diagnose

Reproduce and diagnose bugs through a structured six-phase feedback loop.

Updated Jul 20, 2026
One-click install
npx skills add https://github.com/brandtam/rubber-ducky --skill diagnose-brandtam
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/brandtam/rubber-ducky/tree/main/.agents/skills/diagnose
Command: npx skills add https://github.com/brandtam/rubber-ducky --skill diagnose-brandtam

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Disciplined diagnosis loop helps teams tackle hard bugs and performance regressions by enforcing a structured, repeatable process from reproduction to regression-testing.

Core Features & Use Cases

  • Build a deterministic feedback loop to reproduce and isolate the bug.
  • Follow the six-phase workflow: reproduce, minimise, hypothesise, instrument, fix, regression-test, with explicit checkpoints.
  • Use across codebases and environments to diagnose flaky failures, performance degradations, and tricky regressions.

Quick Start

Describe the bug and its symptoms to begin the diagnosis loop.

Frequently Asked Questions about diagnose

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

FAQPage Schema
How do I debug a hard bug using a structured reproduction loop?

To debug a hard bug, you build a deterministic feedback loop to reproduce, hypothesise, instrument, and fix the issue. This structured diagnosis loop enforces a repeatable process from reproduction to regression-testing across codebases and deployments.

What is the best way to diagnose flaky failures and performance regressions?

Diagnosing flaky failures and performance regressions requires a disciplined six-phase process: build a feedback loop, reproduce, hypothesise, instrument, fix, and regression-test. This loop isolates environmental instabilities and performance degradations through explicit checkpoints.

How do I start finding a bug when I can only describe its symptoms?

You can start finding a bug by describing its symptoms to initiate the diagnosis loop. The process first builds a deterministic feedback loop to reliably reproduce the issue, then minimises the reproduction case to isolate the root cause.

Can I use a structured debugging process across different codebases and test environments?

Yes, this disciplined debugging loop applies across different codebases, tests, and deployments. It uses instrumentation and hypotheses to guide diagnosis, ensuring consistent bug isolation and regression testing regardless of the environment or platform.

What are the limitations of using a formal diagnosis loop for bug hunting?

The diagnosis loop requires building a deterministic feedback loop, meaning it may not suit purely non-reproducible issues without observable symptoms. Effective bug hunting also requires explicit checkpoints and instrumentation setups, adding initial process overhead before reaching a fix.