learn-pj-kickoff

Structure project kickoffs with deliverables, tasks, decisions, and risks.

Updated Mar 12, 2026
One-click install
npx skills add https://github.com/novel-jp/projsight-plugin --skill learn-pj-kickoff
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learn-pj-kickoff
Source: https://github.com/novel-jp/projsight-plugin/tree/main/skills/learn-pj-kickoff
Command: npx skills add https://github.com/novel-jp/projsight-plugin --skill learn-pj-kickoff

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you turn an unclear project idea into a complete kickoff package so AI can work autonomously without missing scope, context, or decision records.

Core Features & Use Cases

  • Theme Selection Support: Helps choose or frame a project theme for a small internal tool, CLI utility, refactor, or custom idea.
  • Deliverable Definition: Captures purpose, scope, and done criteria for the first project artifact.
  • AI Workspace Design: Sets up a project-level CLAUDE.md with overview, tech stack, and coding rules.
  • Technical Decision Records: Records alternative approaches, tradeoffs, and the chosen technology path.
  • Task Breakdown and Risks: Splits the work into linked tasks and registers project risks with mitigations.
  • Use Case: Use it when starting a new engineering exercise and you want a structured kickoff that results in clear planning documents and execution-ready context.

Quick Start

Use the learn-pj-kickoff skill to guide me through choosing a project theme and producing the deliverable, CLAUDE.md, decision record, task breakdown, and risk list for a new kickoff.

Frequently Asked Questions about learn-pj-kickoff

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

FAQPage Schema
How do I structure a project kickoff for AI-assisted development?

To structure a project kickoff for AI-assisted development, define deliverables, create a CLAUDE.md workspace context, document technical decisions, decompose tasks, and register risks. This provides an execution-ready context for autonomous AI work.

What should be included in a technical decision record for a software project launch?

A technical decision record for a software project launch should include alternative approaches, tradeoffs, and the chosen technology path. Documenting these decisions ensures AI agents understand the project context and constraints.

How do I break down project work into tasks and map dependencies?

You break down project work into tasks and map dependencies by decomposing the overall deliverable into linked, actionable items. This task breakdown ensures all execution steps are tracked and connected to the project scope.

Can I use a CLAUDE.md file to set up AI workspace coding rules for a refactor?

Yes, you can use a CLAUDE.md file to set up AI workspace coding rules for a refactor. It captures the project overview, tech stack, and coding rules, providing the necessary context for AI to work autonomously.

What is the best way to register project risks and mitigations before coding starts?

The best way to register project risks and mitigations before coding starts is to identify potential issues during the kickoff phase and document them with specific mitigation strategies. This proactive tracking prevents scope creep during execution.

When do I need a structured kickoff package for a CLI utility or prototype build?

You need a structured kickoff package for a CLI utility or prototype build when you want AI to work autonomously without missing scope or context. It defines done criteria and sets up the working context required for development.