mission-debug

Diagnose failing tests and propose minimal fixes with a structured mission-report.

Updated May 22, 2026
One-click install
npx skills add https://github.com/MathieuDoyon/mission --skill mission-debug
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mission-debug
Source: https://github.com/MathieuDoyon/mission/tree/main/.claude/skills/mission-debug
Command: npx skills add https://github.com/MathieuDoyon/mission --skill mission-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

mission-debug helps you understand why a failing test or unexpected behavior is happening, so you can confidently move forward instead of guessing at fixes.

Core Features & Use Cases

  • Systematic reproduction and evidence capture: Recreates the failure and records observed versus expected outcomes.
  • Hypothesis-driven isolation: Tests one falsifiable cause at a time using the minimum additional experiment or targeted read needed.
  • Minimal-change root-cause reporting: Identifies the likely file/line location and proposes the smallest fix without applying it.

Quick Start

Ask Claude to run mission-debug for your failing behavior and return a mission-report describing the most likely root cause and a minimal proposed change.

Frequently Asked Questions about mission-debug

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

FAQPage Schema
How do I find the root cause of a failing test without guessing at fixes?

Debug a failing test systematically by reproducing the failure, testing one falsifiable hypothesis at a time, and isolating the exact file and line. This produces a structured report with the root-cause location and a minimal proposed fix you can verify.

What is root cause analysis for software debugging and when do I need it?

Root cause analysis for debugging is the process of recreating a failure, capturing observed versus expected outcomes, and isolating the specific code change responsible. You need it when a failing test or unexpected output blocks progress and guessing is inefficient.

What's the best way to debug a regression reported by my testing workflow?

The best way to debug a regression is hypothesis-driven isolation: recreate the failure, test one possible cause at a time using the minimum targeted read or experiment, then document the root-cause location and propose the smallest fix without applying it.

How do I create a verification plan after proposing a code fix?

Create a targeted verification plan by reproducing the original failure, isolating the root-cause location, and defining the minimal set of tests needed to confirm the proposed fix resolves the regression without introducing new issues.

Can I use systematic debugging for unexpected output that isn't covered by a failing test?

Yes, systematic debugging applies to any unexpected output. It recreates the behavior, captures observed versus expected outcomes, isolates the likely cause through targeted experiments, and proposes the smallest fix with a verification plan.

What are the limitations of minimal-change debugging for complex regressions?

Minimal-change debugging isolates a single localizable root cause and proposes the smallest fix, so it may not fully address complex regressions spanning multiple systems or non-localized architectural issues that require broader changes.