writing-plans

Generate self-contained implementation plans with dependencies and testing guidance.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/krwhynot/crispy-crm --skill writing-plans-krwhynot
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-plans
Source: https://github.com/krwhynot/crispy-crm/tree/main/.claude/skills/writing-plans
Command: npx skills add https://github.com/krwhynot/crispy-crm --skill writing-plans-krwhynot

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Generates self-contained, unambiguous implementation plans for AI agents with zero context, applying TDD-oriented planning and clear execution steps.

Core Features & Use Cases

  • CRITICAL: AskUserQuestion across three rounds to gather type, scope, areas, granularity, execution mode, and database/testing constraints.
  • Phase-based plan output: Summary, detailed plan, and optional handoff prompts for /write-plan.
  • Ensures explicit steps, dependencies, and verification criteria are documented before implementation.

Quick Start

Begin with a request like: "Plan the integration for feature X," and the skill will guide you through questions and then produce a formal plan.

Frequently Asked Questions about writing-plans

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

FAQPage Schema
How do I create an implementation plan for an AI agent with no codebase context?

Writing-plans generates self-contained implementation plans by running three rounds of clarifying questions about task scope, granularity, and execution constraints, then outputs detailed steps with explicit dependencies and verification criteria so agents can execute without prior context.

What's the best way to break down a complex feature into executable steps for parallel execution?

The skill applies task breakdown through phase-based planning: it captures dependencies and execution order upfront, enforces TDD-oriented thinking, and documents constraints like database and testing requirements so multiple agents can work in parallel safely.

Can I use this for documenting agent instructions with dependencies and verification?

Yes. The skill produces plans with explicit step dependencies, execution order, and testing guidance built in, plus optional handoff prompts for implementation, making it suitable for handing off work to agents that need complete, unambiguous instructions.

How does this handle zero-context scenarios where the agent knows nothing about the codebase?

Writing-plans enforces a structured workflow—asking about type, scope, areas, and constraints across three rounds—then generates self-contained steps that don't rely on external codebase knowledge, integrating governance checks like constitution compliance to ensure safety.

What happens during the planning phase before implementation starts?

The skill produces a summary of gathered requirements, a detailed plan with step-by-step breakdown, dependencies, and verification criteria, then optionally generates handoff prompts ready for the /write-plan implementation stage.