Verified:GitHub

executing-plans

Execute written implementation plans with review and verification checkpoints.

Updated Jul 11, 2022

Search Tags:#development workflow #implementation planning #ai agent #task execution #software delivery #verification checkpoints

Ready to run

npx skills add https://github.com/zhengfran/dotconfig --skill executing-plans-zhengfran

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: executing-plans
Source: https://github.com/zhengfran/dotconfig/tree/main/tools/ai/agents/skills/executing-plans
Command: npx skills add https://github.com/zhengfran/dotconfig --skill executing-plans-zhengfran

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you reliably turn a prewritten implementation plan into completed work by enforcing step-by-step execution, critical review, and verification checkpoints.

Core Features & Use Cases

Plan-first execution: Load the written implementation plan, review it critically, and pause to resolve questions before proceeding.
Task-by-task progress tracking: Execute each task in the plan as bite-sized steps while marking status and running verifications specified by the plan.
Safe completion workflow: After all tasks pass verification, finish via a required development-branch workflow rather than proceeding on main without consent.
Use Case: You have a detailed feature implementation plan (including verification steps) and need consistent execution in a separate session with review points.

Quick Start

Use the executing-plans skill to run your provided implementation plan end-to-end with review and verification checkpoints.

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 with verification checkpoints?▼

To execute a written implementation plan with verification checkpoints, load the plan into a separate session, review it critically, and run each task step-by-step while pausing on blockers or unclear instructions. The workflow mandates verification before completion and a finishing workflow to safely close the development branch.

What is the best way to ensure task-by-task progress tracking during software delivery?▼

The best way to ensure task-by-task progress tracking during software delivery is to execute each planned task as bite-sized steps, mark their status, and run the specific verifications defined by the plan before moving forward. This enforces strict adherence to plan steps and stop-on-failure behavior.

Can I run my development workflow on the main branch when executing plans?▼

You cannot run your development workflow on the main branch when executing plans. After all tasks pass verification, the workflow requires finishing a development branch using a mandated finishing workflow rather than proceeding on main without consent.

What happens when an AI agent encounters unclear instructions during plan execution?▼

When an AI agent encounters unclear instructions during plan execution, it applies stop-on-failure behavior and pauses to resolve questions before proceeding. This ensures strict adherence to the written plan and prevents deviating from the intended implementation steps.

Do I need a prewritten plan before starting development workflow execution?▼

You need a prewritten plan before starting development workflow execution. The process requires a detailed feature implementation plan that includes verification steps, which is then executed in a separate session with critical review points and verification checkpoints.

Why does my implementation plan execution pause on blockers?▼

Implementation plan execution pauses on blockers to enforce critical review and strict adherence to plan steps. This stop-on-failure behavior ensures unresolved questions or unclear instructions are addressed before proceeding, preventing unchecked errors during software delivery.