sdd:plan

Name: sdd:plan
Availability: InStock
Author: kennyolofsson23-netizen

Parse draft task files and orchestrate parallel agents for implementation-ready planning.

1|Updated Mar 15, 2026

Search Tags:#architecture #software development #verification #multi-agent #parallelization #task refinement

Ready to run

npx skills add https://github.com/kennyolofsson23-netizen/claude-code-config --skill sdd-plan-kennyolofsson23-netizen

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: sdd:plan
Source: https://github.com/kennyolofsson23-netizen/claude-code-config/tree/main/skills/sdd/plan
Command: npx skills add https://github.com/kennyolofsson23-netizen/claude-code-config --skill sdd-plan-kennyolofsson23-netizen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Draft task specifications often lack detailed analysis, architecture, step decomposition, parallel execution planning, and verification, leading to delays and errors when developers start implementation.

Core Features & Use Cases

Parallel Analysis: Runs research, codebase analysis, and business analysis simultaneously.
Architecture Synthesis: Merges findings into a coherent architectural overview.
Decomposition & Parallelization: Breaks the work into implementation steps and reorganizes them for maximum parallel execution.
Verification & Promotion: Adds LLM‑as‑judge checks and moves the refined task from the draft folder to the todo folder.
Use Case: Convert a newly created feature description into a ready‑to‑implement task for a development team.

Quick Start

Run the plan skill on the draft task file .specs/tasks/draft/add-validation.feature.md.

Frequently Asked Questions about sdd:plan

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

FAQPage Schema

How do I turn draft feature files into ready-to-implement software plans?▼

To turn draft feature files into ready-to-implement software plans, you refine draft specifications through multi-stage analysis, architecture synthesis, and task decomposition. This creates verified, implementation-ready steps organized for maximum parallel execution.

What is multi-agent parallelization for software development tasks?▼

Multi-agent parallelization for software development tasks involves orchestrating multiple agents to run research, codebase analysis, and business analysis simultaneously. This merges findings into a coherent architectural overview and organizes work for maximum parallel execution.

How do I refine a draft task specification for a development team?▼

To refine a draft task specification for a development team, you apply multi-stage analysis and judge evaluations to the draft file. This verifies the task specification and promotes it from the draft directory to the todo directory for implementation.

Does task planning with LLM-as-judge checks work for multi-stage analysis?▼

Task planning with LLM-as-judge checks works effectively for multi-stage analysis by verifying draft task specifications before implementation. It evaluates the synthesized architecture and decomposed steps, ensuring the refined task is fully planned and error-free.

What is the best way to decompose software architecture into parallel execution steps?▼

The best way to decompose software architecture into parallel execution steps is by merging simultaneous research and codebase analysis findings. This synthesis breaks work down into implementation steps and reorganizes them for maximum parallel execution.