executing-plans

Execute written implementation plans with stepwise verification in isolated git worktrees.

1|Updated Mar 29, 2026
One-click install
npx skills add https://github.com/dmedina-dev/dev-forge --skill executing-plans-dmedina-dev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: executing-plans
Source: https://github.com/dmedina-dev/dev-forge/tree/main/plugins/forge-superpowers/skills/executing-plans
Command: npx skills add https://github.com/dmedina-dev/dev-forge --skill executing-plans-dmedina-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It enables developers to reliably execute a written implementation plan, ensuring each task is reviewed, verified, and completed before moving on, thereby reducing errors and missed steps.

Core Features & Use Cases

  • Load & Review: Reads the plan file, highlights concerns, and obtains clarification before starting.
  • Task Execution: Marks each task as in_progress, follows detailed steps, runs verifications, and marks completion.
  • Integration: Sets up an isolated git worktree, leverages the finishing-a-development-branch skill for finalization, and works best with subagent support. Use case: A developer has a detailed feature plan and wants the AI to execute it step‑by‑step while keeping the main branch untouched.

Quick Start

Ask the executing-plans skill to run the plan file 'my-feature.plan' and report completion.

Frequently Asked Questions about executing-plans

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

FAQPage Schema
How do I execute a software implementation plan step by step with git worktree isolation?

You can execute a software implementation plan step by step by loading the file, marking tasks in_progress, running verifications, and marking completion before proceeding, ensuring git worktree isolation keeps the main branch untouched.

What is stepwise verification when running a development plan?

Stepwise verification is a process where each task in an implementation plan is reviewed, verified, and marked complete before the next begins, reducing errors and missed steps during software development.

How do I track task progress while an AI runs my feature plan?

Task progress is tracked by marking each task as in_progress when starting, following the detailed steps, running verifications, and then marking completion before moving to the next item in the plan.

Do I need subagent support to run a development plan with stepwise reviews?

Yes, executing plans works best with subagent support. The platform must support subagent-driven development, git worktree setup, and integration with the finishing-a-development-branch skill for finalization.

Can I use git worktree isolation to keep my main branch untouched during plan execution?

Yes, the skill sets up an isolated git worktree for your development tasks. This allows the AI to execute the plan step-by-step while keeping your main branch completely untouched until finalization.

How do I finalize a development branch after executing a plan?

To finalize a development branch after plan execution, the skill integrates with the finishing-a-development-branch skill to complete the workflow and merge your isolated git worktree changes.