whip-debug

Enforce a reproduction-first debugging loop with phase-based verification across backends.

10|1|Updated Feb 5, 2026

Search Tags:#workflow #debugging #verification #reproduction #root-cause-analysis #ai-tools

Ready to run

npx skills add https://github.com/bang9/ai-tools --skill whip-debug

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: whip-debug
Source: https://github.com/bang9/ai-tools/tree/main/whip/skills-codex/whip-debug
Command: npx skills add https://github.com/bang9/ai-tools --skill whip-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates a disciplined debugging process that forces reproduction before fixes, reducing guesswork and misdiagnosis.

Core Features & Use Cases

Reproduction-first workflow: ensures a reliable reproduce-and-validate loop before proposing changes.
Phase-based analysis and verification: coordinates master intake, analysis, fix, and verification with distinct backends (codex, claude) and iterative rounds.
Artifact-driven reviews: requires structured deliverables (reproduction artifacts, root-cause hypotheses, suspect files, and evidence) for informed decisions.

Quick Start

Run the whip-debug workflow on a reported bug to start the repro-first debugging loop.

Frequently Asked Questions about whip-debug

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

FAQPage Schema

What is a reproduction-first debugging workflow and when do I need it?▼

A reproduction-first debugging workflow enforces a strict loop to reliably reproduce a bug before proposing fixes. You need it to reduce guesswork and prevent misdiagnosis when resolving complex software issues.

How do I debug a bug systematically from intake to verification?▼

Systematic debugging coordinates master intake, phase-based analysis, fix, and verification passes. It requires structured artifacts like reproduction steps, root-cause hypotheses, and suspect files for informed decisions before applying changes.

Can I use multi-backend verification to validate root-cause hypotheses?▼

Yes, multi-backend verification validates root-cause hypotheses using different backends like codex and claude. This iterative verification pass checks proposed fixes against evidence to ensure the original bug is resolved.

Does this debugging workflow work for AI toolchains and backend issues?▼

This debugging workflow applies directly to software debugging across AI toolchains, model interactions, and backend workflow issues. It uses phase-driven steps and artifact-driven reviews to identify and resolve bugs in these environments.

What are the limitations of a strict reproduction-first debugging loop?▼

The strict reproduction-first debugging loop requires generating reproducible artifacts before analysis, which may be difficult for intermittent or environment-specific bugs that lack consistent triggers or reliable reproduction steps.

Is whip-debug the best way to enforce reproducible artifact generation?▼

whip-debug enforces reproducible artifact generation by requiring structured deliverables like reproduction artifacts and root-cause hypotheses. This disciplined approach distinguishes it from ad-hoc debugging by reducing guesswork and misdiagnosis.