writing-plans

Creates implementation plans with bite-sized tasks, exact file paths, and verification steps.

1|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/voidful/Aixlarity --skill writing-plans-voidful
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-plans
Source: https://github.com/voidful/Aixlarity/tree/main/.aixlarity/skills/writing-plans
Command: npx skills add https://github.com/voidful/Aixlarity --skill writing-plans-voidful

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Multi-step features often fail because implementation starts without a clear plan, leaving developers to guess at scope, file locations, and how to verify each change. This Skill turns a spec or requirements document into a structured implementation plan where every step is small, explicit, and verifiable. ## Core Features & Use Cases - Structured Plan Format: Every plan includes a goal, current state, bite-sized steps, exact file paths, verification steps, and risks. - Task Granularity Rules: Enforces single-action steps of 2-5 minutes each, so vague items like "implement the feature" get broken down before work begins. - Command Integration: Plans can be saved as markdown or wired into Aixlarity's command system via TOML command definitions. - Use Case: Before delegating a feature to subagents via spawn_agent, generate a plan that lists each file to change and how to confirm each step worked, so reviewers and agents share the same expectations. ## Quick Start Ask the agent to read the spec and create an implementation plan for the feature with bite-sized steps, exact file paths, and verification steps.

Frequently Asked Questions about writing-plans

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

FAQPage Schema
How do I write an implementation plan for a multi-step feature?

Start with a one-sentence goal, describe the current state and relevant files, then list single-action steps of 2-5 minutes each. Include exact file paths relative to the repo root, a verification method per step, and a risks section covering assumptions.

What makes a good task breakdown for coding plans?

Each step must be a single action, such as "write the failing test". Combined items like "add the struct and write tests" are two steps, not one. If someone has to guess what to do, the plan is incomplete.

When should I create an implementation plan before coding?

Create a plan before implementing any multi-step feature and before delegating work to subagents via spawn_agent. Even simple-looking tasks benefit, because unstated assumptions are a common source of bugs.

Can I save implementation plans as reusable commands?

Yes. Plans can be saved as markdown, or defined as TOML command files under .aixlarity/commands/ with a description and a prompt that references architecture docs and accepts arguments.

What are the limitations of plan-first development?

A plan only captures what was known at writing time, so undocumented assumptions can still surface during implementation. The risks section mitigates this by forcing explicit statements of what could go wrong.