# Plan

> Start non-trivial work with a clean plan. Understand → Plan → Estimate → Execute → Verify.

- Skill: `rhythmmittal19/plan` (Agent Skill)
- Install (CLI): `npx skillmds@latest add rhythmmittal19/plan`
- Raw SKILL.md: https://api.skillmd.com/api/skills/rhythmmittal19/plan/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: RhythmMittal19 (https://skillmd.com/u/rhythmmittal19)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/rhythmmittal19/plan

---


## Current Project Context
!`basename $(pwd)`
!`cat package.json 2>/dev/null | head -5 || cat pyproject.toml 2>/dev/null | head -5 || cat Cargo.toml 2>/dev/null | head -5 || echo "No manifest found"`

## Current Branch & Status
!`git branch --show-current 2>/dev/null`
!`git status --short 2>/dev/null | head -10`

## Task
$ARGUMENTS

## Instructions
1. **Understand**: Read relevant files using subagents. Don't guess at architecture — explore first.
2. **Plan**: Create a numbered step-by-step plan. Each step should be independently verifiable.
3. **Estimate**: How many files will change? Any risky areas?
4. **Ask**: If anything is ambiguous, ask ONE clarifying question before proceeding.
5. **Execute**: Only after plan is approved, implement step by step.
6. **Verify**: Run tests after each significant change. Show me the results.

Present the plan and wait for my approval before coding.