review-driven-execution

Execute development packets with implementation, spec compliance, and code quality review gates.

1|Updated Apr 27, 2026
One-click install
npx skills add https://github.com/AIChuY/ai_workflow --skill review-driven-execution
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-driven-execution
Source: https://github.com/AIChuY/ai_workflow/tree/main/review-driven-execution
Command: npx skills add https://github.com/AIChuY/ai_workflow --skill review-driven-execution

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you execute one ready development packet without blurring implementation, specification compliance, and code quality into a single vague step. It keeps status tracking, review notes, and final writeback disciplined so complex changes do not drift or get marked done too early.

Core Features & Use Cases

  • Packet-by-packet execution: Run a single worker-sized task with a clear start, finish, and dependency check.
  • Separate review gates: Validate the implementation, then verify it against the spec, then assess code quality before completion.
  • Ledger and status updates: Keep packet files and index entries synchronized as work moves from ready to in_progress to a final state.
  • Use Case: A developer has one refactor packet ready and wants to implement it, verify it, and record the outcome without losing track of downstream dependencies.

Quick Start

Ask the assistant to execute the ready packet, run the specification and code-quality checks, and update the packet ledger when the work is complete.

Frequently Asked Questions about review-driven-execution

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

FAQPage Schema
How do I execute a development packet with separate code review and spec compliance checks?

Packet execution separates implementation, specification compliance, and code quality into distinct review gates, ensuring validation occurs sequentially before updating the task ledger to a completed state.

What is the best way to track task status and dependencies during packet-based software execution?

Task status tracking uses ledger synchronization to propagate progress from ready to in_progress to a final state across packet files and index entries, validating downstream dependencies before work proceeds.

How do I prevent a complex refactor packet from being marked done before passing quality gates?

Review-driven execution enforces explicit quality gates that assess code quality and spec compliance, preventing premature completion by requiring disciplined final writeback and status validation.

Does packet execution work for single worker-sized tasks or can it batch multiple packets?

Packet execution applies to a single ready development packet, running one worker-sized task with a clear start, finish, and dependency check rather than batching multiple packets simultaneously.

When do I need explicit review gates instead of a single implementation step for software tasks?

Explicit review gates are needed when complex changes risk drifting or early completion, requiring separate specification compliance and code quality checks instead of blurring them into one vague step.