drive-change

Audit code changes through parallel reviews and apply fixes until findings stabilize.

Updated Jun 3, 2026
One-click install
npx skills add https://github.com/0xdeafcafe/skills --skill drive-change
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: drive-change
Source: https://github.com/0xdeafcafe/skills/tree/main/skills/drive-change
Command: npx skills add https://github.com/0xdeafcafe/skills --skill drive-change

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you push a code change from intent to a cleaner, safer result without setting up a full planning workflow. It is designed for quick or already-underway implementation work where you want the assistant to write code if needed, audit the diff, apply fixes, and tell you what still needs attention.

Core Features & Use Cases

  • Direct implementation from intent: If no diff exists yet, it can confirm the requested change and write the implementation directly from the conversation.
  • End-to-end review orchestration: It runs a structured audit pipeline across code quality, tests, feature correctness, security, and UX when relevant.
  • Parallel fix application: It dispatches validated findings into fix-applier agents, loops until findings stabilize, and produces one severity-ordered report.
  • Use Case: You ask the assistant to make a bug fix or small feature happen, and instead of only editing code, it also reviews the resulting change, applies mechanical fixes, escalates judgment calls, and tells you whether the branch is ready for a PR.

Quick Start

Ask the assistant to use drive-change to implement your requested code change and audit the result end to end.

Frequently Asked Questions about drive-change

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

FAQPage Schema
How do I automate code review and bug fixing before opening a pull request?

Automated code review and bug fixing runs structured audits across code quality, tests, security, and UX, dispatching validated findings into fix-applier agents that loop until the diff stabilizes and is ready for a pull request.

What is workflow orchestration for shipping a code change from implementation to audit?

Workflow orchestration for shipping a code change drives implementation through validation and fix application in a single loop, running parallel specialist reviews and iteratively re-reviewing until findings converge into one severity-ordered report.

Can I run a security audit and UX review on an existing diff before creating a pull request?

Yes, you can run security audits and UX reviews on existing diffs; the workflow applies adaptive routing by change size and triggers UX review specifically when UI files change, assessing readiness before opening a pull request.

How do I apply automated fixes to code findings and assess branch readiness?

Automated fixes apply through sensitivity-gated fix agents that process structured finding schemas, escalating judgment calls while applying mechanical fixes, looping until convergence to assess branch readiness.

What's the best way to handle quick feature implementation without a full planning workflow?

Handling quick feature implementation without full planning workflows uses direct implementation from intent, writing code from the conversation and orchestrating an end-to-end audit pipeline to apply fixes and flag remaining attention areas.

When should I not use an automated audit pipeline for code changes?

You should not use an automated audit pipeline for code changes requiring deep architectural planning or heavy manual judgment, as the workflow is optimized for quick features and bug fixes where mechanical fixes and structured reviews are sufficient.