Skip to content
Back to skills

Simple Plan

ASecurity

Quick gut-check before implementing when the user directly asks you to do something. Investigates, proposes a lightweight plan, and implements only after approval. Use this instead of `create-plan` when the change is straightforward.

  • 9 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 23, 2026
ai-agentsgo

Security analysis

A100/100

Scanned September 23, 2026

npx -y skills add greenfield-inc/agent-farm --skill simple-plan --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Simple Plan?

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

Security grade badge for Simple Plan
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/greenfield-inc-simple-plan-agent-farm/badge)](https://www.skillsdirectory.com/skills/greenfield-inc-simple-plan-agent-farm)

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: simple-plan
description: Quick gut-check before implementing when the user directly asks you to do something. Investigates, proposes a lightweight plan, and implements only after approval. Use this instead of `create-plan` when the change is straightforward.
argument-hint: "[what the user wants done]"
---

# Simple Plan

When the user directly asks for a change, investigate first and propose a short
plan before writing code.

## Plan Contents

### Intent and Sources
- Triggering problem, why it matters, intended outcome, constraints and non-goals from the ticket/discussion
- Available source links or artifact paths; flag missing rationale and distinguish proposed assumptions from user decisions

### Current State
- root cause or current behavior
- concrete file references

### Proposed Changes
- what needs to change
- file references where relevant
- task list in implementation order

### Advice
- architectural or implementation guidance when useful

## Process

1. Investigate the codebase first
2. Present the plan to the user
3. Only implement after approval
4. After approval, keep one primary implementation authority by default
5. Keep the user's why, constraints, and non-goals explicit during implementation
6. After implementation, run `implementation-reviewer`
7. Prefer a fresh skeptical second review pass before declaring completion
8. If you have a separate Claude workflow available, it can be the parallel
   second-opinion lane, but Codex remains primary on this path

## Notes

- Keep the plan concise but concrete
- Include file references whenever possible
- If the task is broad or risky, recommend switching to `create-plan`
- Do not implement anything until the user approves

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 ai-agents →

Comments

Loading comments…

Newsletter

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