agent-workflow

Guide AI coding agents through Plan, Implement, and Review phases with approval gates.

15|2|Updated Dec 13, 2025
One-click install
npx skills add https://github.com/d-padmanabhan/agent-engineering-handbook --skill agent-workflow-d-padmanabhan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-workflow
Source: https://github.com/d-padmanabhan/agent-engineering-handbook/tree/main/skills/agent-workflow
Command: npx skills add https://github.com/d-padmanabhan/agent-engineering-handbook --skill agent-workflow-d-padmanabhan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents wasted effort and risky changes by forcing a clear Plan → Implementation → Review workflow with explicit approval gates and mandatory audit-friendly checkpoints.

Core Features & Use Cases

  • Approval-gated planning: Produces a structured plan and requires explicit user approval before any implementation starts.
  • Phase-based execution: Enforces PLAN, IMPLEMENT, and REVIEW boundaries to stop scope creep and unplanned refactors.
  • QA and audit guardrails: Requires QA validation for Level 2+ tasks and directs users to audit/checkpoint requirements for reversible, verifiable work.
  • Use cases: Apply when implementing multi-file features, complex refactors, or any task where the user wants structured scope control and verifiable progress.

Quick Start

Ask the agent to handle your change by saying: "Plan this: add X feature with constraints Y, then wait for my approval before implementing."

Frequently Asked Questions about agent-workflow

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

FAQPage Schema
How do I stop an AI coding agent from making unapproved changes during multi-file feature implementation?

To prevent unapproved code changes during complex refactoring, use an approval-gated workflow that enforces explicit Plan, Implement, and Review boundaries. The agent must generate a structured plan and wait for your explicit approval before writing any code.

What is the best way to manage scope creep when using an AI agent for complex refactoring?

The best way to manage scope creep during complex refactoring is enforcing phase-based execution with strict implementation gates. This approach separates planning, implementation, and review, stopping the agent from performing unplanned refactors outside the agreed scope.

How do I add QA validation and audit checkpoints to an AI software planning workflow?

You add QA validation and audit checkpoints to a software planning workflow by applying a structured execution process with mandatory review phases. This requires QA validation for complex tasks and directs the agent to follow audit-friendly checkpoint procedures for verifiable progress.

Can I use structured context management for feature-level tasks without losing track of audit requirements?

Yes, you can use structured context management for feature-level tasks while maintaining auditability by applying a checkpoint-based workflow. This process enforces explicit implementation gates and references mandatory audit checkpoints to ensure reversible and verifiable work.

Does this structured development workflow support pausing code implementation after an agreed scope is reached?

Yes, this structured development workflow supports pausing code implementation after an agreed scope is reached. It enforces strict implementation gates that stop unapproved coding and prevent the agent from continuing beyond the approved plan boundaries.