions. This ensures safe planning storage and easy retrieval for subsequent implementation phases.

When should I not use autonomous task decomposition for software engineering?

Task decomposition for autonomous agents is not suitable for vaguely scoped tasks lacking a design doc. It requires clearly defined scope to produce verifiable, execution-ready subtasks with explicit affected files and concrete commands.