debugging

Diagnose and remediate software bugs through severity-gated, architecture-aware workflows.

Updated May 3, 2026
One-click install
npx skills add https://github.com/miethe/boxbrain-2 --skill debugging-miethe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: debugging
Source: https://github.com/miethe/boxbrain-2/tree/main/.claude/skills/debugging
Command: npx skills add https://github.com/miethe/boxbrain-2 --skill debugging-miethe

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps diagnose and remediate software bugs without applying an unsafe or disproportionate fix. It routes incidents by severity, preserves architectural intent, and provides structured validation and handoff practices for reliable delivery.

Core Features & Use Cases

  • Severity-Gated Triage: Classify issues as Simple, Moderate, Complex, or Critical based on scope, root-cause clarity, architectural impact, and risk.
  • Delegated Remediation: Route investigation, implementation, architectural review, API validation, accessibility checks, and testing to the appropriate specialist agents.
  • Architecture-Aware Debugging: Use symbol-first investigation, existing code patterns, PRD discovery, contract checks, and cross-layer troubleshooting to avoid introducing inconsistent solutions.
  • Post-Incident Retrospectives: Classify shipped regressions and missed requirements against a failure taxonomy, then append actionable patch-plan addenda to existing plans.
  • Use Case: When a production API failure spans the router, service, database, and frontend, use this Skill to identify the root cause, select comprehensive remediation, coordinate specialists, validate contracts and tests, and document the handoff.

Quick Start

Use the debugging skill to triage the reported issue, determine its severity, identify the root cause, and recommend or coordinate the appropriate remediation workflow.

Frequently Asked Questions about debugging

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

FAQPage Schema
How do I debug a production incident that spans multiple layers like API and database?

Debug cross-layer production incidents by applying severity-gated triage to classify scope and risk. Use symbol-first investigation to identify the root cause across router, service, and database layers, then coordinate specialist agents for comprehensive remediation.

What is the best way to triage software bugs without applying an unsafe fix?

Triage software bugs using a severity-gated workflow that classifies issues as Simple, Moderate, Complex, or Critical. This architecture-aware process evaluates root-cause clarity and architectural impact to prevent disproportionate or unsafe remediation.

How do I conduct a post-merge regression retrospective for shipped bugs?

Conduct a post-merge regression retrospective by classifying shipped regressions and missed requirements against a failure taxonomy. This process generates actionable patch-plan addenda to append to existing project plans for structured follow-up.

Can I use this debugging workflow for frontend and API contract mismatches?

Yes, the debugging workflow supports frontend issues and API contract mismatches. It applies architecture-aware contract checks and delegates accessibility or API validation to specialist agents for targeted code validation.

When should I use comprehensive cross-layer remediation instead of a quick fix?

Use comprehensive cross-layer remediation when a bug's severity classification indicates high architectural impact or unclear root causes. The workflow routes complex incidents to specialist agents for thorough investigation rather than applying isolated quick fixes.

What do I need to validate code changes during root cause analysis?

Validate code changes during root cause analysis by using the workflow's built-in validation gates. These gates ensure existing-pattern alignment, contract verification, and testing are completed before documenting progress and generating handoff artifacts.