gemini-draft-planner

Scan a codebase and generate structured markdown engineering story plans.

Updated Nov 8, 2022
One-click install
npx skills add https://github.com/gvinokur/qatar-prode --skill gemini-draft-planner
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gemini-draft-planner
Source: https://github.com/gvinokur/qatar-prode/tree/main/.gemini/skills/gemini-draft-planner
Command: npx skills add https://github.com/gvinokur/qatar-prode --skill gemini-draft-planner

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

The gemini-draft-planner automates the labor-intensive discovery and initial drafting work required to scope and plan engineering stories by exhaustively mapping a repository and producing a structured plan document.

Core Features & Use Cases

  • Repository discovery: Searches the codebase to identify architecture, server actions, repositories, components, and relevant files to change.
  • Template-driven plan generation: Produces a plans/STORY-[N]-plan.md file that strictly follows the provided plan-template.md structure, including call graph changes, implementation steps, testing strategy, and open questions.
  • Subagent handoff: Acts as a drafting subagent that writes the initial technical plan and then exits for a human or higher-level agent to review and perform git operations.
  • Use Case: When assigned a new feature story, run this Skill to produce a mid-level technical plan that a developer can implement or iterate on in code review.

Quick Start

Use the gemini-draft-planner to scan the repository and create a plans/STORY-123-plan.md draft by providing the story number and context.

Frequently Asked Questions about gemini-draft-planner

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

FAQPage Schema
How do I automate technical spec drafting for new engineering stories?

Automate technical spec drafting by scanning a codebase to map architecture, server actions, repositories, and components, then generating a structured markdown plan at plans/STORY-[N]-plan.md. This eliminates manual discovery work and produces a template-driven engineering story plan.

What is codebase mapping for story scoping and how does it work?

Codebase mapping for story scoping searches the repository to identify affected server actions, repositories, components, and call graph changes. It produces a mid-level technical plan document that a developer can implement or iterate on in code review.

Can I generate a story plan that includes implementation steps and testing strategy?

Yes, you can generate a story plan that includes file changes, implementation steps, testing strategy, and open questions. The plan strictly follows a provided plan-template.md structure and is output to plans/STORY-[N]-plan.md.

Does the generated draft plan perform git operations automatically?

No, the generated draft plan does not perform git operations automatically. It acts as a drafting subagent that writes the initial technical plan and then exits for a human or higher-level agent to review and perform git operations.

What's the best way to identify affected repositories and components across web application projects?

Identify affected repositories and components across web application projects by running an automated repository discovery scan. This maps the codebase architecture and extracts the call graph changes needed to produce a comprehensive technical draft plan.