How does the external plan review process work?

The skill can trigger a hook to send your plan to an external AI reviewer (like Codex or Gemini) for feedback, ensuring the logic is sound and 'LGTM' before proceeding.

What does 'bite-sized task granularity' mean?

Every task in the generated plan is designed to be completed in 2-5 minutes, covering specific actions like writing a single test, running a command, or performing a commit.

Can I choose how the plan is executed?

Yes, after a plan is finalized, the skill offers two options: Subagent-Driven execution for fast iteration within the session, or Parallel Session execution for batch processing.

What is the primary benefit of the Writing Plans skill?

It eliminates implementation ambiguity by forcing a comprehensive, reviewed roadmap that follows TDD principles before a single line of production code is written.

TDD Implementation Planner

Name: TDD Implementation Planner
Author: tajohnson

bytajohnson

0•

Productivity & Workflow

Generates granular, test-driven implementation roadmaps that break complex technical requirements into bite-sized, executable tasks.

The Writing Plans skill provides a structured framework for Claude to architect multi-step technical tasks before implementation begins. It enforces software engineering best practices like Test-Driven Development (TDD), YAGNI, and DRY by producing comprehensive documentation that includes exact file paths, code snippets, and git commit instructions. By integrating an external review process and offering flexible execution strategies—such as subagent-driven development or parallel sessions—it ensures that both human developers and AI agents have a clear, reviewed, and version-controlled roadmap for building robust features.

Key Features

01Automated implementation plan storage in version-controlled directories

02Integrated external AI review hook for plan verification

03Enforced TDD workflow with explicit failing/passing test steps

04Standardized headers covering architecture, stack, and goals

05Granular task breakdown into 2-5 minute actionable segments

060 GitHub stars

Use Cases

01Onboarding AI subagents to specific domain problems with zero context

02Architecting complex new features with clear separation of concerns

03Refactoring legacy codebases using strict test-driven methodologies

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tajohnson/superpowers writing-plans

For use in Claude.ai and ChatGPT

Key Features

01Automated implementation plan storage in version-controlled directories

02Integrated external AI review hook for plan verification

03Enforced TDD workflow with explicit failing/passing test steps

04Standardized headers covering architecture, stack, and goals

05Granular task breakdown into 2-5 minute actionable segments

060 GitHub stars

Use Cases

01Onboarding AI subagents to specific domain problems with zero context

02Architecting complex new features with clear separation of concerns

03Refactoring legacy codebases using strict test-driven methodologies

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tajohnson/superpowers writing-plans

For use in Claude.ai and ChatGPT