Skip to content
Back to skills

Feature Planning

ASecurity

Plan a feature with the owner answering every question — rounds of questions with recommendations, a player walkthrough, sketches for layout decisions, and a short decisions ledger to approve — then turn the approved plan into an epic. Use when the owner wants to plan a feature, mechanic or change themselves — when they invoke it, ask to plan something with you, or accept it as the next step offered in the project's interactive working style. In a hands-off project, prefer feature-planning-ha...

  • 3 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 7, 2026
development

Works with

  • claude code

Security analysis

A100/100

Scanned November 11, 2026

npx -y skills add amramos/game-ai-sdlc --skill feature-planning --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Feature Planning?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Feature Planning
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/amramos-feature-planning/badge)](https://www.skillsdirectory.com/skills/amramos-feature-planning)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: feature-planning
description: Plan a feature with the owner answering every question — rounds of questions with recommendations, a player walkthrough, sketches for layout decisions, and a short decisions ledger to approve — then turn the approved plan into an epic. Use when the owner wants to plan a feature, mechanic or change themselves — when they invoke it, ask to plan something with you, or accept it as the next step offered in the project's interactive working style. In a hands-off project, prefer feature-planning-handsoff unless the owner asks to answer the rounds themselves.
argument-hint: "The feature or change to plan"
---

# Feature planning (interactive)

Load the **`feature-planning-core`** skill (`game-sdlc:feature-planning-core` in Claude Code)
and follow it in **interactive mode**: the owner answers every round.

What this mode adds to the core:

## Make answering cheap

- If a multiple-choice question tool is available (such as `AskUserQuestion`), use it: up to four
  questions per call, your recommendation as the first option and labelled "(Recommended)". Put
  anything longer — the walkthrough, a sketch — in the message before the questions.
- Otherwise write the round as numbered markdown, each with "→ Recommended: …", so the owner can
  answer "yes to all except 2".
- Keep each round to the frontier. If a round would run past about six questions, split it by what
  the owner cares about most first (what the player sees), then the rest.
- The owner reads the ledger, not the plan. Write every answer into the ledger as it's given, so the
  approval at the end is a re-read, not a reconstruction.

## Before asking for approval

1. Show the ledger's decisions table, the walkthrough and the sketches — the three things the owner
   will actually judge.
2. For a big plan, report the two-way review in two lines: decisions missing from the plan (and how
   they were fixed), and invented decisions found (and the questions they became).
3. Ask for approval of the **ledger** explicitly, as one yes/no question. Anything other than a clear
   yes keeps the state at READY FOR APPROVAL.

## After approval

Offer the hand-off from the core skill's last step: record rulings, create the epic with
`issue-tracking`, and build it with `team-lead` — or `team-lead-handsoff` if the project's working
style is hands-off and the owner wants reviewed work merged without them.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in development →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.