executing-plans

Execute written implementation plans with step-by-step verified tasks.

1|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/xxih/ai-harness-zh --skill executing-plans-xxih
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: executing-plans
Source: https://github.com/xxih/ai-harness-zh/tree/main/references/translations/superpowers/skills/executing-plans
Command: npx skills add https://github.com/xxih/ai-harness-zh --skill executing-plans-xxih

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When a development plan already exists, this Skill turns that static plan into a safe, auditable execution flow by loading the plan, performing a critical review, running each step exactly as specified, and enforcing verification and stop conditions to avoid blind changes.

Core Features & Use Cases

  • Plan review before execution: Load a written plan and perform a critical review to surface questions, risks, and required clarifications.
  • Stepwise deterministic execution: Mark tasks in_progress, execute each small step as written, run required validations, and mark tasks completed.
  • Safe completion workflow: Integrates with branching and finishing workflows (e.g., worktree isolation and finishing-a-development-branch) and stops immediately on blockers or unclear instructions.
  • Use Case: Apply when a reviewer-approved implementation plan must be executed in an isolated environment, with mandatory verifications and handoffs to finishing workflows.

Quick Start

Tell the assistant to load the provided plan, perform a critical review, confirm any questions, and then execute the plan step by step with verification using the executing-plans skill.

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 written implementation plan step by step?

Step-by-step plan execution involves loading a written plan, performing a critical review, executing tasks sequentially, running validations, and marking task status transitions from in_progress to completed.

What is the best way to automate task execution in isolated git-worktrees?

Automating task execution in isolated git-worktrees requires running prewritten plans sequentially with mandatory validations, checkpoints, and immediate stops on blockers to ensure safe development workflow isolation.

Why do I need a plan review before starting development workflow execution?

Plan review before execution is required to surface questions, identify risks, and clarify instructions. This critical review prevents blind changes and ensures deterministic stepwise execution of the written plan.

Can I run validation checks during plan execution and stop on blockers?

Yes, validation runs are enforced after each small step. The execution flow stops immediately on blockers, unclear instructions, or failed validations to maintain a safe and auditable development workflow.

Does plan execution work with branching and finishing workflows?

Yes, plan execution integrates with branching and finishing workflows. It supports worktree isolation and hands off completed tasks to finishing workflows for safe branch integration.

What are the limitations of automated plan execution for development tasks?

Automated plan execution requires a prewritten, reviewer-approved plan to function. It cannot proceed with unclear instructions or missing validations and will halt immediately upon encountering execution blockers.