writing-plans

Decompose implementation tasks into atomic, execution-ready steps with verification.

31|8|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/itallstartedwithaidea/agent-skills --skill writing-plans-itallstartedwithaidea
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-plans
Source: https://github.com/itallstartedwithaidea/agent-skills/tree/main/skills/software-dev/writing-plans
Command: npx skills add https://github.com/itallstartedwithaidea/agent-skills --skill writing-plans-itallstartedwithaidea

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Writing complex implementations is slower and riskier when plans are vague, so work gets misaligned, errors compound, and handoffs fail.

Core Features & Use Cases

  • Atomic, time-bounded steps: Decomposes tasks into 2–5 minute actions that can be executed mechanically.
  • Exact artifacts: Specifies file paths, complete code blocks, and insertion points instead of generic guidance.
  • Built-in verification: Adds test/command or UI confirmation after each step to catch failures immediately.
  • Handoff-ready output: Produces a plan that another agent (or a future version of the same agent) can execute without the original conversation.

Example: turning a feature request into an implementation plan that updates schemas, adds migrations, and wires new API route handlers in the correct dependency order—each step verified before moving on.

Quick Start

Ask an AI to create a writing plan for your implementation task with exact file paths, full code changes, and a verification command after each step.

Frequently Asked Questions about writing-plans

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

FAQPage Schema
How do I create a step-by-step implementation plan for multi-file code changes?

Creating a step-by-step implementation plan for multi-file code changes involves decomposing tasks into atomic, 2–5 minute actions with exact file paths, complete code blocks, and per-step verification commands to ensure the codebase remains functional throughout the process.

What is task decomposition for agent handoff in software engineering?

Task decomposition for agent handoff is the process of breaking down implementation work into execution-ready, context-free steps that another agent can mechanically execute without the original conversation, complete with precise file targeting and rollback guidance.

How do I write verifiable steps for a complex feature or refactoring task?

Writing verifiable steps for a complex refactoring task requires specifying exact insertion points, full code changes, and test or UI confirmation commands after each 2–5 minute action to catch failures immediately before proceeding.

Can I use step-by-step planning for multi-file schema updates and API route wiring?

Yes, step-by-step planning supports multi-file schema updates, migrations, and API route wiring by specifying the correct dependency order and adding measurable progress checkpoints with rollback guidance after each atomic action.

What's the best way to structure a handoff-ready implementation plan for another AI agent?

The best way to structure a handoff-ready implementation plan is to define time-bounded atomic steps with exact artifacts, including full file paths, complete code blocks, and built-in verification commands, ensuring execution without prior conversation context.

When should I not use atomic step planning for code changes?

You should avoid atomic step planning for code changes that do not span multiple files or require measurable progress checkpoints, as the overhead of specifying exact file paths and per-step verification is designed for complex, multi-file refactors and agent handoffs.