Can this skill be used for multi-agent workflows?

Yes, it is specifically designed to output plans that can be handed off to subagents for automated execution using the subagent-driven-development pattern.

How granular are the implementation tasks?

Tasks are designed to be 'bite-sized,' representing roughly 2 to 5 minutes of work per step to maximize accuracy and minimize context drift.

What is the primary purpose of the Writing Plans skill?

It creates a structured, step-by-step technical roadmap for a feature before any code is modified, ensuring architectural consistency and test coverage.

How does it support Test-Driven Development (TDD)?

The skill breaks every task into specific steps: writing a failing test, running it to verify failure, implementing minimal code, and verifying the pass.

Where are the generated plans saved?

Plans are automatically saved as Markdown files in the docs/superpowers/plans/ directory, organized by date and feature name.

Writing Plans

Name: Writing Plans
Author: SwervoJake

bySwervoJake

0•

开发者工具

Generates detailed, TDD-focused implementation plans that break down complex requirements into bite-sized, executable tasks for AI agents.

The Writing Plans skill streamlines the transition from brainstorming to execution by creating comprehensive technical roadmaps. It enforces rigorous engineering standards—including Test-Driven Development (TDD), DRY principles, and YAGNI—by decomposing high-level specifications into granular, 2-5 minute tasks. Each plan includes exact file paths, required code snippets, and specific verification commands, ensuring that even agents with zero codebase context can implement features with high reliability and architectural integrity.

主要功能

01Automated plan review loops to validate logic before code is written

020 GitHub stars

03TDD-driven task decomposition with explicit failing and passing test cycles

04Standardized implementation headers for seamless subagent handoff

05Integrated git commit checkpoints for incremental progress tracking

06Precise file mapping to enforce clean architecture and responsibility boundaries

使用场景

01Decomposing a complex feature specification into actionable coding steps

02Standardizing implementation patterns across an autonomous AI-run studio

03Generating detailed handoff documentation for subagent-driven development

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio writing-plans

For use in Claude.ai and ChatGPT

主要功能

01Automated plan review loops to validate logic before code is written

020 GitHub stars

03TDD-driven task decomposition with explicit failing and passing test cycles

04Standardized implementation headers for seamless subagent handoff

05Integrated git commit checkpoints for incremental progress tracking

06Precise file mapping to enforce clean architecture and responsibility boundaries

使用场景

01Decomposing a complex feature specification into actionable coding steps

02Standardizing implementation patterns across an autonomous AI-run studio

03Generating detailed handoff documentation for subagent-driven development

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio writing-plans

For use in Claude.ai and ChatGPT