sdd-plan

Generate phased JSON specifications with file-level dependencies and acceptance criteria.

9|3|Updated Oct 24, 2025
One-click install
npx skills add https://github.com/tylerburleigh/claude-sdd-toolkit --skill sdd-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-plan
Source: https://github.com/tylerburleigh/claude-sdd-toolkit/tree/main/skills/sdd-plan
Command: npx skills add https://github.com/tylerburleigh/claude-sdd-toolkit --skill sdd-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents common AI development pitfalls like hallucinated APIs, context drift, and breaking existing functionality by enforcing a "plan-first" methodology. It ensures every development task begins with a detailed, validated specification, reducing rework and increasing reliability.

Core Features & Use Cases

  • Staged Planning: For complex features, create high-level phase structures for early user approval before diving into detailed task breakdown.
  • Detailed JSON Specifications: Generates precise JSON files with phases, file-level modifications, dependencies, and verification steps.
  • Codebase Analysis Integration: Proactively uses codebase documentation (via Skill(sdd-toolkit:doc-query)) for accurate, context-aware planning.
  • Use Case: When starting a new feature like "Add JWT Authentication," use this skill to generate a comprehensive JSON spec, outlining database changes, service layer implementation, API integration, and testing, ensuring a clear roadmap before writing any code.

Quick Start

Verify sdd CLI is installed

sdd --help

Example: AI generates a markdown outline of phases for user approval.

(This is a conceptual step, the skill generates JSON based on user input)

After approval, generate the complete JSON specification

(This is the core output of the skill, e.g., specs/active/user-auth-001.json)

The specific command to generate the JSON is internal to the skill's logic.

Frequently Asked Questions about sdd-plan

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

FAQPage Schema
How do I create a detailed specification before starting development?

Plan-first development uses structured JSON specifications to outline phases, file modifications, dependencies, and acceptance criteria before writing code. This approach prevents hallucinated APIs, context drift, and breaking changes by enforcing upfront validation and ensuring every task has a clear, auditable roadmap.

What's the best way to break down a complex feature into manageable tasks?

Staged specification planning decomposes features into atomic tasks with file-level dependencies and verification steps. Each phase is documented in JSON with explicit modifications, dependencies, and acceptance criteria, making task execution sequential, verifiable, and production-ready.

How can I ensure my codebase analysis informs my development plan?

Proactive codebase analysis integration uses documentation queries to generate context-aware specifications that reflect existing architecture, APIs, and patterns. This ensures your plan aligns with the current system before implementation begins, reducing rework and integration surprises.

Can I use specification planning for API integrations and system redesigns?

Yes, specification planning covers complex features, large refactors, API integrations, and system redesigns. The JSON specification format accommodates multi-file changes, dependency tracking, and phased rollout, making it suitable for any development task requiring precise, auditable planning.

What problem does planning-first development solve in AI-assisted coding?

Plan-first methodology prevents common AI pitfalls like hallucinated APIs, context drift, and unintended side effects by requiring detailed specifications before generation. It increases reliability and reduces rework by enforcing validation and ensuring all development choices are documented and approved upfront.

How do I verify a development plan is complete and production-ready?

JSON specifications include explicit verification steps and acceptance criteria for each phase. These criteria define testable outcomes, dependencies that must be resolved, and acceptance conditions, ensuring the plan is complete, auditable, and ready for implementation before any code is written.